JP6345335B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program.

There is an image processing apparatus that processes a moving image with a digital computer and detects an object such as an object or a person included in the moving image. In this image processing apparatus, there is a problem that a false detection occurs in which it is determined that there is actually no object. This problem is particularly noticeable in an image processing apparatus that detects a moving object by comparing two images forming a moving image. The incidence of false positives is high for moving images in which the brightness of the image changes due to daylight, illumination light, lightning, etc., and in moving images in which the brightness, hue, contrast, etc. change due to automatic camera control during shooting. Get higher.
In order to cope with such a problem, it is necessary to determine a combination of images that is not suitable for image processing. Japanese Patent Application Laid-Open No. H10-228667 discloses that the area of an object is compared between front and back frames and the area change amount is equal to or greater than a predetermined threshold value, and processing is performed as a possibility of erroneous detection.

JP 2012-123721 A

  Comparing the area of the object in the front and back frames as in Patent Document 1 can be performed only when the object already exists within the frame of the moving image. Therefore, at the moment when the object appears, if the object is included in the rear frame and the object is not included in the previous frame, the possibility of erroneous detection cannot be determined. A similar process can be applied with the object area of the previous frame as zero, but if the area change is small and less than the threshold, it is determined that there is no possibility of false detection. That is, when a small object appears and when a slight disturbance that is not the object appears, the area change amount is small, and it is determined that there is no possibility of erroneous detection. Therefore, in the technique of Patent Document 1, there is a problem that, when there is little disturbance, there is a false detection that determines that there is an object even though there is no object.

  An object of the present invention is to configure an image processing apparatus capable of determining, as a combination of images not suitable for image processing, two images including at least one of the disturbances even when the disturbance is slight.

An image processing apparatus according to the present invention includes:
A first image that is a two image forming a moving image and a second image are acquired, a first feature value for increase that represents a first pixel that is at least one pixel included in the first image, and the first image A second pixel that is at least one pixel included in the two images and is compared with the second feature value for increase that represents the second pixel corresponding to the first pixel; An increase area extraction unit that extracts an area of the second pixel of the second feature value for increase greater than a first threshold as an increase area;
The first image and the second image are acquired, and the first feature value for reduction representing the first pixel is compared with the second feature value for reduction representing the second pixel corresponding to the first pixel. A reduction region extraction unit that extracts, as a reduction region, the region of the second pixel of the second feature value for reduction that is greater than a second threshold value from the first feature value for reduction;
The degree of bias representing the degree of bias between the area of the increase region and the area of the decrease region is calculated, and whether or not the first image and the second image are suitable for image processing based on the calculated bias degree And a determination unit for determining whether or not.

  According to the image processing device of the present invention, whether or not the first image and the second image are suitable for image processing based on the degree of bias representing the degree of bias between the area of the increase region and the area of the decrease region. Since the determination unit is provided, the disturbance can be appropriately detected even if the disturbance has a small area, and therefore a combination of images that are not suitable for image processing can be determined with high accuracy.

1 is a block configuration diagram of an image processing apparatus according to Embodiment 1. FIG. FIG. 2 is a hardware configuration diagram of the image processing apparatus according to the first embodiment. 2 is a flowchart showing an image processing method and image processing of the image processing apparatus according to Embodiment 1. FIG. FIG. 2 is a diagram illustrating a minimum unit of a certain image area according to the first embodiment, where (a) is a diagram in which the minimum unit of the certain image area is a pixel, and (b) is a block in which the minimum unit of the certain image area is a block. FIG. 6C is a diagram in which the minimum unit of a certain image area is the entire image. FIG. 3 is a diagram illustrating an area of an increase region according to the first embodiment, (a) is a diagram illustrating the area of the increase region by the number of images, (b) is a diagram illustrating the area of the increase region by the number of blocks, and (c) is a diagram. The figure which represents the area of an increase area | region with the whole image. 2A and 2B are diagrams showing the area of a reduced region according to the first embodiment, where FIG. 3A is a diagram showing the area of the reduced region by the number of images, FIG. 2B is a diagram showing the area of the reduced region by the number of blocks, and FIG. The figure showing the area of a reduction | decrease area | region with the whole image. FIG. 3 is a block configuration diagram of an image processing apparatus according to a second embodiment. FIG. 9 is a flowchart showing an image processing method and image processing of the image processing apparatus according to the second embodiment. FIG. 6 is a diagram illustrating the area of a change region according to the second embodiment, where (a) illustrates the area of the change region in terms of the number of images, (b) illustrates the area of the change region in terms of the number of blocks, and (c) illustrates The figure showing the area of a change area with the whole image. FIG. 6 is a block configuration diagram of an image processing apparatus according to a third embodiment. FIG. 9 is a flowchart showing an image processing method and image processing of an image processing apparatus according to Embodiment 3.

Embodiment 1 FIG.
In the present embodiment, an image processing apparatus 100 that determines whether two images forming a moving image are a combination of images that are not suitable for image processing will be described.
The combination of images not suitable for image processing is two images or the like in which at least one includes disturbance. When the area of the disturbance is small, even if one of the two images is a two image including the disturbance, there is a possibility that the two images are determined to be suitable for image processing. In the present embodiment, an image processing apparatus that can determine that two images including an image including a disturbance are a combination of images that are not suitable for image processing even when the area of the disturbance is slight will be described. .

*** Explanation of configuration ***
A block configuration of the image processing apparatus 100 according to the present embodiment will be described with reference to FIG.
The image processing apparatus 100 includes an increase area extraction unit 110, a decrease area extraction unit 120, and a determination unit 1300. The determination unit 1300 includes an increase / decrease bias determination unit 130 and an image determination unit 140.

The increased area extraction unit 110 acquires a first image 11 and a second image 12 that are two images forming a moving image. The increase area extraction unit 110 includes a first feature value 1111A for increase that represents the first pixel 111 that is at least one pixel included in the first image 11, and a second that is at least one pixel included in the second image 12. The second increase feature value 1211A representing the pixel 121 is acquired. The increase area extraction unit 110 acquires the second feature value 1211A for increase representing the second pixel 121 corresponding to the first pixel 111, which is the second pixel 121 that is at least one pixel included in the second image 12. .
The increase area extraction unit 110 compares the first increase feature value 1111A and the second increase feature value 1211A. The increase area extraction unit 110 extracts the area of the second pixel 121 of the increase second feature value 1211A that is larger than the first threshold 201 as the increase area 1101 from the increase first feature value 1111A.

The increased area extraction unit 110 compares the second image 12 associated with the new time information of the two images forming the moving image with the first image 11 associated with the old time information. The increase area extraction unit 110 compares the first image 11 and the second image 12, and the amount of increase in the feature value in the area of the first pixel 111 or the second pixel 121, which is a constant image area, sets the first threshold value 201. The image area that exceeds is determined as the increased area 1101. The increase area extraction unit 110 is also referred to as an increase area determination unit.
Here, the increase amount of the feature value is an increase amount from the first increase feature value 1111A to the second increase feature value 1211A.

The reduced area extraction unit 120 acquires the first image 11 and the second image 12. The reduction region extraction unit 120 is a reduction first feature value 1111B that represents the first pixel 111 that is at least one pixel included in the first image 11, and a reduction pixel that represents the second pixel 121 corresponding to the first pixel 111. The second feature value 1211B is acquired. The decrease area extraction unit 120 extracts the area of the second pixel 121 having the decrease second feature value 1211B that is larger than the second threshold value 202 as the decrease amount from the decrease first feature value 1111B as the decrease area 1201.
The reduced area extraction unit 120 compares the first image 11 and the second image 12, and the amount of decrease in the feature value in the area of the first pixel 111 or the second pixel 121, which is a constant image area, sets the second threshold value. The exceeding image area is determined as the reduced area 1201. The reduced area extraction unit 120 is also referred to as a reduced area determination unit.
Here, the reduction amount of the feature value is a reduction amount from the first feature value for reduction 1111B to the second feature value for reduction 1211B.

  Feature values such as the first feature value for increasing 1111A, the second feature value for increasing 1211A, the first feature value for decreasing 1111B, and the second feature value for decreasing 1211B are the first pixel or the second based on the pixel value. It is a value representing the characteristics of the pixel. For example, when the first pixel or the second pixel is one pixel, the pixel value of the pixel may be used as the feature value. Further, when the first pixel or the second pixel is a set of pixels, an average value of pixel values of each pixel in the set of pixels may be used as the feature value. When the first pixel or the second pixel is a set of pixels, the number of pixels that satisfy a predetermined condition for each pixel in the set of pixels may be used as the feature value.

The first pixel 111 may be a single pixel or an aggregate of a plurality of pixels. Further, the first pixel 111 may be all the pixels constituting the first image 11. Similarly, the second pixel 121 may be one pixel or an aggregate of a plurality of pixels. Further, the second pixel 121 may be all the pixels constituting the second image 12.
When the region of the first pixel 111 included in the first image 11 to be compared with the feature value and the region of the second pixel 121 included in the second image 12 are referred to as a fixed image region or a minimum unit of the fixed image region There is.

The determination unit 1300 calculates a degree of bias C that represents the degree of bias between the area of the increasing region 1101 and the area of the decreasing region 1201, and the first image 11 and the second image 12 are based on the calculated degree of bias C. It is determined whether it is suitable for image processing. In the following description, the area of the increasing region 1101 is referred to as the area I of the increasing region, and the area of the decreasing region 1201 is referred to as the area D of the decreasing region.
The determination unit 1300 compares the degree of bias C with the third threshold 203, and determines that the first image 11 and the second image 12 are not suitable for image processing when the degree of bias C is greater than the third threshold 203. . The determination unit 1300 calculates the value of the ratio between the area I of the increasing region and the area D of the decreasing region as the degree of bias C.

The increase / decrease bias determination unit 130 of the determination unit 1300 determines that there is an increase / decrease bias when the bias degree C between the increase region 1101 and the decrease region 1201 exceeds the third threshold value 203.
The degree of bias C between the increasing region 1101 and the decreasing region 1201 is calculated based on the area I of the increasing region and the area D of the decreasing region.
When the increase / decrease bias determination unit 130 determines that there is an increase / decrease bias, the image determination unit 140 of the determination unit 1300 determines that the first image 11 and the second image 12 are image combinations that are not suitable for image processing.

  The first threshold value 201, the second threshold value 202, and the third threshold value 203 may be stored in the image processing apparatus 100 in advance, or may be set by the user at the start of image processing. Further, the first threshold 201 and the second threshold 202 may be different from each other or the same value.

  An example of the hardware configuration of the image processing apparatus 100 according to the present embodiment will be described with reference to FIG.

The image processing apparatus 100 is a computer.
The image processing apparatus 100 includes hardware such as a processor 901, an auxiliary storage device 902, a memory 903, a communication device 904, an input interface 905, and a display interface 906.
The processor 901 is connected to other hardware via the signal line 910, and controls these other hardware.
The input interface 905 is connected to the input device 907.
The display interface 906 is connected to the display 908.

The processor 901 is an IC (Integrated Circuit) that performs processing.
The processor 901 is, for example, a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or a GPU (Graphics Processing Unit).
The auxiliary storage device 902 is, for example, a ROM (Read Only Memory), a flash memory, or an HDD (Hard Disk Drive).
The memory 903 is, for example, a RAM (Random Access Memory).
The communication device 904 includes a receiver 9041 that receives data and a transmitter 9042 that transmits data.
The communication device 904 is, for example, a communication chip or a NIC (Network Interface Card).
The input interface 905 is a port to which the cable 911 of the input device 907 is connected.
The input interface 905 is, for example, a USB (Universal Serial Bus) terminal.
The display interface 906 is a port to which the cable 912 of the display 908 is connected.
The display interface 906 is, for example, a USB terminal or an HDMI (registered trademark) (High Definition Multimedia Interface) terminal.
The input device 907 is, for example, a mouse, a keyboard, or a touch panel.
The display 908 is, for example, an LCD (Liquid Crystal Display).

  The auxiliary storage device 902 stores programs that realize the functions of the increase area extraction unit 110, the decrease area extraction unit 120, the increase / decrease bias determination unit 130, and the image determination unit 140 shown in FIG. Hereinafter, the increase area extraction unit 110, the decrease area extraction unit 120, the increase / decrease bias determination unit 130, and the image determination unit 140 are collectively referred to as “parts”.

The program that realizes the function of the “part” described above is also referred to as an image processing program. The program that realizes the function of “unit” may be a single program or a plurality of programs.
This program is loaded into the memory 903, read into the processor 901, and executed by the processor 901.
Further, the auxiliary storage device 902 also stores an OS (Operating System).
Then, at least a part of the OS is loaded into the memory 903, and the processor 901 executes a program that realizes the function of “unit” while executing the OS.
Although one processor 901 is illustrated in FIG. 2, the image processing apparatus 100 may include a plurality of processors 901.
A plurality of processors 901 may execute a program for realizing the function of “unit” in cooperation with each other.
In addition, information, data, signal values, and variable values indicating the results of the processing of “unit” are stored as files in the memory 903, the auxiliary storage device 902, or a register or cache memory in the processor 901.

The “part” may be provided as “circuitry”.
Further, “part” may be read as “circuit”, “process”, “procedure”, or “processing”. Further, “processing” may be read as “circuit”, “process”, “procedure”, or “part”.
“Circuit” and “Circuitry” include not only the processor 901 but also other types of processing circuits such as a logic IC or GA (Gate Array) or ASIC (Application Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array). It is a concept to include.

  Note that what is called a program product is a storage medium, a storage device, or the like on which a program that realizes the function described as a “part” is recorded. It is what you are loading.

*** Explanation of operation ***
The image processing method and image processing S100 of the image processing apparatus 100 according to the present embodiment will be described with reference to FIG.

<Increase area extraction processing S110>
The increased area extraction unit 110 acquires a first image 11 and a second image 12 that are two images forming a moving image.
As described above, it is assumed that the second image 12 is an image associated with time information that is newer than the time information associated with the first image 11. The first image 11 and the second image 12 may be images extracted from a continuous image forming a moving image, and may be continuous or may not be continuous.
The increased area extraction unit 110 corresponds to the first feature value 1111A for increasing pixels included in the acquired first image 11 and the pixels included in the acquired second image 12 and included in the first image 11. The second feature value 1211A for increasing pixels to be compared is compared. The increase area extraction unit 110 executes an increase area extraction process S110 that extracts, as the increase area 1101, a pixel area of the increase second feature value 1211A in which the increase amount from the increase first feature value 1111A is larger than the first threshold 201. To do.

An image forming a moving image is associated with time information such as a time stamp. Therefore, the two images forming the moving image have one image having newer time information than the other image, and are classified into the first image 11 and the second image 12 having newer time information than the first image 11. Can do.
Two consecutive images forming a 30 fps (frames per second) moving image are associated with different time information by 1/30 second. Even when the time information is lost due to some processing such as image format conversion, the serial number indicating the numbered image in the moving image can be handled as the time information. In this case, if the time information is newer as the value of the serial number is larger, the two images forming the moving image can be classified into the first image 11 and the second image 12.

The increased region extraction process S110 according to the present embodiment will be specifically described with reference to FIG.
A configuration example in which the minimum unit of a certain image area in the increase area extraction processing S110 is a pixel will be described with reference to FIG. Here, the constant image region to be compared, that is, the region of the first pixel 111 is a region of one pixel included in the first image 11. In addition, a constant image region to be compared, that is, the region of the second pixel 121 is a region of one pixel included in the second image 12 and corresponding to the first pixel 111 in the second image 12. That is, each of the first pixel 111 and the second pixel 121 for comparing the amount of increase in the feature value is one pixel.

As shown in (a) of FIG. 4, the increase area extraction unit 110 uses a minimum unit of a certain image area as a pixel. The increase area extraction unit 110 sets the increase amount of the feature value to a value obtained by subtracting the pixel value of the pixel at the same position of the first image 11 from the pixel value of one pixel of the second image 12.
That is, the increase area extraction unit 110 sets the pixel value of the second pixel 121 that is one pixel of the second image 12 as the second feature value 1211A for increase, and is in the same position as the second pixel 121 in the first image 11. The amount of increase in the feature value is calculated using the pixel value of the first pixel 111 as the first feature value 1111A for increase.
The increase area extraction unit 110 compares the increase amount with the first threshold 201 to determine whether the pixel is the increase area 1101 for each pixel.

  With reference to FIG. 4B, a configuration example in which the minimum unit of a certain image area in the increasing area extraction process S110 is a block including a plurality of pixels will be described. Here, the first pixel 111 is a plurality of pixels included in the first image 11, and the second pixel 121 is included in the second image 12, at a position corresponding to the first pixel 111 in the second image 12. A plurality of pixels. That is, each of the first pixel 111 and the second pixel 121 for comparing the amount of increase in the feature value is one block.

As shown in FIG. 4B, the increased area extracting unit 110 sets a minimum unit of a certain image area as a block that is an aggregate of a plurality of pixels. The increase area extraction unit 110 calculates the increase amount of the feature value from the average value of the pixel values in the block of the block of the second image 12 to the pixel value of the pixel in the block of the block at the same position of the first image 11. The average value is subtracted.
The increase area extraction unit 110 sets the average value of the pixel values of the pixels in the block in the block (second pixel 121) in the second image 12 as the second feature value 1211 </ b> A for increase, and the second pixel 121 in the first image 11. The average value of the pixel values of the pixels in the block in the block at the same position (first pixel 111) is set as the first feature value 1111A for increase, and the increase amount of the feature value is calculated.
The increase area extraction unit 110 compares the increase amount with the first threshold 201 and determines whether the block is the increase area 1101 for each block.

In the extraction method in which the minimum unit of a certain image area is a block, threshold processing for each pixel is not required. Therefore, the algorithm is simplified and the processing is simplified compared to the extraction method in which the minimum unit of a certain image area is a pixel. There is an effect of high speed.
In addition, in the extraction method in which the minimum unit of a certain image area is a block, the amount of increase in the feature value is calculated in the second pixel 121 so that the pixel value of the second pixel 121 is the same as that of the first pixel 111. It may be the number of pixels in the block that satisfy the condition that the pixel value is larger than a certain pixel. In this case, the increase area extraction unit 110 does not have a pixel satisfying the above condition in the block that is the first pixel 111, so 0 is set as the first feature value 1111 </ b> A for increase, and the above condition is satisfied in the block that is the second pixel 121. The number of pixels is set as the second feature value for increase 1211A, and the increase amount of the feature value is calculated.
The extraction method in which the minimum unit of a certain image area is a block has an effect that the processing speed is increased when a processing device capable of performing a specific operation at a high speed such as subtraction between images or counting of the number of pixels is installed. The blocks do not necessarily have to be rectangular, and may have any shape.

A configuration example in which the minimum unit of a certain image area in the increased area extraction processing S110 is the entire image will be described with reference to FIG. Here, the first pixel 111 is all the pixels included in the first image 11, and the second pixel 121 is all the pixels included in the second image 12. That is, the minimum unit of a certain image area is the entire image.
As shown in FIG. 4C, the increase area extraction unit 110 sets the minimum unit of a certain image area as the entire image, and determines the increase amount of the feature value as the average value of the pixel values of all the pixels of the second image 12. The average value of the pixel values of all the pixels of the first image 11 is subtracted from the first image 11.
That is, the increase area extraction unit 110 sets the average value of the pixel values of all the pixels of the second image 12 as the increase second feature value 1211A, and sets the average value of the pixel values of all the pixels of the first image 11 to the first increase value. As the feature value 1111A, an increase amount of the feature value is calculated.
The increase area extraction unit 110 determines whether or not the entire image is the increase area 1101.

In the extraction method in which the minimum unit of a certain image area is the entire image, the algorithm is simplified and the processing is faster than the extraction method in which the minimum unit of the certain image area is a block.
Note that the increase area 1101 may be determined by defining the minimum unit of a certain image area and the definition of the increase amount of the feature value other than the method described above. The increased area extraction unit 110 compares the first image 11 and the second image 12, and determines that the image area is the increased area 1101 when the increase amount of the feature value exceeds a threshold value in a certain image area. Any configuration can be used.

<Decrease Area Extraction Processing S120>
The decrease area extraction unit 120 acquires the first image 11 and the second image 12, decreases the first feature value 1111 </ b> B of the first pixel 111 included in the first image 11, and the first image included in the second image 12. The second feature value 1211B for reduction of the two pixels 121 is compared. The reduction area extraction unit 120 extracts the area of the second pixel 121 of the second feature value 1211B for reduction that is larger than the second threshold value 202 as the reduction area 1201 from the first characteristic value for reduction 1111B as the reduction area 1201. S120 is executed.

The reduced area extraction process S120 will be specifically described with reference to FIG.
A configuration example in which the minimum unit of a certain image area in the reduced area extraction process S120 is a pixel will be described with reference to FIG. Here, the fixed image region is a region of the first pixel 111 and a region of the second pixel 121 which are one pixel.
As shown in FIG. 4A, the reduced area extraction unit 120 uses a minimum unit of a certain image area as a pixel. The reduction area extraction unit 120 sets the reduction amount of the feature value to a value obtained by subtracting the pixel value of the pixel at the same position in the second image 12 from the pixel value of the pixel in the first image 11.
That is, the reduction area extraction unit 120 sets the pixel value of the first pixel 111, which is one pixel of the first image 11, as the first feature value 1111B for reduction, and is in the same position as the first pixel 111 in the second image 12. The feature value reduction amount is calculated using the pixel value of the second pixel 121 as the second feature value for reduction 1211B.
The reduction area extraction unit 120 determines whether the pixel is the reduction area 1201 for each pixel.

A configuration example in which the minimum unit of a certain image area in the reduced area extraction process S120 is a block will be described with reference to FIG. Here, the fixed image area is an area of the first pixel 111 and an area of the second pixel 121 that are aggregates of pixels.
As shown in (b) of FIG. 4, the reduced area extraction unit 120 sets a minimum unit of a certain image area as a block. The reduction area extraction unit 120 calculates the reduction amount of the feature value from the average pixel value of the pixels in the block of the block of the first image 11 to the average pixel value of the pixels in the block of the block at the same position of the second image 12. The value minus the value.
The reduction area extraction unit 120 sets the average value of the pixel values of the pixels in the block in the block (second pixel 121) in the second image 12 as the second feature value for reduction 1211 </ b> B, and the second pixel 121 in the first image 11. The average value of the pixel values of the pixels in the block in the block (first pixel 111) at the same position as the first feature value 1111B for reduction is calculated, and the amount of reduction of the feature value is calculated.
The reduction area extraction unit 120 determines whether the block is a reduction area 1201 for each block.

In the extraction method in which the minimum unit of a certain image area is a block, threshold processing for each pixel is not required. Therefore, the algorithm is simplified and the processing is simplified compared to the extraction method in which the minimum unit of a certain image area is a pixel. There is an effect of high speed.
In addition, in the extraction method in which the minimum unit of a certain image area is a block, the amount of decrease in the feature value is set so that the pixel value of the second pixel 121 is the same as that of the first pixel 111 in the second pixel 121. The number of pixels in the block that satisfy the condition that the pixel value is smaller than a certain pixel may be used. In this case, the reduced area extraction unit 120 does not have a pixel that satisfies the above condition in the block that is the first pixel 111, so 0 is set as the first feature value 1111 B for reduction, and the above condition is satisfied in the block that is the second pixel 121. A value obtained by multiplying the number of pixels by −1 is set as a second feature value for reduction 1211B, and a feature value reduction amount is calculated.
The extraction method in which the minimum unit of a certain image area is a block has an effect that the processing speed is increased when a processing device capable of performing a specific operation at a high speed such as subtraction between images or counting of the number of pixels is installed. The blocks do not necessarily have to be rectangular, and may have any shape.

A configuration example in which the minimum unit of a certain image area in the reduced area extraction processing S120 is the entire image will be described with reference to FIG. Here, the fixed image region is a region of the first pixel 111 that is a pixel of the entire image of the first image 11 and a region of the second pixel 121 that is a pixel of the entire image of the second image 12.
As shown in FIG. 4C, the reduced area extraction unit 120 sets the minimum unit of a certain image area as the entire image. The reduction area extraction unit 120 sets the amount of reduction of the feature value to a value obtained by subtracting the average value of the pixel values of all the pixels of the second image 12 from the average value of the pixel values of all the pixels of the first image 11.
That is, the reduction area extraction unit 120 sets the average value of the pixel values of all the pixels of the second image 12 as the second characteristic value for reduction 1211B, and sets the average value of the pixel values of all the pixels of the first image 11 to the first value for reduction. A feature value decrease amount is calculated as the feature value 1111B.
The reduction area extraction unit 120 determines whether or not the entire image is the reduction area 1201.

In the extraction method in which the minimum unit of a certain image area is the entire image, the algorithm is simplified and the processing is faster than the extraction method in which the minimum unit of the certain image area is a block.
Note that the reduction area 1201 may be determined by defining the minimum unit of a certain image area and the definition of the reduction amount of the feature value other than the method described above. The reduction area extraction unit 120 compares the first image 11 and the second image 12 and determines that the image area is the reduction area 1201 when the reduction amount of the feature value exceeds a threshold value in a certain image area. Any configuration can be used.

<Determination Processing S1300>
The determination unit 1300 calculates a degree of bias C that represents the degree of bias between the area I of the increasing region and the area D of the decreasing region, and the first image 11 and the second image 12 are based on the calculated degree of bias C. A determination process S1300 for determining whether or not the image processing is suitable is executed.
The determination process S1300 includes an increase / decrease bias determination process S130 and an image determination process S140.

<Increase / decrease bias determination processing S130>
The increase / decrease bias determination process S <b> 130 will be specifically described with reference to FIGS. 5 and 6.
The increase / decrease bias determination unit 130 is configured to calculate the bias degree C using the area I of the increase region and the area D of the decrease region as shown in the following equation.
C = 0 (when I = D = 0)
C = ∞ (when I> D = 0 or D> I = 0)
C = I / D (when I>D> 0)
C = D / I (when D ≧ I> 0)

  A configuration example in which the area of an image region is represented by the number of pixels in the image region will be described with reference to FIGS. 5A and 6A. When the increase area extraction unit 110 determines the increase area 1101 for each pixel and the decrease area extraction unit 120 determines the decrease area 1201 for each pixel, the area of the image area is represented by the number of pixels, and the degree of bias C is calculated. .

The increase / decrease bias determination unit 130 acquires 158321 pixels as the increase area 1101 as illustrated in FIG. 5A and acquires 20179 pixels as the decrease area 1201 as illustrated in FIG. That is, the area I of the increased region is represented as 158321 pixels, and the area D of the decreased region is 20179 pixels.
The increase / decrease bias determination unit 130 determines that I> D, and calculates the degree of bias C as C = I / D, that is, 158321 / 20179≈7.846.
The increase / decrease bias determination unit 130 compares the calculated bias degree C with the third threshold 203, determines whether there is an increase / decrease bias, and outputs a determination result.

  Next, a configuration example in which the area of the image area is represented by the number of blocks in the image area will be described with reference to FIGS. 5B and 6B. When the increase area extraction unit 110 determines the increase area 1101 in units of blocks and the decrease area extraction unit 120 determines the decrease area 1201 in units of blocks, the degree of bias C is calculated by expressing the area as the number of blocks.

The increase / decrease bias determination unit 130 acquires 12 blocks as the area I of the increased region as shown in FIG. 5B, and acquires 2 blocks as the area D of the decreased region as shown in FIG. 6B. . The increase / decrease bias determination unit 130 determines that I> D, and calculates the degree of bias C as C = I / D, that is, 12/2 = 6.
The increase / decrease bias determination unit 130 compares the calculated bias degree C with the third threshold 203, determines whether there is an increase / decrease bias, and outputs a determination result.

  Next, a configuration example in which the area of the image region is represented by the entire image region will be described with reference to FIG. 5C and FIG. The increase area extraction unit 110 determines the increase area 1101 in the entire image unit, and the decrease area extraction unit 120 determines the decrease area 1201 in the entire image unit. In this case, an image to be determined is an image that is neither the increase area 1101 nor the decrease area 1201, an image that is the increase area 1101, or an image that is the decrease area 1201. The increase / decrease bias determination unit 130 determines that there is an increase / decrease bias when the entire image is only one of the increase region 1101 and the decrease region 1201.

  This determination method is equivalent to a process of calculating the bias degree C by applying the above-described formula and then comparing it with the third threshold value 203. Assume that the increased area extraction unit 110 determines that the increased area 1101 has an area I of the increased area as one image as shown in FIG. Further, it is assumed that the reduced region extraction unit 120 determines that the reduced region 1201 is an image having a reduced region area D of 0 as shown in FIG. In this case, the degree of bias C is ∞ (infinite), and since the degree of bias C exceeds the finite third threshold 203, it is determined that there is an increase / decrease bias. In this way, the increase / decrease bias determination unit 130 can compare the calculated bias degree C with the third threshold 203, determine whether there is an increase / decrease bias, and output the determination result.

  The definition of the degree of bias C is not limited to the above formula. The degree of bias C may be calculated by a conditional branch or calculation formula that is more complicated than the above formula. Further, the increase / decrease bias determination unit 130 may have any configuration as long as it is configured to determine that there is an increase / decrease bias when the degree of bias of the increase region 1101 and the decrease region 1201 exceeds a threshold value.

<Image determination processing S140>
The image determination unit 140 determines whether the first image 11 and the second image 12 are suitable for image processing based on the determination result output from the increase / decrease bias determination unit 130. The image determination unit 140 determines that the first image 11 and the second image 12 are a combination of images that are not suitable for image processing when the determination result output from the increase / decrease bias determination unit 130 has an increase / decrease bias. When the determination result output from the increase / decrease bias determination unit 130 indicates that there is no increase / decrease bias, the image determination unit 140 determines that the first image 11 and the second image 12 are a combination of images suitable for image processing.

  Note that the increase / decrease bias determination unit 130 may calculate the bias degree C and output the calculated bias degree C to the image determination unit 140. The image determination unit 140 acquires the degree of bias C from the increase / decrease bias determination unit 130, compares the acquired degree of bias C with the third threshold value 203, and the first image 11 and the second image 12 are suitable for image processing. It may be configured to determine whether or not a combination of images.

  This is the end of the description of the image processing S100 of the image processing apparatus 100 according to the present embodiment.

*** Explanation of effects ***
The image processing apparatus according to the present embodiment acquires a second image associated with new time information and a first image associated with old time information among the two images forming the moving image. The increase area extraction unit compares the acquired first image and second image, and determines that the image area is an increase area when the increase amount of the feature value exceeds a threshold value in a certain image area. The reduction area extraction unit compares the first image and the second image, and determines that the image area is a reduction area when the reduction amount of the feature value exceeds a threshold value in a certain image area. Also, the increase / decrease bias determination unit determines that there is an increase / decrease bias when the degree of bias of the increase region and the decrease region exceeds a threshold value. The image determination unit determines that the first image and the second image are a combination of images that are not suitable for image processing when the increase / decrease bias determination unit determines that there is an increase / decrease bias.

  As described above, in the image processing apparatus according to the present embodiment, the degree of bias between the increase region and the decrease region is a criterion for determining whether the two images are a combination of images that are not suitable for image processing. Therefore, the image processing apparatus displays the first image and the second image when the area of the increase region or the decrease region is small and has a tendency to be biased toward either the increase region or the decrease region. It can be determined that the combination of images is not suitable for processing. When a slight disturbance occurs in the image, if the disturbance is caused by one cause, the change in the pixel value due to the disturbance is often biased to either an increasing trend or a decreasing trend. Therefore, the image processing apparatus according to the present embodiment has an effect that even if the disturbance is slight, at least one of the images including the disturbance can be determined as a combination of images not suitable for image processing.

Embodiment 2. FIG.
In the present embodiment, differences from the first embodiment will be mainly described.
The same components as those described in Embodiment 1 are denoted by the same reference numerals, and the description thereof may be omitted.

A block configuration of the image processing apparatus 100a according to the present embodiment will be described with reference to FIG.
As shown in FIG. 7, the image processing apparatus 100 a according to the present embodiment includes a changed region extraction unit 150 in addition to the configuration of the image processing apparatus 100 described in the first embodiment. Further, the determination unit 1300a according to the present embodiment includes a large change determination unit 160 in addition to the configuration of the determination unit 1300 described in the first embodiment.

The change area extraction unit 150 acquires the first image 11 and the second image 12, changes the first feature value 1111C representing the first pixel 111, and the change representing the second pixel 121 corresponding to the first pixel 111. The second feature value 1211C for use is acquired. The change area extraction unit 150 has the degree of change representing the degree of change from the first change feature value 1111C to the second change feature value 1211C greater than the fourth threshold value of the second pixel 121 of the second change feature value 1211C. An area is extracted as a change area 1501. When the change degree of the feature value exceeds the fourth threshold value 204 in a certain image area, the change area extraction unit 150 determines that the image area is a change area. The change area extraction unit 150 is also referred to as a change area determination unit.
Here, the change degree of the feature value is a degree of change from the first change feature value 1111C to the second change feature value 1211C.

The determination unit 1300a determines whether the first image 11 and the second image 12 are suitable for image processing based on the degree of bias C and the area M of the change region.
The determination unit 1300a compares the degree C of bias with the third threshold value 203, and compares the area M of the change region with the fifth threshold value 205. The determination unit 1300a determines that the first image 11 and the second image 12 are not suitable for image processing when the degree of bias C is larger than the third threshold 203 or when the area M of the change region is larger than the fifth threshold 205. Is determined.

The image processing method and image processing S100a of the image processing apparatus 100a according to the present embodiment will be described with reference to FIG.
The increase area extraction process S110 and the decrease area extraction process S120 in FIG. 8 are the same as the increase area extraction process S110 and the decrease area extraction process S120 in FIG.

<Change Area Extraction Processing S121>
The change region extraction unit 150 is a region of the second pixel 121 of the second change feature value 1211C that is larger than the fourth threshold value 204 by subtracting the first change feature value 1111C from the second change feature value 1211C. Is executed as a change area 1501.

With reference to FIG. 4, the change area extraction processing S121 according to the present embodiment will be described in detail.
A configuration example in which the minimum unit of a certain image area is a pixel in the change area extraction process S121 will be described with reference to FIG. As shown in FIG. 4A, the change area extraction unit 150 subtracts the pixel value of the pixel at the same position in the first image 11 from the pixel value of the pixel in the second image 12 as the degree of change in the feature value. The absolute value of the measured value.
That is, the change area extraction unit 150 sets the pixel value of the second pixel 121, which is one pixel of the second image 12, as the second feature value for change 1211C, and is in the same position as the second pixel 121 in the first image 11. The degree of change in the feature value is calculated using the pixel value of the first pixel 111 as the first feature value for change 1111C.
The change area extraction unit 150 compares the degree of change with the fourth threshold value 204 and determines whether or not the change area 1501 for each pixel.

A configuration example in which the minimum unit of a certain image area is a block in the change area extraction process S121 will be described with reference to FIG. As shown in FIG. 4B, the change area extraction unit 150 sets the change degree of the feature value in the second pixel 121 at the same position as the pixel value of the first pixel 111 and the second pixel 121. The number of pixels in the block that satisfies the condition that the absolute value of the difference between the pixel values of pixels exceeds a certain threshold value. In this case, since the block that is the first pixel 111 has no pixel that satisfies the above condition, the change area extraction unit 150 sets 0 as the first change feature value 1111C, and the block that is the second pixel 121 satisfies the above condition. The number of pixels is set as the second feature value for change 1211C, and the change degree of the feature value is calculated.
The change area extraction unit 150 compares the degree of change with the fourth threshold value 204 and determines whether or not the area is the change area 1501 for each block. With this configuration, even if the pixel value may fluctuate greatly in a very small number of pixels due to random noise, there is an effect that the influence of random noise can be suppressed by determining in block units. The blocks do not necessarily have to be rectangular, and may have any shape.

A configuration example in which the minimum unit of a certain image area in the change area extraction process S121 is the entire image will be described with reference to FIG. As illustrated in FIG. 4C, the change area extraction unit 150 sets the change degree of the feature value at the same position in the second image 12 as the pixel value of the pixel of the first image 11 in the second pixel 121. Let the number of pixels satisfy the condition that the absolute value of the difference between the pixel values of the pixels exceeds a certain threshold. In this case, since the first image 11 that is the first pixel 111 has no pixel that satisfies the above condition, the change region extraction unit 150 sets 0 as the first feature value 1111C for change and the second image that is the second pixel 121. 12, the number of pixels satisfying the above condition is set as the second change feature value 1211C, and the change degree of the feature value is calculated.
The change area extraction unit 150 compares the degree of change with the fourth threshold value 204 and determines whether or not the entire image is the change area 1501.
This configuration has an effect that the algorithm is simplified and the processing is faster than the configuration example of the extraction method in which the minimum unit of a certain image area is a block.
Note that the change region 1501 may be determined by defining the minimum unit of a certain image region and the definition of the change degree of the feature value other than the method described above. The change area extraction unit 150 compares the first image 11 and the second image 12 and determines that the image area is the change area 1501 when the change degree of the feature value exceeds the fourth threshold value 204 in a certain image area. If it is, it can be set as arbitrary structures.

  Note that the fourth threshold value 204, the fifth threshold value 205, or other threshold values are set in advance and stored in the storage device of the image processing apparatus 100a. Alternatively, the fourth threshold value 204, the fifth threshold value 205, or other threshold values may be set by the user at the start of the image processing S100a. The fourth threshold value 204 is set in advance according to what value the degree of change is calculated.

  The change area extraction unit 150 outputs the extracted change area 1501 to the determination unit 1300a.

<Determination Processing S1300a>
The determination unit 1300a executes determination processing S1300a for determining whether the first image 11 and the second image 12 are suitable for image processing based on the degree of bias C and the area M of the change region.
The determination process S1300a includes an increase / decrease bias determination process S130, a large change determination process S131, and an image determination process S140a.

  The increase / decrease bias determination process S130 is the same as the increase / decrease bias determination process S130 of FIG.

<Large change determination process S131>
The large change determination unit 160 acquires the change region 1501 output from the change region extraction unit 150. The large change determination unit 160 determines that there is a large change when the area M of the change region exceeds the fifth threshold value 205. The large change determination unit 160 determines whether there is a large change or no large change based on the change region 1501 and outputs a determination result.

The large change determination process S131 will be described with reference to FIG.
FIG. 9A shows a configuration in which the area M of the change region is represented by the number of pixels of the change region 1501. It is assumed that the change area extraction unit 150 determines the change area 1501 as shown in FIG. 9A, and the area M of the change area is expressed by 197688 pixels. The large change determination unit 160 determines that there is a large change when 197688 pixels exceed the fifth threshold value 205.

  FIG. 9B is a configuration example in which the change area 1501 is determined in units of blocks. When the change region extraction unit 150 is configured to determine the change region 1501 in units of blocks, the large change determination unit 160 may determine whether there is a large change by representing the area M of the change region by the number of blocks. . It is assumed that the change area extraction unit 150 determines the change area 1501 as shown in FIG. 9B, and the area M of the change area is represented by 19 blocks. The large change determination unit 160 determines that there is a large change when 19 blocks exceed the fifth threshold value 205.

  FIG. 9C is a configuration example in which the change area 1501 is determined in units of the entire image. When the change area extraction unit 150 is configured to determine the change area 1501 in units of the entire image, if the entire image is the change area 1501, it is determined that there is a large change. This process is equivalent to a configuration in which it is determined that there is a large change when the area 1 of the entire image unit exceeds a fifth threshold value 205 (for example, 0). When the change area extraction unit 150 determines the change area 1501 as shown in (c) of FIG. 9 and the area M of the change area is one image, it is determined that there is a large change because the fifth threshold value 205 exceeds 0. The As described above, the large change determination unit 160 may have any configuration as long as it is configured to determine that there is a large change when the area M of the change region exceeds the fifth threshold.

<Image determination processing S140a>
The determination unit 1300a determines that the first image 11 and the second image 12 are not suitable for image processing when the degree of bias C is larger than the third threshold 203 or when the area M of the change region is larger than the fifth threshold 205. Is determined.
The image determination unit 140a determines whether the first image 11 and the second image 12 are suitable for image processing based on the determination result output from the increase / decrease bias determination unit 130 and the determination result output from the large change determination unit 160. Determine whether or not.
When the determination result output from the increase / decrease bias determination unit 130 indicates no bias and the determination result output from the large change determination unit 160 indicates no significant change, the image determination unit 140 a The image 12 is determined to be a combination of images suitable for image processing.

  This is the end of the description of the image processing S100a of the image processing apparatus 100a according to the present embodiment.

  The image processing apparatus 100a described in the present embodiment not only determines the first image and the second image as a combination of images that are not suitable for image processing when the degree of bias of the increase region and the decrease region exceeds a threshold value. Even when the area of the change region exceeds the threshold, the first image and the second image are determined to be a combination of images that are not suitable for image processing. Since the area of the change area becomes large when a disturbance over a wide range occurs on the image, the image processing apparatus having this configuration has at least one of a case where the disturbance is small or a wide range. Has an effect that two images including disturbance can be discriminated as a combination of images not suitable for image processing.

Next, the image processing S100b of the image processing device 100b having a function different from that of the image processing S100a of the image processing device 100a will be described.
In the determination process S1300a of the image processing S100a described above, the increase / decrease bias determination unit 130 has the same configuration as the increase / decrease bias determination unit 130 in the image processing apparatus 100 according to Embodiment 1. However, in the image processing S100b described below, the increase / decrease bias determination unit 130b acquires the change area 1501 in addition to the increase area 1101 and the decrease area 1201. Then, the increase / decrease bias determination unit 130b refers to the increase area 1101, the decrease area 1201, and the change area 1501 to determine whether there is an increase / decrease bias.

<Increasing Area Extraction Processing S110b>
As shown in FIG. 4B, the increase area extraction unit 110b sets a minimum unit of a certain image area as a block. The increase area extraction unit 110b is configured such that a difference obtained by subtracting the pixel value of the pixel at the same position of the first image 11 from the pixel value of the pixel of the second image 12 exceeds the sixth threshold value 206. The number of pixels in the block that satisfies Then, the increase area extraction unit 110b determines whether or not the increase amount of the feature value exceeds the seventh threshold value 207, thereby determining whether or not the increase area 1101 for each block.

<Decrease Area Extraction Processing S120b>
As shown in FIG. 4B, the reduced area extraction unit 120b sets a minimum unit of a certain image area as a block. The reduction area extraction unit 120b is configured such that the difference between the pixel value of the pixel in the first image 11 and the pixel value of the pixel at the same position in the second image 12 exceeds the sixth threshold value 206 as the reduction amount of the feature value. The number of pixels in the block that satisfies Then, the reduction area extraction unit 120b determines whether or not the block is the reduction area 1201 for each block by determining whether or not the reduction amount of the feature value exceeds the seventh threshold value 207.

<Change Area Extraction Processing S121b>
As shown in FIG. 4B, the change area extraction unit 150b sets a minimum unit of a certain image area as a block. The change area extraction unit 150b sets the degree of change of the feature value so that the absolute value of the difference between the pixel value of the pixel of the first image 11 and the pixel value of the pixel at the same position of the second image 12 exceeds the sixth threshold value 206. The number of pixels in the block that satisfies the condition. Then, the change area extraction unit 150b determines whether the block is the change area 1501 for each block by determining whether the change degree of the feature value exceeds the seventh threshold value 207.

<Increase / decrease bias determination processing S130b>
In the increase area extraction process S110b, the decrease area extraction process S120b, and the change area extraction process S121b described above, the increase area 1101 and the decrease area 1201 inevitably become the change area 1501. That is, the relationship between the area I of the increasing region, the area D of the decreasing region, and the area M of the changing region is M ≧ I + D.

The increase / decrease bias determination unit 130b calculates the bias degree C using the area I of the increase region, the area D of the decrease region, and the area M of the change region as shown in the following equation.
C = 0 (when I = D = 0)
C = ∞ (when I> D = 0 or D> I = 0)
C = I / M (when I>D> 0)
C = D / M (when D ≧ I> 0)

  The increase / decrease bias determination unit 130b determines whether or not there is a bias by comparing the bias degree C with the third threshold value 203. The increase / decrease bias determination unit 130b determines that there is a bias when the ratio of the area I of the increase region or the area D of the decrease region to the area M of the change region exceeds the third threshold value 203. The ratio of the area I of the increase region or the area D of the decrease region to 0 (less than 1 and less than 1) can be set as the third threshold value 203.

Next, the image processing S100c of the image processing device 100c different from the image processing S100a and S100b of the image processing devices 100a and 100b described in the present embodiment will be described.
In the image processing S100a and S100b of the image processing apparatuses 100a and 100b described in the present embodiment, the change region extraction units 150a and 150b are configured to operate independently of the increase region extraction unit 110 and the decrease region extraction unit 120. . However, the image processing apparatus 100c may include a change region extraction unit 150c that operates using the processing of the increase region extraction unit 110 and the decrease region extraction unit 120.

<Change Area Extraction Processing S121c>
The increase area extraction unit 110 determines that the image area is the increase area 1101 when the increase amount of the feature value exceeds the first threshold 201 in a certain image area. Further, the reduction area extraction unit 120 determines that the image area is the reduction area 1201 when the reduction amount of the feature value exceeds the second threshold value 202 in a certain image area. At this time, the first threshold 201 and the second threshold 202 are preferably equal.
The change area extraction unit 150c may set the change degree of the feature value as a larger value of the feature value increase amount and the feature value decrease amount. Then, the change area extraction unit 150c may execute a change area extraction process S121c that determines the image area as the change area 1501 when the change degree of the feature value exceeds the fourth threshold value 204 in a certain image area.

This change area extraction process S121c has an effect that the algorithm for calculating the change degree of the feature value is simplified and the process becomes faster. The change area 1501 determined by the change area extraction process S121c is a union of the increase area 1101 and the decrease area 1201. Therefore, the change area extraction process S121c is equivalent to a process in which the change area extraction unit determines that the union of the increase area 1101 and the decrease area 1201 is the change area 1501. The change area extraction process S121c is also equivalent to the following process. In this process, the change degree of the feature value of the image area included in the increase area 1101 or the decrease area 1201 is set to 1, otherwise the change degree of the feature value is set to 0, and the change degree of the feature value exceeds a threshold value less than 1. In this case, the image area is determined to be a change area.
In this change area extraction process S121c, a change in which the increase or decrease is not clear is not evaluated as a change. Therefore, the static noise indicating the property that the increase and decrease are mixed at random is not regarded as a change, and a change in luminous intensity or the like. Temporary disturbance due to the cause will be regarded as a change. Therefore, according to this change area extraction processing S121c, there is an effect that two images including the disturbance in at least one of them can be efficiently determined as a combination of images not suitable for image processing.

Embodiment 3 FIG.
In the present embodiment, differences from Embodiment 2 will be mainly described.
In the present embodiment, an image processing apparatus 100d that suppresses erroneous detection that occurs when a moving object is detected by determining whether two images forming a moving image are a combination of images that are not suitable for image processing will be described. .
In this embodiment, the same components as those described in Embodiment 2 are denoted by the same reference numerals, and the description thereof may be omitted.

A block configuration of the image processing apparatus 100d according to the present embodiment will be described with reference to FIG.
As shown in FIG. 10, the image processing apparatus 100d according to the present embodiment includes a moving object detection unit 170 in addition to the configuration of the image processing apparatus 100a described in the second embodiment.

  The moving body detection unit 170 acquires the first image 11 and the second image 12, and acquires a determination result 1301 as to whether or not the first image 11 and the second image 12 are suitable for image processing from the determination unit 1300a. . The moving object detection unit 170 executes a moving object detection process using the first image 11 and the second image 12 when it is determined that the first image 11 and the second image 12 are suitable for image processing, and the executed moving object The result of the detection process is set as a detection result 1701. When it is determined that the first image 11 and the second image 12 are not suitable for image processing, the moving object detection unit 170 sets a detection failure in the detection result 1701 without executing the moving object detection process.

The image processing S100d of the image processing apparatus 100d according to the present embodiment will be described with reference to FIG.
The increase area extraction process S110, the decrease area extraction process S120, the change area extraction process S121, the increase / decrease bias determination process S130, the large change determination process S131, and the image determination process S140 are the same as those described in the second embodiment.

<Moving object detection process S150>
The moving body detection unit 170 acquires the first image 11 and the second image 12, and acquires a determination result 1301 as to whether or not the first image 11 and the second image 12 are suitable for image processing from the determination unit 1300a. . The moving object detection unit 170 executes a moving object detection process using the first image 11 and the second image 12 when it is determined that the first image 11 and the second image 12 are suitable for image processing, and the moving object detection process is performed. Is set as a detection result 1701. When it is determined that the first image 11 and the second image 12 are not suitable for image processing, the moving object detection unit 170 sets detection failure to the detection result 1701.
In the moving object detection process S150, the moving object detection process is performed only when the image determination unit 140 does not determine that the first image 11 and the second image 12 are a combination of images not suitable for image processing.

  Alternatively, the moving object detection unit 170 compares the first image 11 and the second image 12 to detect a moving object. When the determination result 1301 output from the image determination unit 140 determines that the first image 11 and the second image 12 are a combination of images that are not suitable for image processing, the moving object detection unit 170 may invalidate the moving object detection. . That is, the moving object detection unit 170 discards the content of the detection result 1701 of the moving object detection process only when the image determination unit 140 determines that the first image 11 and the second image 12 are a combination of images that are not suitable for image processing. The moving object may not be detected.

  In addition to the moving object detection process as described above, when the first image 11 and the second image 12 are determined to be a combination of images that are not suitable for image processing, the moving object detection process becomes invalid. Any processing method can be used. The fact that the moving object detection process is invalid means that even if the moving object detection process is executed, it is not executed or the moving object detection process is not executed.

  The image processing apparatus 100d according to the present embodiment compares the first image 11 and the second image 12, and determines whether the first image 11 and the second image 12 are a combination of images that are not suitable for image processing. . Then, the image processing apparatus 100d executes processing for invalidating the moving object detection when the first image 11 and the second image 12 are a combination of images that are not suitable for image processing. For this reason, according to the image processing device 100d, it is possible to suppress erroneous detection in which at least one of the first image 11 and the second image 12 is processed as a correctly detected moving object detected from two images including disturbance. There is an effect to.

  In the image processing apparatus described in the first to third embodiments, the pixel value may be any component value. For example, if the target is a black and white image, the pixel value can be a value of a luminance (shading) component. If the target is an RGB color image, the pixel value can be any one of a red (R) component, a green (G) component, and a blue (B) component. If the target is a YCbCr color image, the pixel value can be any one of a luminance (Y) component, a blue color difference (Cb) component, and a red color difference (Cr) component.

  The image processing apparatus described in Embodiments 1 to 3 can also be applied to two images that form a moving image generated by performing some processing on the moving image. For example, each image forming a black and white moving image is replaced with an image obtained by extracting only a high frequency component in the image. Two images forming a moving image obtained by this replacement can be handled by the image processing apparatus described in the first to third embodiments. In this case, the pixel value handled by the image processing apparatus described in the first to third embodiments is the value of the high frequency component of the image. Thus, in the image processing apparatus described in the first to third embodiments, the pixel value can be a value of an arbitrary component.

  In the image processing apparatus described in the first to third embodiments, the second image associated with the new time information and the first image associated with the old time information among the two images forming the moving image are used. It was configured to work. In order to continuously operate the image processing apparatus, another control apparatus for controlling the image processing apparatus is prepared, and the image processing apparatus is operated while sequentially changing the first image and the second image. The control device may be configured to perform control. In particular, an image processing apparatus that operates in real time on a live moving image by configuring the second image as the latest image and the first image as the second image used in the operation one step before the sequential control. realizable.

In the image processing apparatus described in the first to third embodiments, the increase area extraction unit, the decrease area extraction unit, the change area extraction unit, the increase / decrease bias determination unit, the large change determination unit, the image determination unit, and the moving object detection unit are respectively provided. The image processing apparatus is configured as an independent functional block. However, the image processing apparatus may not be configured as described above, and the configuration of the image processing apparatus is arbitrary. The increase region extraction unit, the decrease region extraction unit, and the change region extraction unit may be realized by one functional block, and the increase / decrease bias determination unit, the large change determination unit, and the image determination unit may be realized by one functional block. The functional blocks of the image processing apparatus are arbitrary as long as the functions described in the first to third embodiments can be realized. These functional blocks may be configured in any other combination. Alternatively, the image processing apparatus may be configured with an arbitrary block configuration.
The image processing apparatus may be an image processing system including a plurality of apparatuses instead of a single apparatus.

Further, although Embodiments 1 to 3 have been described, a plurality of these three embodiments may be partially combined. Alternatively, one of these three embodiments may be partially implemented. In addition, these three embodiments may be implemented in any combination as a whole or in part.
In addition, said embodiment is an essentially preferable illustration, Comprising: It does not intend restrict | limiting the range of this invention, its application thing, or a use, A various change is possible as needed. .

  11 First image, 12 Second image, 100, 100a, 100b, 100c, 100d Image processing device, 110, 110b Increased region extraction unit, 111 First pixel, 120, 120b Decrease region extraction unit, 121 Second pixel, 130 , 130b Increase / decrease bias determination unit, 140, 140a Image determination unit, 150, 150a, 150b, 150c Change area extraction unit, 160 Large change determination unit, 170 Moving object detection unit, 201 First threshold value, 202 Second threshold value, 203 Third Threshold, 204 fourth threshold, 205 fifth threshold, 206 sixth threshold, 207 seventh threshold, 901 processor, 902 auxiliary storage device, 903 memory, 904 communication device, 905 input interface, 906 display interface, 907 input device, 908 Display, 910 signal line, 91 , 912 cable, 9041 receiver, 9042 transmitter, 1101 increasing area, 1111A increasing first feature value, 1111B decreasing first feature value, 1111C changing first feature value, 1201 decreasing area, 1211A increasing second feature value, 1211B Second characteristic value for reduction, 1211C Second characteristic value for change, 1301 judgment result, 1300, 1300a, 1300b, 1300c judgment unit, 1501 change area, 1701 detection result, C bias degree, D area of decrease area, I increase Area area, M Change area area, S100, S100a, S100b, S100c, S100d Image processing, S110, S110b Increase area extraction process, S120, S120b Decrease area extraction process, S121, S121b, S121c Change area extraction process, S1 0, S130b decrease deviation determination process, S131 very assessment process, S140, S140a image determining process, S150 motion detection process, S1300, S1300a determination process.

Claims (12)

A first image that is a two image forming a moving image and a second image are acquired, a first feature value for increase that represents a first pixel that is at least one pixel included in the first image, and the first image A second pixel that is at least one pixel included in the two images and is compared with the second feature value for increase that represents the second pixel corresponding to the first pixel; An increase area extraction unit that extracts an area of the second pixel of the second feature value for increase greater than a first threshold as an increase area;
The first image and the second image are acquired, and the first feature value for reduction representing the first pixel is compared with the second feature value for reduction representing the second pixel corresponding to the first pixel. A reduction region extraction unit that extracts, as a reduction region, the region of the second pixel of the second feature value for reduction that is greater than a second threshold value from the first feature value for reduction;
The degree of bias representing the degree of bias between the area of the increase region and the area of the decrease region is calculated, and whether or not the first image and the second image are suitable for image processing based on the calculated bias degree and a determining unit,
The determination unit
When both the area of the increase area and the area of the decrease area are zero, it is determined that the first image and the second image are suitable for image processing, and either the area of the increase area or the area of the decrease area is An image processing apparatus that determines that the first image and the second image are not suitable for image processing when only one of them is zero . The determination unit
The bias degree and a third threshold value are compared, and when the bias degree is larger than the third threshold value, it is determined that the first image and the second image are not suitable for image processing. Image processing device. The image processing apparatus further includes:
The first image and the second image are acquired, and the first characteristic value for change representing the first pixel is compared with the second characteristic value for change representing the second pixel corresponding to the first pixel. Then, a region of the second pixel of the second feature value for change that is greater than a fourth threshold is a change region that represents a degree of change from the first feature value for change to the second feature value for change. A change area extraction unit for extracting;
The determination unit
The image processing apparatus according to claim 1, wherein it is determined whether the first image and the second image are suitable for image processing based on the degree of bias and the area of the change region. The determination unit
The bias degree is compared with a third threshold value, and the area of the change region is compared with a fifth threshold value. When the bias degree is larger than the third threshold value, or the area of the change region is a fifth threshold value. The image processing apparatus according to claim 3, wherein the first image and the second image are determined to be unsuitable for image processing when larger. The determination unit
5. The image processing apparatus according to claim 1, wherein a value of a ratio between an area of the increased region and an area of the decreased region is calculated as the degree of bias. The determination unit
5. The ratio of the addition area obtained by adding the area of the increase area and the area of the decrease area and the area of the increase area or the decrease area is calculated as the degree of bias. An image processing apparatus according to 1. The first pixel is one pixel included in the first image,
7. The device according to claim 1, wherein the second pixel is included in the second image and is one pixel at a position corresponding to the first pixel in the first image in the second image. Image processing apparatus. The first pixel is a plurality of pixels included in the first image,
7. The first pixel according to claim 1, wherein the second pixel is included in the second image, and is a plurality of pixels at positions corresponding to the first pixel in the first image in the second image. Image processing apparatus. The first pixels are all pixels included in the first image,
The image processing apparatus according to claim 8, wherein the second pixel is all pixels included in the second image. The image processing apparatus includes:
The first image and the second image are acquired, and a determination result as to whether or not the first image and the second image are suitable for image processing is acquired from the determination unit, and the first image and the second image are acquired. When it is determined that the second image is suitable for image processing, a moving object detection process using the first image and the second image is executed, and a result of the moving object detection process is set as a detection result. 10. The image processing apparatus according to claim 1, further comprising a moving body detection unit that sets detection failure in the detection result when it is determined that one image and the second image are not suitable for image processing. . The increase area extraction unit acquires a first image and a second image that are two images forming a moving image, and indicates a first increase pixel that represents a first pixel that is at least one pixel included in the first image. The feature value is compared with an increase second feature value representing the second pixel corresponding to the first pixel, which is a second pixel that is at least one pixel included in the second image, and the increase value Extracting an area of the second pixel of the second feature value for increase greater than a first threshold as an increase area from the first feature value;
A reduction area extraction unit acquires the first image and the second image, and a first feature value for reduction that represents a first pixel that is at least one pixel included in the first image; and the first pixel Is compared with a second feature value for reduction representing the second pixel corresponding to the second feature value of the second feature value of the second feature value for reduction greater than a second threshold value. Extract the area as a decreasing area,
The determination unit calculates a degree of bias representing a degree of bias between the area of the increased region and the area of the decreased region, and based on the calculated degree of bias, both the area of the increased region and the area of the decreased region are zero. In this case, it is determined that the first image and the second image are suitable for image processing, and when only one of the area of the increase region and the area of the decrease region is zero, the first image and An image processing method for determining that the second image is not suitable for image processing. A first image that is a two image forming a moving image and a second image are acquired, a first feature value for increase that represents a first pixel that is at least one pixel included in the first image, and the first image A second pixel that is at least one pixel included in the two images and is compared with the second feature value for increase that represents the second pixel corresponding to the first pixel; An increase area extraction process for extracting, as an increase area, the area of the second pixel of the second feature value for increase greater than a first threshold;
The first image and the second image are acquired, the first feature value for reduction representing the first pixel that is at least one pixel included in the first image, and the second corresponding to the first pixel The second feature value for reduction representing a pixel is compared, and the region of the second pixel of the second feature value for reduction that is larger than the second threshold value is extracted as a reduction region from the first feature value for reduction. Reduction area extraction processing to
Calculate the degree of bias representing the degree of bias between the area of the increased region and the area of the decreased region, and based on the calculated degree of bias, when the area of the increased region and the area of the decreased region are both zero, When it is determined that the first image and the second image are suitable for image processing, and only one of the area of the increase region and the area of the decrease region is zero, the first image and the second image Is a program that causes a computer to execute a determination process that determines that is not suitable for image processing .

Images