Image processing apparatus, image processing method, and image processing program

Abstract

An object of the present invention is to improve the accuracy of image processing for determining whether or not an image related to a predetermined object is included in a target image. An image processing apparatus 1 includes a sample distribution index calculation unit 25 that calculates a sample distribution index indicating the variation in a sample image S1 of a plurality of sample features extracted from a sample image S1 related to a predetermined object, a match distribution index calculation unit 26 that calculates a match distribution index indicating the variation in a sample image S1 of a plurality of match features that match a plurality of query features extracted from a query image Q1 or Q2 that is a determination target as to whether or not the sample image S1 is included among the plurality of sample features, and an exclusion determination unit 27 that determines that the sample image S1 is not included in the query image Q1 or Q2 when the difference between the sample distribution index and the match distribution index exceeds a predetermined reference.

Description

Technical Field

[0001] The present disclosure relates to image processing technology.

Background Art

[0002] Conventionally, an acquisition unit that acquires a first image and a second image, a feature point extraction unit that extracts feature points of each of the first image and the second image, and a feature point extracted from the first image and the second image A matching unit that associates the extracted feature points to detect corresponding points between the images, an outermost contour extraction unit that extracts the outermost contour from each of the first image and the second image, and the outermost contour and the number of corresponding points Based on the above, a detection unit that detects a similar region, which is a partial region similar to each other between the first image and the second image, from each of the first image and the second image has been proposed (see Patent Document 1).

[0003] In addition, various techniques for performing image matching by collating features extracted from images have been proposed (see Non-Patent Documents 1 and 2).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Non-Patent Documents

[0005]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] Conventionally, various techniques have been proposed for recognizing a predetermined object (such as an imaged object or characters) included in an image, but there is still room for improvement in the recognition accuracy of the object in the image. In view of the above problems, an object of the present disclosure is to improve the recognition accuracy of a predetermined object included in an image.

Means for Solving the Problems

[0007] An example of the present disclosure includes sample distribution index calculation means for calculating a sample distribution index indicating the variation in a sample image of a plurality of sample features extracted from a sample image related to a predetermined object, and among the plurality of sample features, a plurality of match features that match a plurality of query features extracted from a query image that is a determination target as to whether an image related to the predetermined object is included, match distribution index calculation means for calculating a match distribution index indicating the variation in the sample image, and exclusion determination means for determining that the query image does not include an image related to the predetermined object when the difference between the sample distribution index and the match distribution index exceeds a predetermined reference. The image processing apparatus includes the above.

[0008] The present disclosure can be understood as an image processing apparatus, a system, a method executed by a computer, or a program to be executed by a computer. Further, the present disclosure can also be understood as a program recorded on a recording medium readable by a computer or other devices, machines, etc. Here, the recording medium readable by a computer or the like refers to a recording medium that accumulates information such as data and programs by an electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer or the like.

Advantages of the Invention

[0009] According to the present disclosure, it is possible to improve the recognition accuracy of a predetermined object included in an image.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Modes for Carrying Out the Invention

[0011] Hereinafter, embodiments of an image processing apparatus, method, and program according to the present disclosure will be described with reference to the drawings. However, the embodiments described below are illustrative of the embodiments and do not limit the image processing apparatus, method, and program according to the present disclosure to the specific configurations described below. In practice, specific configurations according to the implementation mode may be appropriately adopted, and various improvements and modifications may be made.

[0012] Conventionally, feature matching has been performed for comparing images, and various image processing techniques for determining whether a target image contains an image related to a predetermined object have been proposed. However, although the conventionally proposed techniques have a certain effect in detecting similar regions, which are partially similar regions between images, from each image, there is room for improvement in performing matching that takes into account the positional relationship and distribution status between features in a huge number of feature groups.

[0013] In view of the above problems, an object of the present disclosure is to improve the accuracy of image processing for determining whether a target image contains an image related to a predetermined object.

[0014] According to the image processing apparatus, method, and program according to the present embodiment, it becomes possible to perform matching that takes into account the positional relationship and distribution status between features in a feature group, and it becomes possible to improve the accuracy of image processing for determining whether a target image contains an image related to a predetermined object.

[0015] In this embodiment, an embodiment will be described in the case where the technology according to the present disclosure is implemented to check whether an advertisement as a predetermined object is installed correctly in accordance with an instruction for advertisement posting. However, the technology according to the present disclosure can be widely used to determine whether an image related to a predetermined object is included in a target image, and the application target of the present disclosure is not limited to the examples shown in the embodiment. For example, the object may not be a physical object, but an object drawn in a virtual space or an object drawn using drawing software. Also, for example, the image may not be an optically captured image, but an image in which a virtual space is drawn or an image drawn using drawing software.

[0016] <Configuration of the System> FIG. 1 is a schematic diagram showing the configuration of the system according to this embodiment. The system according to this embodiment includes an image processing apparatus 1, an imaging apparatus 81, and a user terminal 9 that can communicate with each other by being connected to a network.

[0017] The image processing apparatus 1 is a computer including a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage device 14 such as an EEPROM (Electrically Erasable and Programmable Read Only Memory) or an HDD (Hard Disk Drive), a communication unit 15 such as a NIC (Network Interface Card), and the like. However, regarding the specific hardware configuration of the image processing apparatus 1, appropriate omission, replacement, or addition can be made according to the embodiment. Also, the image processing apparatus 1 is not limited to a device consisting of a single housing. The image processing apparatus 1 may be realized by a plurality of devices using so-called cloud or distributed computing technologies.

[0018] The imaging device 81 obtains a query image, which will be described later, by imaging the inspection target. As the imaging device 81, a general digital camera or other device capable of recording the light incident from the target may be used, and its specific configuration is not limited. In the present embodiment, the inspection target is a store in which an advertisement as a predetermined object is installed.

[0019] The user terminal 9 is a terminal device used by the user. The user terminal 9 is a computer including a CPU, a ROM, a RAM, a storage device, a communication unit, an input device, an output device, etc. (illustrations are omitted). However, regarding the specific hardware configuration of the user terminal 9, appropriate omission, replacement, or addition can be made according to the implementation mode. Also, the user terminal 9 is not limited to a device consisting of a single housing. The user terminal 9 may be realized by a plurality of devices using so-called cloud or distributed computing technologies, etc. The user uses various services provided by the system according to the present embodiment via these user terminals 9.

[0020] FIG. 2 is a diagram showing an outline of the functional configuration of the image processing apparatus 1 according to the present embodiment. In the image processing apparatus 1, the program recorded in the storage device 14 is read into the RAM 13 and executed by the CPU 11, and each hardware provided in the image processing apparatus 1 is controlled, so that the query image acquisition unit 21, the feature extraction unit 22, the matching feature identification unit 23, the object determination unit 24, the sample distribution index calculation unit 25, the matching distribution index calculation unit 26, the exclusion determination unit 27, and the notification unit 28 are provided. In the present embodiment and other embodiments described later, each function provided in the image processing apparatus 1 is executed by the CPU 11, which is a general-purpose processor, but a part or all of these functions may be executed by one or a plurality of dedicated processors.

[0021] The query image acquisition unit 21 acquires a query image to be inspected, in which an installed object (an advertisement in this embodiment) and an installation target of the object (a store in this embodiment) are simultaneously imaged. The method for acquiring the query image is not limited, but in this embodiment, an example of acquiring a query image imaged using the imaging device 81 via the user terminal 9 will be described.

[0022] FIG. 3 is a diagram showing an example of a query image in this embodiment. In this figure, a query image Q1 in which an advertisement A1 is imaged together with a store, and a query image Q2 in which an advertisement A2 that is an advertisement related to the same product or service as the advertisement A1 but has a different layout is imaged together with the store are shown.

[0023] As described above, in this embodiment, an embodiment in which the technology according to the present disclosure is implemented to check whether an advertisement as a predetermined object is installed correctly in accordance with an instruction for advertisement posting will be described. When a company posts an advertisement in a store or the like, even if the products or services targeted by the advertisement are the same, for a plurality of types of advertisements prepared for the advertisement, in order to obtain the intended effect, its posting rules (posting position, posting time, etc.) may be specified in detail. For example, for each size, layout, and content of the advertisement, it is instructed where it should be posted, such as on an outdoor signboard of the store, at the store entrance, on a product shelf inside the store, or on a sideboard.

[0024] A person in charge of checking whether an advertisement is installed correctly in accordance with an instruction for advertisement posting takes a photo so that the advertisement and its installation target are included in the same image at the location where the advertisement is posted (for example, a store), and uses this as a query image.

[0025] The feature extraction unit 22 extracts a plurality of features (feature points. Hereinafter, the features extracted from the query image are referred to as "query features") from the query image Q1 or Q2 that is the object to be determined whether it includes an image related to a predetermined object. Further, in the present embodiment, the feature extraction unit 22 further extracts a plurality of features (hereinafter, the features extracted from the sample image are referred to as "sample features") from the sample image S1 related to a predetermined object. Note that the types of algorithms and descriptors used for feature extraction from images are not limited, and any algorithms and descriptors currently known and those to be developed in the future may be used.

[0026] FIG. 4 is a diagram showing an example of the sample image S1 related to the object in the present embodiment and the features extracted from the sample image S1. In this figure, the sample image S1 of the advertisement A1 described above is shown. Since the object according to the present embodiment is an advertisement posted in a store, as the sample image, data for creating an advertisement (for example, printing data) may be used, or photo data obtained by imaging an actually created advertisement may be used.

[0027] The "+" in the figure indicates the position of the sample features extracted from the sample image S1. Regarding the feature extraction from the query image Q1 or Q2, since it is substantially the same as the feature extraction from the sample image except that the image input to the extractor is a query image instead of a sample image, the illustration is omitted.

[0028] The matching feature specifying unit 23 determines whether each of the plurality of sample features extracted from the sample image S1 matches any of the plurality of query features extracted from the query image Q1 or Q2, and thereby specifies a plurality of sample features (hereinafter, referred to as "matching features") that match any of the plurality of query features among the plurality of sample features.

[0029] The object determination unit 24 determines whether the query image Q1 or Q2 contains an image related to a predetermined object, based on a plurality of matching features, by a method different from that of the exclusion determination unit 27 described later. Here, the specific method for the object determination unit 24 to determine whether the query image Q1 or Q2 contains an image related to a predetermined object may be any method different from that of the exclusion determination unit 27 described later, and the specific determination method is not limited. In the present embodiment, the match feature identification unit 23 and the object determination unit 24 identify a plurality of match features using a learned machine learning model, and determine whether the query image Q1 or Q2 contains an image related to a predetermined object.

[0030] Specifically, for example, the match feature identification unit 23 and the object determination unit 24 collate features by attempting to optimize the feature transformation (a homography matrix in the case of a planar object, an essential matrix in the case of a three-dimensional object) between the sample image S1 and the query image Q1 or Q2, and determine that a match is established when optimization is possible (when it is a valid transformation or matrix). In this way, it can be determined whether the query image Q1 or Q2 contains the sample image S1. Since a more specific matching method is a prior art as shown in Non-Patent Documents 1 and 2, the description thereof is omitted.

[0031] The sample distribution index calculation unit 25 calculates a sample distribution index indicating the variation of a plurality of sample features extracted from the sample image S1 related to a predetermined object in the sample image S1. In the present embodiment, the sample distribution index calculation unit 25 calculates the density of the sample features included in each grid when the sample image S1 is divided into a plurality of grids, and calculates the standard deviation of the density of the sample features included in each grid, thereby calculating the sample distribution index. In the present embodiment, an example of using the standard deviation as the distribution index is described, but as the distribution index, known statistical quantities such as variance, which indicate the variation from the average value, may be adopted.

[0032] FIG. 5 is a diagram showing an outline of calculation of the density of sample features in the present embodiment. In the present embodiment, the sample image S1 is equally divided into 16 grids arranged in a 4-row and 4-column matrix, and the density of sample features for each grid is calculated. Specifically, the sample distribution index calculation unit 25 counts the number of features included in each grid and calculates the ratio to the number of features included in the entire image, thereby calculating the density of features for each grid. That is, in the present embodiment, the feature density d i of grid i is calculated according to the following formula.

Equation

[0033] Then, the sample distribution index calculation unit 25 calculates the standard deviation of the feature density of the sample image S1 according to the calculated feature density for each grid. That is, in the present embodiment, the standard deviation sd of the feature density of the sample image S1 is calculated according to the following formula. Here, n is the number of grids (in the present embodiment, n = 16), and d μ is the average density.

Equation

[0034] The match distribution index calculation unit 26 calculates a match distribution index indicating the variation in the sample image S1 of a plurality of match features that match a plurality of query features extracted from the query image Q1 or Q2 among the plurality of sample features. In the present embodiment, the match distribution index calculation unit 26 calculates the density of match features included in each grid when the sample image S1 is divided into a plurality of grids, and calculates the standard deviation of the density of match features included in each grid, thereby calculating the match distribution index.

[0035] FIG. 6 is a diagram showing an overview of the calculation of the density of match features in the present embodiment. The example shown in the upper part shows the density of match features for each grid that matches with the query image Q1 illustrated in FIG. 3, and the example shown in the lower part shows the density of match features for each grid that matches with the query image Q2. Regarding the match distribution index as well, the sample image S1 is equally divided into 16 grids arranged in a 4-row and 4-column matrix, and the density of match features for each grid is calculated. Since the more specific method for calculating the match distribution index (standard deviation of the density of match features) is the same as the method for calculating the sample distribution index described above, the description thereof is omitted.

[0036] FIG. 7 is a diagram showing an overview of comparing the standard deviations of feature densities for each of the sample image S1, query images Q1, and Q2 in the present embodiment. Note that since this figure is for explaining an overview of comparing the standard deviations, the numerical values are not described for the scales of the vertical and horizontal axes. In this figure, the vertical axis indicates the feature density of the grid, and the horizontal axis indicates the deviation from the average value when the center of the horizontal axis is the average value of the feature density. According to this figure, it can be seen that the difference between the standard deviation of the sample image S1 and the standard deviation of the query image Q2 is relatively large compared to the difference between the standard deviation of the sample image S1 and the standard deviation of the query image Q1.

[0037] When the difference between the sample distribution index and the match distribution index exceeds a predetermined standard, the exclusion determination unit 27 determines that the query image Q1 or Q2 does not include an image (sample image S1) related to a predetermined object. When the absolute value of the difference between the standard deviation as the sample distribution index and the standard deviation as the match distribution index exceeds a predetermined threshold value, the exclusion determination unit 27 determines that the query image Q1 or Q2 does not include an image related to a predetermined object regardless of the determination result by the object determination unit 24. That is, in the present embodiment, the exclusion determination unit 27 is a functional unit for determining the correctness of the determination result by the object determination unit 24.

[0038] FIG. 8 is a diagram showing the relationship between the determination result by the object determination unit 24, the standard deviation of the feature density, and the difference in the standard deviation for each of the sample image S1, the query images Q1 and Q2 in the present embodiment. As can be seen by referring to FIGS. 3 and 4, the advertisement A1 related to the sample image S1 is captured in the query image Q1, and the advertisement A1 related to the sample image S1 is not captured in the query image Q2. However, although the advertisement A2 captured in the query image Q2 is different from the advertisement A1 related to the sample image S1, since only the layout is different and there are common points in its elements, the object determination unit 24 according to the present embodiment determines that the advertisement A1 related to the sample image S1 is captured in both the query image Q1 and the query image Q2 (that is, the determination result of the query image Q2 by the object determination unit 24 is incorrect).

[0039] Therefore, the exclusion determination unit 27 compares the distribution index (standard deviation) of the sample features calculated by the sample distribution index calculation unit 25 with the distribution index (standard deviation) of the match features related to each of the query images Q1 and Q2 calculated by the match distribution index calculation unit 26. Then, it can be seen that the difference between the sample distribution index and the match distribution index for the query image Q2 is relatively large compared to the difference for the query image Q1. Since the difference for the query image Q2 exceeds a predetermined threshold, the exclusion determination unit 27 determines that the advertisement related to the sample image S1 is not captured in the query image Q2 regardless of the determination result by the object determination unit 24.

[0040] The notification unit 28 notifies the user of the determination result by the object determination unit 24 and / or the determination result by the exclusion determination unit 27.

[0041] <Flow of processing> Next, the flow of processing executed by the image processing apparatus according to the present embodiment will be described. Note that the specific content and processing order of the processing described below are an example for implementing the present disclosure. The specific processing content and processing order may be appropriately selected according to the embodiment of the present disclosure.

[0042] FIG. 9 is a flowchart showing the flow of the inspection process according to the present embodiment. The processes shown in this flowchart are executed upon receiving an instruction to start the inspection from the user.

[0043] In the inspection process described in this flowchart, first, a first matching process (step S1) is executed to perform a first matching of the feature groups of the query image and a predetermined sample image based on a learned model. Then, the result of the first matching by the first matching process is determined (step S2). When the result of the first matching indicates that the query image is similar to the sample image (YES in step S2), a second matching process (step S3) is executed to perform a second matching regarding the distribution of the feature group of the query image. When the results of the first matching process and / or the second matching process are obtained, the determination result of the installation state is notified to the user (step S4). The notification unit 28 notifies the user of the determination result in step S1 and / or the determination result in step S3. Thereafter, the processes shown in this flowchart end.

[0044] Hereinafter, the details of the processes (steps S101 to S105) included in the first matching process (step S1) will be described.

[0045] In step S101, a query image is acquired. The operator uses the imaging device 81 to image the inspection target (in this embodiment, a store where an advertisement is installed as a predetermined object), and inputs the image data of the obtained query image into the image processing device 1. The imaging method and the method of inputting the image data into the image processing device 1 are not limited. In this embodiment, the inspection target is imaged using the imaging device 81, and the image data transferred from the imaging device 81 to the user terminal 9 via communication or a recording medium is further transferred to the image processing device 1 via the network, so that the image data of the query image is input into the image processing device 1. When the query image acquisition unit 21 acquires the query image, the process proceeds to step S102.

[0046] In steps S102 and S103, features related to the query image and features related to the sample image are acquired. The feature extraction unit 22 extracts query features from the query image acquired in step S101 (step S102). Then, the match feature identification unit 23 acquires sample features extracted from the sample image (step S103). Note that the sample features acquired in step S103 may have been extracted and stored in advance by the feature extraction unit 22 or the like before the start of the processing according to this flowchart, or may have been extracted by the feature extraction unit 22 during the processing according to this flowchart. Thereafter, the process proceeds to step S104.

[0047] In steps S104 and S105, it is determined whether the query image includes the sample image. The match feature identification unit 23 determines whether each of the sample features acquired in step S103 matches the query features extracted in step S102, thereby identifying match features that match any of the query features among the sample features (step S104). Then, the object determination unit 24 determines whether the query image includes an image related to a predetermined object based on the matching result in step S104 (step S105). As described above, in the present embodiment, a method of determining whether the sample image and the query image match is adopted by using a method of determining whether the conversion of features between the sample image and the query image can be optimized. However, the matching method adopted in the first matching process may be a method different from the second matching process described later, and the matching method adopted in the first matching process is not limited to the example shown in the present disclosure. Thereafter, the first matching process ends.

[0048] Hereinafter, the details of the processes (steps S301 to S304) included in the second matching process (step S3) will be described.

[0049] In steps S301 and S302, the standard deviations of the sample features and the standard deviations of the match features included in a plurality of grids into which the sample image is divided are obtained. The exclusion determination unit 27 obtains the standard deviation of the density of the sample features included in each grid when the sample image is divided into a plurality of grids (in this embodiment, 16 grids arranged in a 4-row and 4-column matrix) (step S301). Note that the standard deviation of the sample features obtained in step S301 may be one that has been calculated and stored in advance by the sample distribution index calculation unit 25 or the like before the start of the processing according to this flowchart, or may be one calculated by the sample distribution index calculation unit 25 during the processing according to this flowchart. Further, the match distribution index calculation unit 26 calculates the density of the match features included in each grid when the sample image is divided into a plurality of grids (in this embodiment, 16 grids arranged in a 4-row and 4-column matrix), and calculates the standard deviation of the density of the match features included in each grid (step S302). Then, the processing proceeds to step S303.

[0050] In steps S303 and S304, based on the standard deviation, it is verified whether the sample image is included in the query image. The exclusion determination unit 27 calculates the difference between the standard deviation of the sample features and the standard deviation of the match features, and compares the calculated difference with a preset threshold (step S303). As a result of the comparison, if the calculated difference exceeds the threshold, the exclusion determination unit 27 determines that the object related to the sample image is not imaged in the query image obtained in step S101 (step S304). Then, the second matching process ends.

[0051] That is, according to the processing shown in this flowchart, when the result of the first matching process is affirmative (indicating that the query image is similar to the sample image), the second matching process is executed, and based on the distribution of the features, the determination result of the first matching process is verified.

[0052] <Effect> According to the image processing apparatus, method, and program according to this embodiment, by providing the configuration described above, it is possible to improve the accuracy of image processing for determining whether a target image includes an image related to a predetermined object.

[0053] <Variation> In the embodiment described above, for the correction of the determination result by the first matching process, the second matching process considering the positional relationship and distribution status between the features according to the present disclosure was to be performed. However, the matching process considering the positional relationship and distribution status between the features according to the present disclosure is not limited to the use of correcting the determination result by other matching processes. In this variation, the description of the configuration common to the embodiment described above is omitted, and the differences will be described.

[0054] FIG. 10 is a diagram showing an outline of the functional configuration of the image processing apparatus 1b according to the variation. In the image processing apparatus 1b, the program recorded in the storage device 14 is read into the RAM 13 and executed by the CPU 11, and each hardware provided in the image processing apparatus 1b is controlled, so that it functions as an image processing apparatus including a query image acquisition unit 21, a feature extraction unit 22, a matching feature identification unit 23, a sample distribution index calculation unit 25, a matching distribution index calculation unit 26, an exclusion determination unit 27, and a notification unit 28. That is, in the image processing apparatus 1b according to this embodiment, the object determination unit 24 is omitted from the image processing apparatus 1 shown in FIG. 2.

[0055] FIG. 11 is a flowchart showing the flow of the inspection process according to the variation. The process shown in this flowchart is executed on the occasion of receiving an instruction to start the inspection by the user, similar to the inspection process described with reference to FIG. 9.

[0056] The processes shown from step S501 to step S504 are substantially the same as the processes from step S101 to step S104 of the inspection process described with reference to FIG. 9. The processes shown from step S505 to step S509 are substantially the same as the processes of step S301 to step S304 and step S4 of the inspection process described with reference to FIG. 9. That is, in the inspection process shown in this flowchart, steps S105 and S2 are omitted from the inspection process described with reference to FIG. 9.

[0057] <Other variations> Also, in the above-described embodiment, an example in which a sample image is equally divided into 16 grids arranged in a 4-row and 4-column matrix, and the density of sample features for each grid is calculated has been described. However, the method of dividing the grids is not limited to the example in this embodiment. For example, a division method in which the sample image is equally divided into 9 grids arranged in a 3-row and 3-column matrix may be adopted, or other division methods may be adopted.

Description of reference numerals

[0058] 1 Image processing apparatus

Claims

1. Sample distribution index calculation means for calculating a sample distribution index indicating the variation in a sample image of a plurality of sample features extracted from a sample image related to a predetermined object; Match distribution index calculation means for calculating a match distribution index indicating the variation in the sample image of a plurality of match features that match a plurality of query features extracted from a query image that is a determination target as to whether an image related to the predetermined object is included among the plurality of sample features; Exclusion determination means for determining that an image related to the predetermined object is not included in the query image when the difference between the sample distribution index and the match distribution index exceeds a predetermined criterion; An image processing apparatus comprising the same.

2. The sample distribution index calculation means calculates, as the sample distribution index, the standard deviation of the plurality of sample features in the sample image; The match distribution index calculation means calculates, as the match distribution index, the standard deviation of the plurality of match features in the sample image; The image processing apparatus according to Claim 1.

3. The sample distribution index calculation means calculates the density of the sample features included in each grid when the sample image is divided into a plurality of grids, and calculates the standard deviation of the density of the sample features included in each grid, thereby calculating the sample distribution index; The match distribution index calculation means calculates the density of the match features included in each grid when the sample image is divided into a plurality of grids, and calculates the standard deviation of the density of the match features included in each grid, thereby calculating the match distribution index; The image processing apparatus according to Claim 2.

4. The exclusion determination means determines that an image related to the predetermined object is not included in the query image when the difference between the standard deviation as the sample distribution index and the standard deviation as the match distribution index exceeds a predetermined threshold; The image processing apparatus according to Claim 3.

5. The apparatus further comprises object determination means for determining whether an image related to the predetermined object is included in the query image by a method different from that of the exclusion determination means based on the plurality of match features. When the difference between the sample distribution index and the match distribution index exceeds a predetermined criterion, the exclusion determination means determines that the query image does not include an image related to the predetermined object regardless of the determination result by the object determination means. The image processing apparatus according to claim 1.

6. The object determination means determines whether the query image includes an image related to the predetermined object based on the plurality of match features using a learned machine learning model. The image processing apparatus according to claim 5.

7. The apparatus further comprises match feature specifying means for specifying, among the plurality of sample features, a plurality of match features that match any of the plurality of query features by determining whether each of the plurality of sample features matches the plurality of query features. The image processing apparatus according to claim 1.

8. The apparatus further comprises feature extraction means for extracting the plurality of query features from the query image. The image processing apparatus according to claim 1.

9. The feature extraction means further extracts the plurality of sample features from the sample image. The image processing apparatus according to claim 8.

10. A computer A sample distribution index calculation step of calculating a sample distribution index indicating the variation in the sample image of a plurality of sample features extracted from a sample image related to a predetermined object; A match distribution index calculation step of calculating a match distribution index indicating the variation in the sample image of a plurality of match features that match a plurality of query features extracted from a query image that is a determination target as to whether the image related to the predetermined object is included among the plurality of sample features; An exclusion determination step of determining that the query image does not include an image related to the predetermined object when the difference between the sample distribution index and the match distribution index exceeds a predetermined criterion; An image processing method for executing the above.

11. A computer A sample distribution index calculation means for calculating a sample distribution index indicating the variation in the sample image of a plurality of sample features extracted from a sample image related to a predetermined object; Match distribution index calculation means for calculating a match distribution index indicating the variation in the sample image of a plurality of match features that match a plurality of query features extracted from a query image that is a determination target as to whether an image related to the predetermined object is included among the plurality of sample features; Exclusion determination means for determining that the image related to the predetermined object is not included in the query image when the difference between the sample distribution index and the match distribution index exceeds a predetermined reference; An image processing program that functions as.

Images