JP5780127B2 - 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 capable of improving the accuracy of image similarity determination.

  Conventionally, there is a technique for automatically determining the similarity of data of two images. For example, a color histogram is used as a determination element in determining the similarity of images. However, when similarity determination is performed using such determination elements, all image data of similar colors are determined to be similar. For this reason, it is difficult to determine dissimilarity between two image data having different angles of view, movement of the entire image, and white balance, resulting in a decrease in determination accuracy.

Therefore, for the data of two images having different angles of view, movement of the whole image, and white balance, the degree of similarity is determined by the number of corresponding points between the two images in order to realize the determination of dissimilarity. The technique of determining is also conventionally used.
However, when such a technique is used, if there is an abnormal value at the corresponding point between the two images and there is an error in the corresponding point itself, another problem arises that the determination accuracy is reduced. Yes.

As a technique capable of solving this problem, Patent Document 1 discloses a technique for detecting corresponding points in a region of a subject that is considered to be the same between two images.
Further, in the technique described in Patent Document 1, when an inclination of a corresponding point occurs in a region of a subject that is considered to be the same between two images, the corresponding point is determined based on the difference in the rotation angle of the corresponding point. It is determined whether they are normal values (hereinafter also referred to as inliers) or abnormal values (hereinafter also referred to as outliers), and the determination result is used for determining the similarity.

JP 2007-334895 A

  However, in the technique described in Patent Document 1, for example, when only a part of the subject rotates, some subjects whose positions are different due to the rotation or the like are not counted as corresponding points. For this reason, there is a case where it is determined as a dissimilar image where it should originally be determined as a similar image in which the same subject exists.

  The present invention has been made in view of such circumstances, and provides an image processing apparatus, an image processing method, and a program capable of improving the accuracy of similarity determination of images including a common subject. Objective.

In order to achieve the above object, an image processing apparatus according to an aspect of the present invention includes an acquisition unit that acquires data of a plurality of images, and a plurality of similar pixel regions with respect to the data of the plurality of images acquired by the acquisition unit. Based on the distances calculated by the distance calculation means, the distance calculation means for calculating the distances in the coordinate space between the plurality of similar pixel areas specified by the specification means, respectively. , Based on pixel areas corresponding to distances equal to or less than the threshold set by the threshold setting means among the distances calculated by the threshold setting means for setting a threshold used for similarity determination and the distance calculating means, Similarity determination means for determining whether the data of the plurality of images acquired by the acquisition means are similar to each other.

  According to the present invention, it is possible to improve the accuracy of similarity determination of images including a common subject.

It is a schematic diagram which shows the example of the image of 2 sheets used as determination object for demonstrating determination of the similarity of an image. 1 is a block diagram illustrating a hardware configuration of an image processing apparatus according to an embodiment of the present invention. It is a functional block diagram which shows the functional structure for performing an image similarity determination process among the functional structures of the image processing apparatus of FIG. It is a functional block diagram which shows the functional structure for performing the feature-value extraction process among the functional structures of the image processing apparatus of FIG. 4 is a flowchart illustrating a flow of image similarity determination processing executed by the image processing apparatus of FIG. 2 having the functional configuration of FIG. 3. It is a flowchart explaining the flow of the feature-value extraction process of step S2 among image similarity determination processes.

First, before describing the embodiment of the present invention, an outline of the principle of the present invention will be described with reference to FIG.
FIG. 1 is a schematic diagram illustrating an example of two images to be determined for explaining determination of similarity between images.
In FIG. 1, the first image p1 on the left is an image that is a similar determination source, and the second image p2 on the right is an image that is a similar determination target.

  In the example of FIG. 1, the apparatus determines whether the data of the first image p1 and the second image p2 are similar.

In the first image p1, an object ob1-1 placed on the left side of the image, an object ob2-1 placed on the center of the image, and an object ob3-1 placed on the right side of the image are subjects. include.
On the other hand, the second image p2 includes an object ob1-2 arranged on the left side of the image, an object ob2-2 arranged on the center of the image, and an object ob3-2 arranged on the right side of the image. It is included as a subject.

  The object ob1-1 in the first image p1 and the object ob1-2 in the second image p2 are images of the same writing instrument. Further, each of the feature points c1-1 to c4-1 in the object ob1-1 of the first image p1 corresponds to each of the feature points c1-2 to c4-2 in ob1-2 of the second image p2. The positions of these corresponding points are substantially the same. Thus, the object ob1-1 in the first image p1 and the object ob1-2 in the second image p2 are arranged at substantially the same position in the image.

  It can be seen visually that the object ob3-1 in the first image p1 and the object ob3-2 in the second image p2 are images of the same mobile phone. Also, each of the feature points c7-1 and c8-1 in the object ob3-1 of the first image p1 corresponds to each of the feature points c7-2 and c8-2 in the ob3-2 of the second image p2. It can be seen visually that the positions of these corresponding points are substantially the same. Thus, it can be seen visually that the object ob3-1 in the first image p1 and the object ob3-2 in the second image p2 are arranged at substantially the same position in the image.

  On the other hand, the object ob2-1 in the first image p1 and the object ob23-2 in the second image p2 are the same wristwatch image, but are arranged at different positions in the image. This can be seen visually. That is, each of the characteristic portions c5-1 and c6-1 in the object ob2-1 of the first image p1 and each of the characteristic portions c5-2 and c6-2 in the object ob2-2 of the second image p2. Although it is a corresponding point, it can be seen visually that the positions of these corresponding points are different. Specifically, visually, it can be seen that the object ob1-1 in the first image and the object ob1-2 in the second image are arranged at positions rotated by approximately 180 degrees in the image. .

  As described above, in the example of FIG. 1, the first image p <b> 1 and the second image p <b> 2 are the positional relationship between the wristwatch objects ob <b> 2-1 and ob <b> 2-2 among a plurality of identical subjects such as a writing instrument, a mobile phone, and a wristwatch. Although they have different compositions, it can be seen that they are similar.

However, when the similarity between the data of the first image p1 and the second image p2 is determined using the conventional technique described above in the [Background art] column, the position is rotated by 180 degrees. The different objects ob2-1 and ob2-2 are determined to be dissimilar, and as a result, the first image p1 and the second image p2 as a whole may be determined to be dissimilar.
That is, in the conventional technique, since the feature points that should originally be the corresponding points between the two images are determined not to be the corresponding points due to a difference in rotation angle or the like, as a result, the two images May be determined to be dissimilar as a whole.

Therefore, according to an embodiment of the present invention, even when a part of an object in the image is moved, for example, even when a part of the object in the image is rotated, it can be determined as the same object. This is a technique that can realize the following series of processes.
That is, in the present embodiment, a similar pixel area is specified from each of a plurality of images. Here, the pixel region refers to a region composed of one or more pixels. That is, one pixel, for example, a feature point (pixel) is an example of the pixel region. The apparatus calculates the distance between the identified pixel areas. The apparatus determines whether or not each calculated distance is equal to or less than a threshold, and determines whether or not the plurality of images as a whole are similar based on pixel areas corresponding to the distance determined to be equal to or less than the threshold. Here, the threshold value is set based on each of the distances between the specified pixel regions. Such a series of processing is hereinafter referred to as “image similarity determination processing”. That is, a method capable of realizing the image similarity determination process is an embodiment of the present invention.
By applying such a method of the present invention to an image processing apparatus, the similarity of images can be determined with high accuracy. Hereinafter, an embodiment of an image processing apparatus to which the method of the present invention is applied will be described with reference to FIG. 2 and subsequent drawings.

FIG. 2 is a block diagram showing a hardware configuration of the image processing apparatus according to the embodiment of the present invention.
The image processing apparatus 1 is configured as a personal computer, for example.

  The image processing apparatus 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an image processing unit 14, a bus 15, an input / output interface 16, and an input. A unit 17, an output unit 18, a storage unit 19, a communication unit 20, and a drive 21 are provided.

  The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 19 to the RAM 13.

  The RAM 13 appropriately stores data necessary for the CPU 11 to execute various processes.

  The image processing unit 14 is configured by a DSP (Digital Signal Processor), a VRAM (Video Random Access Memory), and the like, and performs various image processing on image data in cooperation with the CPU 11.

  The CPU 11, ROM 12, RAM 13, and image processing unit 14 are connected to each other via a bus 15. An input / output interface 16 is also connected to the bus 15. An input unit 17, an output unit 18, a storage unit 19, a communication unit 20, and a drive 21 are connected to the input / output interface 16.

The input unit 17 includes a mouse, a keyboard, and the like, and inputs various types of information in accordance with user instruction operations.
The output unit 18 includes a display, a speaker, and the like, and outputs images and sounds.
The storage unit 19 is composed of a hard disk, a DRAM (Dynamic Random Access Memory), or the like, and stores various image data.
The communication unit 20 controls communication performed with other devices (not shown) via a network including the Internet.

  A removable medium 31 made of a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is appropriately attached to the drive 21. The program read from the removable medium 31 by the drive 21 is installed in the storage unit 19 as necessary. The removable medium 31 can also store various data such as image data stored in the storage unit 19 in the same manner as the storage unit 19.

Next, a functional configuration for executing the above-described image similarity determination process among the functional configurations of the image processing apparatus 1 will be described with reference to FIG.
FIG. 3 is a functional block diagram showing a functional configuration for executing the image similarity determination process among the functional configurations of the image processing apparatus 1 of FIG.

When the image processing apparatus 1 executes the image similarity determination process, as shown in FIG. 3, in the image processing unit 14, a pair image acquisition unit 41, a feature amount extraction unit 42, a similarity calculation unit 43, The similar image determination unit 44 functions. In the CPU 11, the output control unit 51 functions.
However, FIG. 3 is an example, and a part of the function of the image processing unit 14 may be delegated to the CPU 11, and conversely, a part of the function of the CPU 11 may be delegated to the image processing unit 14.
Further, an image data storage unit 61 is provided as an area of the storage unit 19 of the image processing apparatus 1. Note that the image data storage unit 61 is provided as one area of the storage unit 19, and may be provided as another area of the removable media 31, for example.

  The image data storage unit 61 stores, for example, data of a plurality of images used for the similar image determination process.

  The pair image acquisition unit 41 acquires data of two images (hereinafter referred to as “pair images”) as similarity determination targets from the image data storage unit 61 and supplies the acquired data to the feature amount extraction unit 42.

The feature quantity extraction unit 42 extracts a plurality of feature points included in each pair image from the pair image data acquired by the pair image acquisition unit 41 and searches for corresponding points between the pair images. Next, the feature quantity extraction unit 42 classifies the corresponding point pairs that are to be counted and the corresponding point pairs that are not to be counted, based on the distance between the corresponding point pairs, and sets the corresponding point pairs that are to be counted. The number is output as a feature value.
Further details of the feature quantity extraction unit 42 will be described later with reference to FIG.

The similarity calculation unit 43 calculates the pair image similarity based on the feature amount output from the feature amount extraction unit 42.
Specifically, in this embodiment, the similarity calculation unit 43 calculates the pair image similarity by weighting the feature amount with a weighting coefficient. Here, the weighting coefficient is not particularly limited, but in the present embodiment, a value calculated in advance using a machine learning algorithm such as adaboost (Adaptive Boosting) is employed.

  The similar image determination unit 44 determines whether or not the pair images are similar based on the similarity calculated by the similarity calculation unit 43 (similarity determination), and the result (hereinafter referred to as “similarity determination result”). Stored in the storage unit 19.

  The output control unit 51 executes control for outputting the pair image similarity determination result and the pair image data to the output unit 18.

Further, details of the functional configuration of the feature amount extraction unit 42 among the functional configurations of the image processing apparatus 1 will be described below.
FIG. 4 is a functional block diagram showing details of the functional configuration of the feature quantity extraction unit 42 in the image processing apparatus 1 of FIG.

  When the image processing apparatus 1 executes the image similarity determination process, the feature amount extraction unit 42 of the image processing unit 14 includes a feature point extraction unit 81, a corresponding point search unit 82, a vector, as shown in FIG. The calculation unit 83, the threshold setting unit 84, the count unit 85, the corresponding point determination unit 86, and the feature amount output unit 87 function.

  An image information data storage unit 62 is provided as an area of the storage unit 19 of the image processing apparatus 1. Note that the image information data storage unit 62 is provided as one area of the storage unit 19, and may be provided as another area of the removable media 31, for example.

  The image information data storage unit 62 stores, for example, various image information data such as vector values, feature point counts, and feature amounts, which will be described later, used for feature amount extraction processing.

The feature point extraction unit 81 extracts feature points included in the pair image for each data of the pair image acquired by the pair image acquisition unit 41.
The method of extracting feature points by the feature point extracting unit 81 is not particularly limited, and any known method that will appear or will appear in the future can be adopted. However, in the present embodiment, a feature point extraction method using SURF (Speeded-Up Robust Features) processing is employed.

  The corresponding point search unit 82 searches for corresponding points between the pair images (hereinafter, referred to as “corresponding point pair”) from among the plurality of feature points extracted by the feature point extraction unit 81.

Of the corresponding point pairs searched by the corresponding point search unit 82, the vector calculation unit 83 uses the coordinates of one corresponding point in the coordinate space common between the pair images as the start point and the coordinates of the other corresponding point as the end point. A value based on the length of the vector, that is, a value representing a distance between corresponding point pairs (hereinafter simply referred to as “vector value”) is calculated.
The method for calculating the vector value is not particularly limited, and any known method that will appear in the future or that will appear in the future can be adopted. However, in this embodiment, a method using the Euclidean distance calculated by the sum of squares is adopted.
Further, the vector calculation unit 83 causes the image information data storage unit 62 to store the calculated vector value.

Based on the vector values for all corresponding point pairs calculated by the vector calculating unit 83, the threshold setting unit 84 is an index (determination criterion) of an abnormal distance (meaning that it cannot be said that the distance between corresponding points). Is obtained as a threshold value.
In the present embodiment, the threshold value setting unit 84 obtains a median value (median value) among the values of each vector of all corresponding point pairs, and a value obtained by adding or multiplying the median value by a certain number α is used as the threshold value. Set.

  The counting unit 85 counts the total number of corresponding point pairs to be counted among all the corresponding point pairs calculated by the vector calculating unit 83. The count unit 85 stores the count value in the image information data storage unit 62.

The corresponding point determination unit 86 determines whether each of all the corresponding point pairs calculated by the vector calculation unit 83 is a corresponding point pair to be counted or a corresponding point pair to be excluded from counting. To do. The corresponding point pair to be counted is hereinafter referred to as “inlier”. On the other hand, corresponding point pairs that are not counted are hereinafter referred to as “outliers”.
Specifically, the corresponding point determination unit 86 determines that the vector value (the value indicating the distance between the corresponding point pairs) of the corresponding point pair to be determined is equal to or less than the threshold value, and determines that the threshold value is inlier. If it exceeds, it is determined as an outlier.

  The feature amount output unit 87 counts the count unit 85 when the determination by the corresponding point determination unit 86 is completed for all the corresponding point pairs calculated by the vector calculation unit 83 and the counting by the counting unit 85 ends. The value is output as a feature value.

Next, the flow of image similarity determination processing executed by the image processing apparatus 1 according to the present embodiment will be described with reference to FIG.
FIG. 5 is a flowchart for explaining the flow of the image similarity determination process executed by the image processing apparatus 1 of FIG. 2 having the functional configuration of FIG.

A user who operates the input unit 17 of the image processing apparatus 1 can perform a predetermined operation instructing the start of the image similarity determination process.
The image similarity determination process is triggered by such a predetermined operation by the user, and the following process is executed.

  In step S <b> 1, the pair image acquisition unit 41 acquires pair image data from the image data storage unit 61.

In step S <b> 2, the feature amount extraction unit 42 outputs the feature amount of the pair image data acquired in the process of step S <b> 1.
Such processing of the feature amount extraction unit 42 in step S2 is hereinafter referred to as “feature amount extraction processing”. Details of the feature amount extraction processing will be described later with reference to FIG.

  In step S3, the similarity calculation unit 43 weights the feature amount output by the feature amount extraction process of step S2 with the weighting coefficient calculated by the abodost, thereby obtaining the pair image data acquired in the process of step S1. The similarity for is calculated.

In step S4, the similar image determination unit 44 determines whether the similarity calculated in the process of step S3 is greater than a predetermined threshold.
If the similarity exceeds the threshold, it is determined as YES in step S4, and the process proceeds to step S5.

  In step S5, the similar image determination unit 44 determines that the pair images are similar images. When this process ends, the process proceeds to step S7. However, the process of step S7 will be described later.

  On the other hand, when the similarity calculated by the process of step S3 is below a threshold value, it determines with it being NO in step S4, and a process progresses to step S6.

  In step S6, the similar image determination unit 44 determines that the pair images are not similar images (non-similar images). When this process ends, the process proceeds to step S7.

In step S <b> 7, the output unit 18 causes the output control unit 51 to output the processing result of step S <b> 5 or step S <b> 6, that is, the similarity determination result, from the output unit 18 together with the pair image data. Although the output form of the determination result is not particularly limited, in the present embodiment, a message image indicating the determination result is displayed on the display unit of the output unit 18 together with the pair image.
Thereby, the image similarity determination process ends.

Next, a detailed flow of the feature amount extraction process in step S2 in such an image similarity determination process will be described with reference to FIG.
FIG. 6 is a flowchart showing a detailed flow of the feature amount extraction process of step S2 in the image similarity determination process of FIG.

  In step S31, the feature point extraction unit 81 extracts feature points included in the pair image from the pair image data.

  In step S32, the corresponding point search unit 82 searches for corresponding points between the pair images based on the feature points extracted in the process of step S31.

  In step S33, the vector calculation unit 83 sequentially sets each of all the corresponding point pairs searched in the process of step S32 as a processing target, and determines the target of the processing target from the distance between images of the corresponding pair of processing targets. A vector value of the corresponding point pair (a value indicating the Euclidean distance between the corresponding point pairs) is calculated.

  In step S34, the threshold setting unit 84 sets a threshold based on the vector values for all the corresponding point pairs calculated by the processing in step S33.

  In step S35, the count unit 85 sets the inlier count value i to “0” (i = 0). The inlier count value i is a value counted by the counting unit 85.

  In step S36, the corresponding point determination unit 86 sets a predetermined one of all corresponding point pairs searched in the process of step S32 as a corresponding point pair to be processed.

In step S37, the corresponding point determination unit 86 determines whether the vector value of the corresponding point pair to be processed is equal to or less than a threshold value.
If the vector value is less than or equal to the threshold value, it is determined as YES in step S37, and the process proceeds to step S38.

  In step S38, the corresponding point determination unit 86 determines that the corresponding point pair to be processed is an inlier.

  In step S39, the count unit 85 increments the inlier count value i by “1” (i = i + 1). When this process ends, the process proceeds to step S42. However, the processing after step S42 will be described later.

On the other hand, if the vector value of the corresponding point pair to be processed exceeds the threshold value, it is determined as NO in step S37, and the process proceeds to step S40.

  In step S40, the corresponding point determination unit 86 determines that the corresponding point pair to be processed is an outlier.

  In step S41, the corresponding point determination unit 86 excludes outliers. That is, the corresponding pair of processing targets determined to be outliers is excluded from the count target. When this process ends, the process proceeds to step S42.

In step S42, the corresponding point determination unit 86 determines whether all corresponding point pairs have been set as processing targets.
If all corresponding point pairs have not yet been set as processing targets, NO is determined in step S42, and the process returns to step S36. Thereafter, until all the corresponding point pairs are set as processing targets, the loop processing from step S36 to step S42 is repeated. When the processing target point pair is an inlier, the inlier count value i is “1”. ”Is incremented (i = i + 1), and if it is an outlier, it is excluded from the count target.
If all the corresponding point pairs are set as processing targets, YES is determined in step S42, and the process proceeds to step S43.

  In step S43, the feature amount output unit 87 outputs the inlier count value i as a feature point. Thereafter, the feature point extraction process ends, and the process returns to step S3.

As described above, the image processing apparatus 1 includes the pair image acquisition unit 41, the corresponding point search unit 82, the vector calculation unit 83, the threshold setting unit 84, and the similar image determination unit 44.
The pair image acquisition unit 41 acquires a pair image (a pair image that is an example of a plurality of images) that is a plurality of images to be determined.
The corresponding point search unit 82 specifies a plurality of similar pixel regions for the data of the plurality of images acquired by the pair image acquisition unit 41.
The vector calculation unit 83 calculates the distance between the image areas specified by the corresponding point search unit 82.
The threshold setting unit 84 sets a threshold based on the distance calculated by the vector calculation unit 83.
The similar image determination unit 44 is acquired by the pair image acquisition unit 41 based on pixel areas corresponding to distances equal to or less than the threshold set by the threshold setting unit 84 among the distances calculated by the vector calculation unit 83. It is determined whether the data of the plurality of images are similar.

  Accordingly, the image processing apparatus 1 acquires the pair image acquisition unit 41 based on pixel areas corresponding to distances equal to or less than the threshold set by the threshold setting unit 84 among the distances calculated by the vector calculation unit 83. It is determined whether the data of the plurality of images obtained are similar. For this reason, in the image processing apparatus 1, even when a part of the object in the image is moving (for example, when a part of the object in the image is rotating), it is determined that they are similar. Can do. Therefore, the image processing apparatus 1 can improve the accuracy of similarity determination of images including a common subject.

The image processing apparatus 1 further includes a corresponding point determination unit 86 that determines whether each distance calculated by the vector calculation unit 83 is equal to or less than the threshold set by the threshold setting unit 84.
The similar image determination unit 44 determines whether or not the data of the plurality of images acquired by the pair image acquisition unit 41 is similar based on pixel areas corresponding to the distance determined by the corresponding point determination unit 86 to be equal to or less than the threshold. Determine. For this reason, the image processing apparatus 1 can perform similar determination with higher accuracy.

  The similar image determination unit 44 excludes pixel regions corresponding to the distances determined to exceed the threshold by the threshold setting unit 84 from the distances calculated by the vector calculation unit 83 from the determination targets. It is determined whether or not the data of a plurality of images are similar. For this reason, the image processing apparatus 1 determines that the objects are similar even when some of the objects in the image are moving (for example, when some of the objects in the image are rotating). Can do.

  The threshold setting unit 84 sets a threshold based on any one of the median value, average value, and mode value of the distances calculated by the vector calculation unit 83. Thereby, the image processing apparatus 1 can determine an abnormal value.

  In addition, this invention is not limited to the above-mentioned embodiment, The deformation | transformation in the range which can achieve the objective of this invention, improvement, etc. are included in this invention.

  In the above-described embodiment, feature points are extracted by SURF processing in consideration of processing speed and processing load. However, when improvement of extraction accuracy is prioritized, feature points are extracted by SIFT processing. You may comprise as follows.

  In the above-described embodiment, the spatial distance between pixels is calculated using the Euclidean distance, but the present invention is not limited to this. For example, the spatial distance between the pixels may be calculated by a Manhattan distance calculation method that is calculated by the sum of absolute values. Good.

  In the above-described embodiment, the threshold value is obtained by adding a certain number of α to the median value of the distances of all corresponding point pairs. However, the present invention is not limited to this. For example, instead of the median value, You may comprise so that a frequent value may be used.

In the above-described embodiment, the image processing apparatus 1 to which the present invention is applied has been described using a digital camera as an example, but is not particularly limited thereto.
For example, the present invention can be applied to general electronic devices having an image similarity determination processing function. Specifically, for example, the present invention can be applied to a notebook personal computer, a printer, a television receiver, a video camera, a portable navigation device, a mobile phone, a portable game machine, and the like.

The series of processes described above can be executed by hardware or can be executed by software.
In other words, the functional configurations of FIGS. 3 and 4 are merely examples, and are not particularly limited. That is, it is sufficient if the image processing apparatus 1 has a function capable of executing the above-described series of processes as a whole, and what functional blocks are used to realize this function is particularly the example shown in FIGS. It is not limited to.
In addition, one functional block may be constituted by hardware alone, software alone, or a combination thereof.

When a series of processing is executed by software, a program constituting the software is installed on a computer or the like from a network or a recording medium.
The computer may be a computer incorporated in dedicated hardware. The computer may be a computer capable of executing various functions by installing various programs, for example, a general-purpose personal computer.

  The recording medium including such a program is not only constituted by the removable medium 31 of FIG. 2 distributed separately from the apparatus main body in order to provide the program to the user, but also in a state of being incorporated in the apparatus main body in advance. It is comprised with the recording medium etc. which are provided in. The removable medium 31 is composed of, for example, a magnetic disk (including a floppy disk), an optical disk, a magneto-optical disk, or the like. The optical disk is composed of, for example, a CD-ROM (Compact Disk-Read Only Memory), a DVD (Digital Versatile Disk), or the like. The magneto-optical disk is configured by an MD (Mini-Disk) or the like. In addition, the recording medium provided to the user in a state of being preliminarily incorporated in the apparatus main body includes, for example, the ROM 12 in FIG. 2 in which the program is recorded, the hard disk included in the storage unit 19 in FIG.

In the present specification, the step of describing the program recorded on the recording medium is not limited to the processing performed in time series along the order, but is not necessarily performed in time series, either in parallel or individually. The process to be executed is also included.
Further, in the present specification, the term “system” means an overall device configured by a plurality of devices, a plurality of means, and the like.

  As mentioned above, although several embodiment of this invention was described, these embodiment is only an illustration and does not limit the technical scope of this invention. The present invention can take other various embodiments, and various modifications such as omission and replacement can be made without departing from the gist of the present invention. These embodiments and modifications thereof are included in the scope and gist of the invention described in this specification and the like, and are included in the invention described in the claims and the equivalent scope thereof.

The invention described in the scope of claims at the beginning of the filing of the present application will be appended.
[Appendix 1]
An acquisition means for acquiring data of a plurality of images;
A plurality of image data acquired by the acquisition unit, a specifying unit for specifying a plurality of similar pixel regions,
Distance calculating means for calculating the distance between the pixel areas specified by the specifying means;
Threshold setting means for setting a threshold based on the distances calculated by the distance calculating means;
Of the distances calculated by the distance calculating means, the data of the plurality of images acquired by the acquiring means are similar based on pixel areas corresponding to distances equal to or less than the threshold set by the threshold setting means. Similarity determination means for determining whether or not to do,
An image processing apparatus comprising:
[Appendix 2]
A threshold determination unit that determines whether each distance calculated by the distance calculation unit is equal to or less than a threshold set by the threshold setting unit;
The similarity determination unit determines whether the data of the plurality of images acquired by the acquisition unit are similar based on pixel areas corresponding to the distance determined to be equal to or less than the threshold by the threshold determination unit. ,
The image processing apparatus according to appendix 1, wherein
[Appendix 3]
The similarity determination unit excludes pixel areas corresponding to the distances determined to exceed the threshold by the threshold determination unit from the distances calculated by the distance calculation unit, Determine whether the data of multiple images are similar,
The image processing apparatus according to Supplementary Note 2, wherein
[Appendix 4]
The threshold setting means sets the threshold based on any one of a median value, an average value, and a mode value of the distances calculated by the distance calculation means;
The image processing apparatus according to any one of supplementary notes 1 to 3, wherein
[Appendix 5]
An image processing method executed by an image processing apparatus,
An acquisition step of acquiring a plurality of images;
A specifying step of specifying a plurality of similar pixel regions from a plurality of images acquired by the acquiring step;
A distance calculating step of calculating a distance between the image regions specified by the specifying step;
A threshold setting step for setting a threshold based on the distance calculated by each of the distance calculating steps;
A similarity determination step for determining whether or not pixel regions corresponding to a distance equal to or smaller than the threshold set by the threshold setting step among the distances calculated by the distance calculation step are similar to each other;
An image processing method comprising:
[Appendix 6]
Computer
Acquisition means for acquiring a plurality of images;
A specifying unit that specifies a plurality of similar pixel regions from among a plurality of images acquired by the acquiring unit;
Distance calculating means for calculating the distance between the image areas specified by the specifying means;
Threshold setting means for setting a threshold based on the distances calculated by the distance calculating means;
Similarity determination means for determining whether or not pixel regions corresponding to distances equal to or less than the threshold set by the threshold setting means among the distances calculated by the distance calculation means are similar to each other,
A program characterized by having the means as follows.

  DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 11 ... CPU, 12 ... ROM, 11 ... RAM, 14 ... Image processing part, 15 ... Bus, 16 ... Input / output interface, 17. ··· input unit, 18 ··· output unit, 19 ··· storage unit, 20 ··· communication unit, 21 ··· drive, 31 ··· removable media, 41 ··· pair image acquisition unit, 42 · · ..Feature amount extraction unit, 43... Similarity calculation unit, 44... Similar image determination unit, 51... Output control unit, 61. , 81... Feature point extraction unit, 82... Corresponding point search unit, 83... Vector calculation unit, 84. Part, 87 ... similarity output part

Claims (6)

An acquisition means for acquiring data of a plurality of images;
A plurality of image data acquired by the acquisition unit, a specifying unit for specifying a plurality of similar pixel regions,
Distance calculating means for calculating distances in a coordinate space between a plurality of similar pixel areas specified by the specifying means;
Threshold setting means for setting a threshold used for similarity determination based on the distances calculated by the distance calculating means;
Of the distances calculated by the distance calculating means, the data of the plurality of images acquired by the acquiring means are similar based on pixel areas corresponding to distances equal to or less than the threshold set by the threshold setting means. Similarity determination means for determining whether or not to do,
An image processing apparatus comprising: Distance coordinate space which are respectively calculated by the distance calculation means further includes a threshold determination means for determining whether the following set threshold by the threshold value setting means,
The similarity determining means determines based on the pixel regions with each other corresponding to the distance it is determined that the threshold value or less, whether the data of the plurality of images acquired by the acquisition unit is more similar to the threshold determination unit To
The image processing apparatus according to claim 1. The similarity determination unit excludes pixel areas corresponding to the distance determined to exceed the threshold by the threshold determination unit from the distances in the coordinate space respectively calculated by the distance calculation unit from the determination target. And determining whether or not the data of the plurality of images is similar.
The image processing apparatus according to claim 2. The threshold value setting means sets the threshold value based on any one of a median value, an average value, and a mode value of distances in the coordinate space calculated by the distance calculation means,
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. An image processing method executed by an image processing apparatus,
An acquisition step of acquiring a plurality of images;
A specifying step of specifying a plurality of similar pixel regions from a plurality of images acquired by the acquiring step;
A distance calculating step for calculating a distance on a coordinate space between a plurality of similar image areas specified by the specifying step;
A threshold value setting step for setting a threshold value used for similarity determination based on the distances calculated by the distance calculation step;
A similarity determination step for determining whether or not pixel regions corresponding to a distance equal to or smaller than the threshold set by the threshold setting step among the distances calculated by the distance calculation step are similar to each other;
An image processing method comprising: Computer
Acquisition means for acquiring a plurality of images;
A specifying unit that specifies a plurality of similar pixel regions from among a plurality of images acquired by the acquiring unit;
Distance calculating means for calculating distances in a coordinate space between a plurality of similar image areas specified by the specifying means;
Threshold setting means for setting a threshold used for similarity determination based on the distances calculated by the distance calculating means,
Similarity determination means for determining whether or not pixel regions corresponding to distances equal to or less than the threshold set by the threshold setting means among the distances calculated by the distance calculation means are similar to each other,
A program characterized by functioning as

Images