JP6798304B2 - Image processing device, program - Google Patents

Description

The present invention relates to an image processing apparatus and a program used for image processing.

In Patent Document 1, the first image and the second image, which are two images forming a moving image, are acquired, and the degree of bias between the area of the feature value increasing region representing the pixel and the area of the feature value decreasing region is determined. It is disclosed to calculate the degree of bias represented. Based on this degree of bias, it is determined whether or not the first image and the second image are suitable for image processing. As a result, even if the disturbance has a small area, the disturbance can be appropriately detected, so that the combination of images that is not suitable for image processing can be determined with high accuracy.

International Publication No. 2016/139717

There is an image processing device that processes a moving image with a digital computer and detects a person included in the moving image. In such an image processing device, a person is detected by comparing the first image and the second image taken continuously. Specifically, the first image and the second image are compared to detect a portion where the pixel value has changed, and a person is detected. However, if a person detects a change in pixel value caused by a disturbance, a false detection will occur. An image containing disturbance must be determined as an image that is not suitable for image processing for detecting a person.

Disturbances occur, for example, when the lights flash, the lights that were off are turned on, or the brightness is amplified or reduced by internal processing of the camera. Therefore, disturbances often cause changes in pixel values over a wide area of the image. Therefore, when a change in the pixel value over a wide range of the image is found, it is conceivable to determine that the image is not suitable for image processing.

However, for example, when there are people in the entire room, it is necessary to detect that "there are many people", but it is misjudged as a disturbance because a wide range of pixel changes occur. It was. When the change in pixel value is widespread, there is a need for a technique for determining whether the change is due to the existence of "many people" or due to disturbance.

The present invention has been made to solve the above-mentioned problems, and when a change in pixel value is found in a wide range of an image, it is disturbed whether the change is due to the existence of "many people". It is an object of the present invention to provide an image processing apparatus and a program capable of determining whether or not the image is due to the above.

In the image processing apparatus according to the present invention, the second image taken after the first image is compared with the first image, and the increased region is caused by disturbance for each increased region where the pixel value is increased. The pixel value is calculated by comparing the second image with the first image and the increasing region disturbance calculation unit that calculates the increased region disturbance that is the generated probability and calculates the total of the obtained increased region disturbance. For each reduction region, which is a reduced region, a reduction region disturbance degree calculation unit that calculates the reduction region disturbance degree, which is the probability that the reduction region is caused by a disturbance, and calculates the total of the obtained reduction region disturbance degree, and the reduction region disturbance degree calculation unit. When the sum of the sum of the disturbance degrees of the increasing region and the sum of the disturbance degrees of the decreasing region exceeds a predetermined value, it is determined that the combination of the first image and the second image is not suitable for image processing. It includes a determination unit. Then, the total sum of the increasing area disturbances is smaller than the increasing area disturbance calculated for the total increased area by obtaining the total increased area which is the total area of the increasing areas, and the total of the decreasing area disturbances is said to be. The total reduced area, which is the total area of the reduced regions, is obtained and is smaller than the reduced region disturbance degree calculated for the total reduced area.
In the program according to the present invention, the second image taken after the first image is compared with the first image, and the increased region is generated by disturbance for each increased region where the pixel value is increased. The increase region disturbance degree, which is the accuracy, is calculated, and the total increase region disturbance degree obtained is obtained, and the total increase area in which the sum total of the increase region disturbance degree is the sum of the areas of the increase region is obtained. The process of calculating so as to be smaller than the calculated increase region disturbance degree, and comparing the second image with the first image, the decrease region is generated by the disturbance for each decrease region which is the region where the pixel value is decreased. The reduced region disturbance degree, which is the certainty, is calculated, and the total of the obtained reduced region disturbance degree is calculated, and the total reduced area, in which the total of the reduced region disturbance degree is the total area of the reduced region, is obtained. When the process of calculating so as to be smaller than the reduced region disturbance degree calculated for, the sum of the increased region disturbance degree and the sum of the reduced region disturbance degree exceeds a predetermined value, the first A computer is made to execute a process of determining that the combination of the first image and the second image is not suitable for image processing.

According to the present invention, when the area of the increasing region where the pixel value is increased and the area of the decreasing region where the pixel value is decreased are small, the degree of disturbance of the increasing region and the degree of disturbance of the decreasing region are suppressed to be low, and the areas of the increasing region and the decreasing region are large. In this case, high values are calculated as the degree of disturbance in the increasing area and the degree of disturbance in the decreasing area. Therefore, when the pixel value increase region or the pixel value decrease region covers a wide range, it can be accurately determined whether it is due to "a lot of people" or "disturbance".

It is a block diagram of the image processing apparatus which concerns on embodiment. It is a flowchart which shows the operation of the image processing apparatus which concerns on embodiment. It is a figure which shows the example of the 1st image, 2nd image and the pixel value increase / decrease map. It is a figure which shows the configuration which realized the image processing apparatus which concerns on embodiment by hardware, and the configuration which realized by software.

The image processing apparatus and the program according to the embodiment of the present invention will be described with reference to the drawings. The same or corresponding components may be designated by the same reference numerals and the description may be omitted.

Embodiment.
FIG. 1 is a block diagram of an image processing device according to an embodiment. The image processing device includes an increase region disturbance degree calculation unit 12 and a decrease region disturbance degree calculation unit 14. Images constituting a moving image are transmitted one after another to the increasing region disturbance degree calculation unit 12 and the decreasing region disturbance degree calculation unit 14. FIG. 1 shows that the first image and the second image were transmitted to the increasing region disturbance degree calculation unit 12 and the decreasing region disturbance degree calculation unit 14. The first image and the second image are two consecutive images constituting a moving image.

The outputs of the increasing area disturbance degree calculation unit 12 and the decreasing area disturbance degree calculation unit 14 are transmitted to the determination unit 20. The determination unit 20 determines whether or not the combination of the first image and the second image is suitable for image processing from the outputs of the increase region disturbance degree calculation unit 12 and the decrease region disturbance degree calculation unit 14. The determination result is transmitted to the control unit 22. When the determination unit 20 determines that the combination of the first image and the second image is suitable for image processing, the control unit 22 detects a person with these images and controls the lighting according to the detection result. On the other hand, when the determination unit 20 determines that the combination of the first image and the second image is not suitable for image processing, the control unit 22 does not detect a person with these images.

FIG. 2 is a flowchart showing the operation of the image processing apparatus according to the embodiment of the present invention. The operation of the image processing apparatus will be described in detail with reference to this flowchart.

(Image acquisition)
First, in step S1, the first image and the second image are transmitted from the camera to the increasing region disturbance degree calculation unit 12 and the decreasing region disturbance degree calculation unit 14. The first image and the second image are two images constituting a moving image. The first image is an image taken first, and the second image is an image taken after the first image. For example, the first image and the second image are two images acquired at intervals of a fraction of a second or a few tenths of a second. FIG. 3 shows examples of the first image and the second image. The first image includes a black region R1, a white region R2 within the region R1, and a white region R3 to the right of the region R1. The second image includes a triangular white region R4, a black region R5 next to the region R4, a white region R6 within the region R5, and a white region R7 next to the region R5.

(Increase area and decrease area)
Then, the process proceeds to step S2. In step S2, the increasing region disturbance degree calculation unit 12 compares the second image with the first image and identifies the increasing region, which is the region where the pixel value has increased. A small number of large areas of increase are identified when there is disturbance due to blinking lights, turning on lights that were off, or amplifying or reducing brightness by internal processing of the camera. On the other hand, if there are "many people" and they move individually, many small areas of increase are identified.

Further, in step S2, the reduction region disturbance degree calculation unit 14 compares the second image with the first image to specify the reduction region, which is the region where the pixel value is reduced. A small number of large reduction areas are identified when there is disturbance due to blinking lights, turning off lights that were on, or amplifying or reducing brightness by internal processing of the camera. On the other hand, if there are "many people" and they move individually, many small areas of decline are identified.

FIG. 3 shows a pixel value increase / decrease map. The pixel value increase / decrease map is created by the increase area disturbance degree calculation unit 12 and the decrease area disturbance degree calculation unit 14. The pixel value increase / decrease map is a map that compares the second image with the first image and specifies an increase region that is a portion where the pixel value is increased and a decrease region that is a portion where the pixel value is decreased. FIG. 3 shows two increasing regions I1 and I2 and one decreasing region D1.

In step S2, specifying an increase region, a decrease region, or a region that is neither an increase region nor a decrease region for each pixel may increase the amount of calculation, which is not preferable. Therefore, the increase region disturbance degree calculation unit 12 and the decrease region disturbance degree calculation unit 14 divide the first image and the second image into an aggregate of blocks having a plurality of predetermined pixels as constituent elements, and divide each block into blocks. It may be determined whether or not it is an increase region or a decrease region. As a result, the amount of calculation can be reduced. For example, a region composed of a total of 100 pixels having 10 pixels in the horizontal direction and 10 pixels in the vertical direction can be regarded as one block.

Three types of methods can be considered as specific methods for determining whether or not each block is an increase area or a decrease area. In the first method, the increasing region disturbance degree calculation unit 12 sets the block as an increasing region when the average value of the pixel values in the block increases, and the decreasing region disturbance degree calculating unit 14 sets the average value of the pixel values in the block. When is reduced, the block is set as the reduced area. The pixel value may increase or decrease in a very small number of pixels due to noise. By calculating the average value of the pixel values in the block, the effect of canceling the influence of noise can be expected.

In the second method, when the number of pixels whose pixel values have increased in the block exceeds the threshold value, the increasing region disturbance degree calculation unit 12 sets the block as the increasing region, and the decreasing region disturbance degree calculating unit 14 uses the decreasing region disturbance degree calculating unit 14 in the block. When the number of pixels whose pixel value has decreased exceeds the threshold value, the block is set as the decrease region. According to this method, even if the pixel value decreases in one pixel in the block due to noise, it can be determined that it is an increase region when the pixel value increases in a plurality of pixels in the block. Further, even if the pixel value increases in one pixel in the block due to noise, it can be determined that the area is decreased when the pixel value decreases in a plurality of pixels in the block.

In the third method, the increasing region disturbance degree calculation unit 12 blocks the block when the value obtained by subtracting the number of pixels whose pixel value has decreased from the number of pixels whose pixel value has increased in the block is larger than a predetermined value. Is the increasing area. Further, the reduction area disturbance degree calculation unit 14 reduces the block when the value obtained by subtracting the number of pixels whose pixel value has increased from the number of pixels whose pixel value has decreased in the block is larger than a predetermined value. And. This method can also reduce the influence of noise.

(Increased area disturbance and decreased area disturbance)
Then, the process proceeds to step S3. In step 3, the increasing region disturbance degree calculation unit 12 calculates the increasing region disturbance degree for each increasing region, and calculates the total sum of the obtained increasing region disturbance degree. The degree of disturbance in the increasing region represents the probability that the increasing region is caused by the disturbance. As mentioned above, based on the idea that there are a few large areas of increase in the presence of disturbances, and a large number of small areas of increase when there are "many people" and they move individually. , The larger the area of the increasing area, the larger the increasing area disturbance degree is given. Therefore, it is highly probable that the increased region to which a large increase region disturbance degree is given is caused by the disturbance. On the contrary, the increased region to which a small increase region disturbance degree is given has a low probability of being caused by disturbance and a high probability of being caused by human movement.

The sum of the increase areas when a small number of large increase areas are generated by the disturbance is the increase area when a large number of small increase areas occur due to the existence of "many people" and their individual movements. Make it larger than the sum of the degree of disturbance. For example, the sum of the areas of the increase area when several large increase areas are generated due to the disturbance and the total area of the increase areas when many small increase areas are generated due to the movement of "many people" If they are equal, the sum of the former increasing area disturbances is made larger than the sum of the latter increasing area disturbances. In other words, the sum of the increased area disturbances is the total increased area, which is the total area of the increased areas, and is smaller than the increased area disturbance calculated for the total increased area. It is preferable that the total sum of the degree of disturbance in the increasing region becomes smaller than the degree of disturbance in the increasing region calculated for the total increased area as the number of increasing regions increases. As a result, it is possible to reduce the sum of the degree of disturbance in the increased area when "a lot of people" move and a large number of increased areas with a small area are generated.

Further, in step S3, the reduction region disturbance degree calculation unit 14 calculates the reduction region disturbance degree for each reduction region, and calculates the sum of the obtained reduction region disturbance degree. The reduced region disturbance degree represents the probability that the reduced region is caused by the disturbance. Large reduced region A reduced region with a degree of disturbance is more likely to be caused by disturbance. On the contrary, the reduced area to which a small degree of disturbance is given is highly likely to be caused by the movement of a person. The method for calculating the degree of disturbance in the decreasing region is the content obtained by replacing "increase" in the degree of disturbance in the increasing region with "decrease". That is, the total sum of the reduced area disturbances is smaller than the reduced area disturbance calculated for the total reduced area by obtaining the total reduced area, which is the total area of the reduced areas. It is preferable that the total sum of the reduced region disturbances is smaller than the reduced region disturbance calculated for the total reduced area as the number of reduced regions increases.

Since various specific calculation methods for the degree of disturbance in the increasing region and the degree of disturbance in the decreasing region can be considered, some examples are given. For example, the increasing region disturbance degree calculation unit 12 may calculate the increasing region disturbance degree for each increasing region by using a table that gives a larger increasing region disturbance degree as the area of the increasing region increases. Further, the reduced region disturbance degree calculation unit 14 may calculate the reduced region disturbance degree for each reduced region by using a table that gives a larger reduced region disturbance degree as the area of the reduced region increases. For example, if the area of the increasing area or the decreasing area is less than A1, the disturbance degree S1 is given, if the area is A1 or more and less than A2, the disturbance degree S2 larger than the disturbance degree S1 is given, and if the area is A2 or more, the disturbance degree S2 is given. The degree S3 may be given. The calculation process can be simplified by using the table.

There is another calculation method. The increase region disturbance degree calculation unit 12 may use a value obtained by multiplying the area of the increase region a plurality of times as the increase region disturbance degree. That is, a power with an area as the base and a natural number larger than 1 as an exponent (or exponentiation) may be used as the degree of disturbance in the increasing region. The reduction region disturbance degree calculation unit 14 may use a value obtained by multiplying the area of the reduction region a plurality of times as the reduction region disturbance degree. That is, a power with an area as the base and a natural number larger than 1 as an exponent (or exponentiation) may be used as the degree of disturbance in the decreasing region. Further, in the power calculation, the exponent may be a real number as well as a natural number. However, the index must be greater than 1. This makes the sum of the increasing area disturbances smaller than the increasing area disturbances calculated for the total increased area by finding the total increased area, which is the total area of the increasing areas, and also reducing the total decreasing area disturbances. This is because the total reduced area, which is the total area of the regions, is calculated and made smaller than the reduced region disturbance degree calculated for the total reduced area.

In any case, the larger the area of the increasing region is, the larger the increasing region disturbance degree is given, and the larger the area of the decreasing region is, the larger the decreasing region disturbance degree is given. This can be expressed by a mathematical formula as follows.
Number of areas to increase: n
Area of i-th increase region: Ai (i = 1, 2, ... n)
Disturbance degree in the i-th increasing region: Si (i = 1, 2, ... n)
For any i, j, k, if Ak = Ai + Aj, then Sk> Si + Sj holds.
The value obtained by dividing 1 by the area of the increasing region or the area of the decreasing region is used as the degree of disturbance in the increasing region or the degree of disturbance in the decreasing region, and the square root of the area of the increasing region or the area of the decreasing region is the degree of disturbance in the increasing region or the disturbance in the decreasing region. There is no such thing as a degree.

By the way, for example, it is assumed that the first increase region is close to a circle and the second increase region has an elongated shape having the same area as the first increase region and a longer outer circumference than the first increase region. In this case, the first increase region is highly probable to be caused by disturbance, while the second increase region is highly probable to be caused by the movement of a person. That is, when there are two increasing regions having the same area, it is preferable to increase or decrease the degree of disturbance of the increasing region according to their outer circumferences. Therefore, it is preferable that the increasing region disturbance degree calculation unit 12 obtains the outer circumference for each increasing region, and the smaller the outer circumference, the higher the increasing region disturbance degree. Since the same argument holds for the reduced region, it is preferable that the reduced region disturbance degree calculation unit 14 obtains the outer circumference for each reduced region, and the smaller the outer circumference, the higher the reduced region disturbance degree.

Since various concrete means for reflecting the outer circumference of the increasing region and the decreasing region in the increasing region disturbance degree and the decreasing region disturbance degree can be considered, some examples will be given. For example, the degree of disturbance of the increasing region having a certain area is increased as the outer circumference of the increasing region is shorter, and is increased as the area of the increasing region is larger. In this case, a large increase region disturbance degree is given to the increase region having a short outer circumference and a large area. For example, the increased region disturbance degree can be calculated by using a table in which the length of the outer circumference is associated with the increased region disturbance degree. Alternatively, the value obtained by dividing 1 by the outer circumference and adding it to the disturbance degree calculated from the area can be used as the increase region disturbance degree. In this way, for the first increasing region and the second increasing region having the same area as described above, the increasing region disturbance degree of the first increasing region becomes larger than the increasing region disturbance degree of the second increasing region.

The technique for reflecting the outer circumference of the increasing region in the degree of disturbance in the increasing region is expressed by a mathematical formula as follows.
Number of areas to increase: n
Area of i-th increase region: Ai (i = 1, 2, ... n)
Outer circumference of the i-th increasing region: Bi (i = 1, 2, ... n)
Disturbance degree in the i-th increasing region: Si (i = 1, 2, ... n)
For any i, j, if Ai = Aj and Bi <Bj, then Si> Sj.
If Ai> Aj and Bi = Bj, then Si> Sj.
The reduction region can be calculated in the same manner.

The accuracy of the disturbance degree can be improved by not only determining the disturbance degree of the increase region according to the area of the increase region but also calculating the disturbance degree of the increase region in consideration of the outer circumference of the increase region as described above. For example, the degree of disturbance of the increasing region is calculated according to the area of the increasing region, the degree of disturbance of the increasing region is calculated from the outer circumference of the increasing region, and these are added together or the product of these is calculated before the sum is calculated. It may be the degree of disturbance in the increasing area of.

(Judgment by the judgment unit)
In step S4, the increasing region disturbance degree calculation unit 12 transmits the sum total of the increasing region disturbance degree to the determination unit 20, and the decreasing region disturbance degree calculation unit 14 transmits the total sum of the decreasing region disturbance degree to the determination unit 20. Then, the determination unit 20 determines that the combination of the first image and the second image is suitable for image processing when the sum of the sum of the disturbance degrees of the increasing region and the sum of the disturbance degrees of the decreasing region exceeds a predetermined value. Judge that there is no. In this case, the process proceeds to step S5, and the determination unit 20 transmits a determination result that the combination of the first image and the second image is not suitable for human detection to the control unit 22. In response to this, the control unit 22 does not detect a person by combining the first image and the second image. On the other hand, the determination unit 20 states that the combination of the first image and the second image is suitable for image processing when the sum of the sum of the disturbance degrees of the increasing region and the sum of the disturbance degrees of the decreasing region does not exceed a predetermined value. The determination result is transmitted to the control unit 22. In response to this, the control unit 22 detects a person by combining the first image and the second image, and controls the lighting based on the detection result.

As described above, according to the image processing apparatus according to the embodiment of the present invention, when a change in the pixel value is found in a wide range of the image due to the presence of "many people", the degree of disturbance in the increasing region and the decrease are found. Area disturbance is kept low. On the other hand, when a change in the pixel value is found in a wide range of the image due to the disturbance, a high value is calculated as the degree of disturbance in the increasing region and the degree of disturbance in the decreasing region. Therefore, when the increasing area or the decreasing area covers a wide area, it can be accurately determined whether it is due to "a lot of people" or "disturbance". By using this image processing device in combination with a device for determining the presence or absence of a moving object, it is possible to reduce an error in determining the presence or absence of a moving object.

In a conventional lighting device that tracks a person and illuminates the place where the person is, when a person enters from the vicinity of the entrance of the room, the person near the entrance is detected and the surroundings of the person are illuminated by the illumination. After that, when a lot of people enter the room and the increasing area and the decreasing area increase, it is judged that it is a disturbance and the person is not detected. If a person goes to the back of the room when the person is no longer detected, the lights will not turn on in the back of the room. On the other hand, according to the image processing apparatus according to the embodiment of the present invention, even if "a lot of people" enter the room, it can be determined that it is not a disturbance but "a lot of people", so that this problem is solved. it can.

(Modification example)
The image processing apparatus according to the embodiment of the present invention can be variously modified without losing its characteristics. In the embodiment of the present invention, the region where the pixel value is increased as compared with the first image is defined as the increasing region, and the region where the pixel value is decreased is defined as the decreasing region. The "region in which the pixel value has increased" is a broad concept including a region in which the value calculated by a specific calculation method from the pixel value such as the average value of the pixel values has increased. The “region in which the pixel value is reduced” is a broad concept including a region in which the value calculated by a specific calculation method is reduced from the pixel value such as the average value of the pixel values.

If the region where the pixel value has increased is simply set as the increasing region and the region where the pixel value has decreased is set as the decreasing region, the pixel whose pixel value has increased or decreased due to noise is determined to be the increasing region or the decreasing region. Therefore, if the region where the pixel value increase amount is equal to or greater than the threshold value is defined as the increase region and the region where the pixel value decrease amount is equal to or greater than the threshold value is defined as the decrease region, erroneous detection due to noise can be prevented. For example, a region in which the pixel value is increased by 10 or more is defined as an increasing region, and a region in which the pixel value is decreased by 10 or more is defined as a decreasing region. Further, as described above, it is also effective as a noise countermeasure to divide the first image and the second image into blocks including a plurality of pixels and determine whether each block is an increase region or a decrease region.

FIG. 4 is a diagram showing a configuration in which the image processing device according to the embodiment of the present invention is realized by hardware and a configuration realized by software. The upper part of FIG. 4 shows the image processing device realized by hardware, and the lower part of FIG. 4 shows the image processing device realized by software. The first image and the second image are received by the receiving device 30 or the receiving device 40. Each function of the increase region disturbance degree calculation unit 12, the decrease region disturbance degree calculation unit 14, the determination unit 20, and the control unit 22 is realized by a processing circuit. Even if the processing circuit is a processing circuit 32 composed of dedicated hardware, a CPU 42 (Central Processing Unit, central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer) that executes a program stored in the memory 44 , Processor, DSP). The program that realizes each function of the increasing area disturbance degree calculation unit 12, the decreasing area disturbance degree calculation unit 14, the determination unit 20, and the control unit 22 may be one program or may be composed of a plurality of programs. Such programs are delivered, for example, over a network.

When the processing circuit is dedicated hardware, the processing circuit 32 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. The functions of each part of FIG. 1 may be realized by the processing circuit 32, or the functions of each part may be collectively realized by the processing circuit.

When the processing circuit is a CPU, the functions of each part are realized by software, firmware, or a combination of software and firmware. The software and firmware are described as programs and stored in the memory 44. The processing circuit realizes the functions of each part by reading and executing the program stored in the memory. It can be said that a program causes a computer to execute the procedure or method of each part. Here, the memory corresponds to, for example, non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, EEPROM, magnetic disk, flexible disk, optical disk, compact disk, mini disk, DVD and the like.

It should be noted that some of the functions of each part may be realized by dedicated hardware and some may be realized by software or firmware. In this way, the processing circuit can realize each of the above-mentioned functions by hardware, software, or a combination thereof.

12 Increase area disturbance calculation unit, 14 Decrease area disturbance calculation unit, 20 Judgment unit, 22 Control unit

Claims (10)

The second image taken after the first image is compared with the first image, and for each increase region in which the pixel value is increased, the increase region disturbance degree, which is the probability that the increase region is caused by the disturbance, is determined. An increasing area disturbance calculation unit that calculates and calculates the sum of the obtained increasing area disturbances,
The second image is compared with the first image, and the reduction region disturbance degree, which is the probability that the reduction region is caused by the disturbance, is calculated for each reduction region, which is the region where the pixel value is reduced, and the obtained reduction is obtained. A reduced area disturbance calculation unit that calculates the sum of the area disturbances,
When the sum of the total disturbance degree of the increasing region and the total sum of the disturbance degree of the decreasing region exceeds a predetermined value, it is determined that the combination of the first image and the second image is not suitable for image processing. Equipped with a judgment unit to
The total sum of the increased area disturbances is smaller than the increased area disturbance calculated for the total increased area by obtaining the total increased area which is the total area of the increased areas.
An image processing apparatus characterized in that the total sum of the reduced region disturbance degrees is smaller than the reduced region disturbance degree calculated for the total reduced area by obtaining the total reduced area which is the total area of the reduced regions. The increase region disturbance degree calculation unit calculates the increase region disturbance degree for each increase region by using a table that gives the increase region disturbance degree that increases as the area of the increase region increases.
The claim is characterized in that the reduction region disturbance degree calculation unit calculates the reduction region disturbance degree for each reduction region by using a table that gives the reduction region disturbance degree that increases as the area of the reduction region increases. The image processing apparatus according to 1. The increase region disturbance degree calculation unit uses a value obtained by raising the area of the increase region as the increase region disturbance degree.
The image processing apparatus according to claim 1, wherein the reduced region disturbance degree calculation unit uses a value obtained by raising the area of the reduced region as the reduced region disturbance degree. The increasing region disturbance degree calculation unit increases the increasing region disturbance degree as the outer circumference of the increasing region becomes smaller.
The image processing apparatus according to any one of claims 1 to 3, wherein the reduced region disturbance degree calculation unit increases the reduced region disturbance degree as the outer circumference of the reduced region becomes smaller. The increase region disturbance degree calculation unit and the decrease region disturbance degree calculation unit divide the first image and the second image into an aggregate of blocks having a plurality of predetermined pixels as constituent elements, and for each of the blocks. The image processing apparatus according to any one of claims 1 to 4, wherein it is determined whether or not the region is an increase region or a decrease region. The increase region disturbance degree calculation unit uses the block in which the average value of the pixel values has increased as the increase region.
The image processing apparatus according to claim 5, wherein the reduced region disturbance degree calculation unit uses the block in which the average value of pixel values is reduced as a reduced region. The increase region disturbance degree calculation unit sets the block in which the number of pixels whose pixel value has increased exceeds the threshold value as the increase region.
The image processing apparatus according to claim 5, wherein the reduced region disturbance degree calculation unit uses the block in which the number of pixels whose pixel values have been reduced exceeds a threshold value as the reduced region. The increase region disturbance degree calculation unit sets the block in which the value obtained by subtracting the number of pixels whose pixel value has decreased from the number of pixels whose pixel value has increased is larger than a predetermined value as the increase region.
The reduction region disturbance calculation unit is characterized in that the block in which the value obtained by subtracting the number of pixels whose pixel value has increased from the number of pixels whose pixel value has decreased is larger than a predetermined value is defined as the reduction region. The image processing apparatus according to claim 5. As the number of the increased regions increases, the sum of the increased region disturbances becomes smaller than the increased region disturbance calculated for the total increased area by obtaining the total increased area which is the total area of the increased regions.
The total of the reduced region disturbances is characterized in that as the number of the reduced regions increases, the total reduced area, which is the total area of the reduced regions, is calculated and becomes smaller than the reduced region disturbance calculated for the total reduced area. The image processing apparatus according to any one of claims 1 to 8. The second image taken after the first image is compared with the first image, and for each increased area where the pixel value is increased, the increased area disturbance degree which is the probability that the increased area is caused by the disturbance is calculated. The sum of the increased region disturbances calculated and obtained is smaller than the increased region disturbance calculated for the total increased area by obtaining the total increased area in which the sum of the increased region disturbances is the total area of the increased regions. And the process of calculating so that
The second image is compared with the first image, and the reduction area disturbance degree, which is the probability that the reduction area is caused by the disturbance, is calculated for each reduction area, which is the area where the pixel value is reduced, and the obtained reduction is obtained. The total of the region disturbances is calculated so that the sum of the reduced region disturbances is smaller than the reduced region disturbance calculated for the total reduced area by obtaining the total reduced area which is the total area of the reduced regions. ,
When the sum of the sum of the disturbance degrees of the increasing region and the sum of the disturbance degrees of the decreasing region exceeds a predetermined value, it is determined that the combination of the first image and the second image is not suitable for image processing. A program that lets a computer execute the processing to be performed.

Images