Information processing apparatus, information processing method, and program

Abstract

To provide an information processing device, an information processing method, and a program capable of more accurately evaluating validity of an object region extracted from an image.SOLUTION: An information processing device includes an extraction unit and an evaluation unit. The extraction unit extracts a first object region including an object from an image. The processing circuitry calculates an evaluation value for evaluating validity of the first object region based on a first evaluation criterion and a second evaluation criterion different from the first evaluation criterion.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] An embodiment of the present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Labor-saving measures are being implemented through the automation of logistics. For example, there are systems that use picking robots to automate picking tasks. Picking robot systems use image recognition technology to recognize the position of objects such as goods in containers. However, if the recognition is incorrect, there is a risk that the object may be destroyed by undesired robot operation. Therefore, there is a need for technology to evaluate the validity of object regions extracted by image recognition. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7242882 [Non-patent literature]

[0004] [Non-Patent Document 1] Test-time Augmentation for Active Learning in Medical Image Segmentation,TAAL [MICCAI-DALI 2022] Summary of the Invention [Problem to be solved by the invention]

[0005] However, with conventional techniques, it has been difficult to accurately evaluate the validity of an object region extracted from an image. [Means for solving the problem]

[0006] According to an embodiment, an information processing apparatus includes an extraction unit and an evaluation unit. The extraction unit extracts a first object region including an object from an image. The evaluation unit calculates an evaluation value for evaluating the validity of the first object region based on a first evaluation criterion and a second evaluation criterion different from the first evaluation criterion. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 2 is a diagram showing an example of the functional configuration of the information processing apparatus according to the embodiment. [Figure 2] FIG. 2 is a diagram showing an example of the functional configuration of an evaluation unit according to the embodiment. [Figure 3] 10A and 10B are diagrams illustrating an example of processing by an extension unit and an extraction unit according to the embodiment. [Figure 4] FIG. 6 is a diagram showing an example of processing by a first evaluation unit according to the embodiment. [Figure 5] FIG. 10 is a diagram showing an example of processing by a second evaluation unit according to the embodiment. [Figure 6] 1 is a flowchart illustrating an example of an information processing method according to an embodiment. [Figure 7] FIG. 10 is a diagram showing an example of the functional configuration of an evaluation unit according to a modified example of the embodiment. [Figure 8] FIG. 10 is a diagram for explaining an example of object information according to a modified example of the embodiment. [Figure 9] FIG. 1 is a diagram illustrating an example of a hardware configuration of an information processing apparatus according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of an information processing device, an information processing method, and a program will be described in detail with reference to the accompanying drawings.

[0009] First, an example of the functional configuration of the information processing apparatus according to the embodiment will be described.

[0010] [Example of functional configuration] 1 is a diagram illustrating an example of the functional configuration of an information processing device 100 according to an embodiment. The information processing device 100 according to the embodiment includes an image acquisition unit 1, an extension unit 2, an extraction unit 3, an input unit 4, an object information acquisition unit 5, an object information DB 6, an evaluation unit 7, and a display unit 8.

[0011] The image acquisition unit 1 acquires one or more images including an object to be evaluated. For example, the images are one or more photographed images of the object to be evaluated. The object to be evaluated may be any object. In the embodiment, an example will be described in which the object is a product.

[0012] In addition, the method for acquiring the captured image may be any method. In the embodiment, the captured image is described as an RGB image, but the captured image may be a depth image, a PCD (Point Cloud Data), or an image captured by various sensors (such as an infrared camera) in addition to the RGB image.

[0013] Specifically, for example, the image acquisition unit 1 acquires an image of the product to be evaluated from an image database in which photographed images are stored. Alternatively, for example, the image acquisition unit 1 may take an image using a photography system that photographs the product to be evaluated.

[0014] The expansion unit 2 generates a data-augmented image by performing data expansion processing on each captured image acquired by the image acquisition unit 1. The data expansion processing may be any processing that performs a predetermined conversion processing. For example, the data expansion processing includes at least one of predetermined conversion processing such as horizontal flipping, vertical flipping, rotation, enlargement, reduction, noise addition, padding, geometric transformation, and color conversion. The noise addition processing is, for example, processing that adds noise to an image to reduce the quality of the image.

[0015] The extraction unit 3 extracts an object region (first object region) showing an object (in this embodiment, a commodity) from the captured image. The extraction unit 3 also extracts an object region (second object region) showing an object (in this embodiment, a commodity) from each data augmented image generated by the augmentation unit 2.

[0016] Any method may be used for extracting an object region by the extraction unit 3. For example, the extraction unit 3 extracts an object region using an instance segmentation technique such as Mask R-CNN (Convolutional Neural Network).

[0017] The details of the processing by the expansion unit 2 and extraction unit 3 will be described later with reference to FIG.

[0018] The input unit 4 is a UI (User Interface) for inputting input information indicating an object to be evaluated. The input unit 4 may have any configuration. For example, the input unit 4 is a UI such as a keyboard and a mouse. Alternatively, for example, the input unit 4 may be an interface for reading product data saved in a predetermined format such as a CSV file.

[0019] The object information acquisition unit 5 acquires object information (in the embodiment, product information) regarding the object (in the embodiment, product) to be evaluated. The object information includes at least one of identification information for identifying the object, the size of the object, the shape of the object, the number of objects, and the texture of the object.

[0020] For example, the identification information for identifying an object is the ID of the object. For example, the size of an object is represented by the external dimensions and volume of the object. For example, if the object is a product, the texture of the object is an image of the product's packaging.

[0021] The object information may further include the mass of the object, CAD data (three-dimensional shape data) of the object, and the like.

[0022] For example, the object information acquisition unit 5 may acquire input information including object information from the input unit 4. Furthermore, for example, the object information acquisition unit 5 may acquire object information registered in the object information DB 6 in accordance with the input information from the input unit 4.

[0023] The object information DB (in the embodiment, product information DB) 6 is a database that stores object information (in the embodiment, product information). The object information DB 6 may be stored in a storage device such as an HDD (Hard Disk Drive) of the information processing device 100, or may be stored in a data server device connected to the information processing device 100.

[0024] The evaluation unit 7 calculates an evaluation value for evaluating the validity of the object region based on the object region extracted by the extraction unit 3 and the object information acquired by the object information acquisition unit 5. Details of the processing by the evaluation unit 7 will be described later with reference to FIG.

[0025] The display unit 8 displays evaluation information based on the evaluation value output from the evaluation unit 7. The configuration of the display unit 8 may be arbitrary. For example, the display unit 8 is a display monitor that displays display information including the object region to be evaluated and evaluation information based on the evaluation value of the object region. The method of displaying the display information may be arbitrary. For example, the display unit 8 may display the evaluation value as is as the evaluation information. Also, for example, the display unit 8 may display display information including evaluation information based on the evaluation value (for example, text information indicating whether the evaluation result is appropriate or not).

[0026] For example, the object to be evaluated can be explicitly specified by the input unit 4, and the display unit 8 can display the evaluation result of the specified object, thereby switching the subsequent processing depending on whether the evaluation result is valid or not. For example, in a picking robot system that picks objects, if there is a high possibility that the extraction result of the object region contains an error (for example, if the evaluation value is below a threshold), measures can be taken such as having the processing be done by a person rather than by a robot.

[0027] 2 is a diagram illustrating an example of the functional configuration of the evaluation unit 7 according to the embodiment. The evaluation unit 7 according to the embodiment includes a first evaluation unit 71, a second evaluation unit 72, and a calculation unit 73.

[0028] The first evaluation unit 71 calculates an evaluation value for evaluating the validity of the object region based on the first evaluation criterion. Specifically, the first evaluation unit 71 calculates the first evaluation value based on the object region extracted by the extraction unit 3. Details of the processing by the first evaluation unit 71 will be described later with reference to FIGS. 3 and 4.

[0029] The second evaluation unit 72 calculates an evaluation value for evaluating the validity of the object region based on a second evaluation criterion different from the first evaluation criterion. Specifically, the second evaluation unit 72 calculates the second evaluation value based on the object information acquired by the object information acquisition unit 5. Details of the processing by the second evaluation unit 72 will be described later with reference to FIG.

[0030] The calculation unit 73 calculates an evaluation value based on the first and second evaluations. Any method may be used to calculate the evaluation value based on the first and second evaluations. For example, the calculation unit 73 calculates the evaluation value by obtaining a weighted sum, an average value, a minimum value, a maximum value, or the like from the first and second evaluation values.

[0031] Fig. 3 is a diagram showing an example of processing by the expansion unit 2 and extraction unit 3 of the embodiment. In the example of Fig. 3, the expansion unit 2 performs data expansion processing on the captured image by performing left-right flip processing, up-down flip processing, and 180° rotation processing, thereby obtaining three data-augmented images. The example of Fig. 3 shows a case where data expansion processing is performed on a captured image taken from above of a container containing four objects (e.g., products) with rectangular top surfaces.

[0032] The extraction unit 3 extracts an object region from each of the captured image and three data-augmented images. After extracting the object region from each of the three data-augmented images, the extraction unit 3 performs data augmentation processing to restore the position and orientation of the image to those of the original captured image. In the example of Fig. 3, four extraction results (1) to (4) are obtained for one captured image.

[0033] The type of data augmentation processing (the number of data augmented images) may be arbitrary. Furthermore, the extraction unit 3 may extract the object region from the data augmented image without extracting the object region from the captured image. That is, in the example of FIG. 3, the extraction unit 3 may extract extraction results (2) to (4) from each of the three data augmented images.

[0034] If the extraction process by the extraction unit 3 is robust even for data-augmented images after data augmentation processing, that is, if the object region extraction results are consistent, it can be considered that a valid object region has been extracted from the captured image.

[0035] Based on this idea, the first evaluation unit 71 uses, as a first evaluation criterion, the degree of match between an object region (first object region) extracted from a captured image and an object region (second object region) extracted from at least one data-augmented image. The greater the degree of match, the higher the evaluation value calculated by the first evaluation unit 71 for the first object region.

[0036] Specifically, in the example of FIG. 3, the first evaluation unit 71 performs evaluation based on the degree of match between the extraction results (1) to (4). For example, the first evaluation unit 71 calculates image features from each of the extraction results (1) to (4) and calculates the degree of match based on the similarity of the image features. Specifically, the first evaluation unit 71 may calculate the degree of match between the feature amounts using cosine similarity or the like. Note that although an example has been described in which cosine similarity is used as the evaluation index for the similarity of the image features, other evaluation indexes, such as the distance between image features, may also be used.

[0037] Furthermore, for example, the first evaluation unit 71 may calculate the degree of match based on the overlap rate of the extraction results (1) to (4) (masks of the object region).

[0038] FIG. 4 is a diagram illustrating an example of processing by the first evaluation unit 71 according to the embodiment. The example in FIG. 4 illustrates an example of calculating the degree of match based on the overlap rate. Specifically, the example in FIG. 4 illustrates an example of evaluating the degree of match between extraction result (1) and extraction result (2). First, the first evaluation unit 71 associates object regions A to D obtained in extraction results (1) and (2) and calculates their overlap rate (IoU: Intersection over Union). For example, the first evaluation unit 71 determines the average value of the IoU calculated for each of the object regions A to D as the degree of match. The mode, maximum value, minimum value, median, or the like may be used as the statistic of the IoU calculated for each of the object regions A to D.

[0039] Although an example has been described in which IoU is used as an evaluation index for the overlap rate, other evaluation indexes may be used.

[0040] The first evaluation unit 71 calculates the degree of match for each combination of the extraction results (1) to (4) (4C2=6 combinations), and sets the average value of the degrees of match as the evaluation value. The statistical value of the degree of match calculated for each combination may be the mode, maximum value, minimum value, median value, or the like.

[0041] The combinations for calculating the degree of match may be such that the extraction result (1) is used as a reference and statistics of the degree of match are calculated for three combinations (2) to (4).

[0042] Fig. 5 is a diagram showing an example of processing by the second evaluation unit 72 in the embodiment. In the example of Fig. 5, the extraction result of the object region obtained for an image containing three objects to be evaluated (products in the embodiment) and the evaluation thereof will be described. In this extraction result, extracted regions E and F are obtained. Extracted region E is an example where the region was correctly extracted. On the other hand, extracted region F is an example where two adjacent products were mistakenly extracted as one product.

[0043] The second evaluation unit 72 uses, as the second evaluation criterion, the degree of similarity between the object features indicated by the object region (first object region) extracted from the captured image and the object information indicating the object features. The second evaluation unit 72 calculates a higher evaluation value for the first object region as the degree of similarity between the object features indicated by the first object region and the object features indicated by the object information increases.

[0044] Specifically, in the example of Fig. 5, the second evaluation unit 72 calculates the expected size of the product on the image based on object information (e.g., shape information such as the product's external dimensions) and evaluates the validity of extraction areas E and F. If the expected size is as shown in Fig. 5, extraction area E is evaluated as having a valid result. On the other hand, extraction area F is larger than the expected size, and therefore is evaluated as having an abnormal result.

[0045] The second evaluation unit 72 compares the expected size with the area of ​​each of the extraction regions E and F, and calculates a lower evaluation value the greater the difference in area. In other words, the second evaluation unit 72 calculates a higher evaluation value the smaller the difference in area. Alternatively, the second evaluation unit 72 may calculate the evaluation value based on the ratio between the expected size and the area of ​​each of the extraction regions E and F, or the expected size and the aspect ratio of the length and width of each of the extraction regions E and F.

[0046] In the first evaluation unit 71, if the extraction results (1) to (4) are similarly incorrect, as in the case of extraction area F in Figure 5, for example, it is not possible to detect that extraction area F is abnormal. However, by evaluating the extraction area based on object information as in Figure 5, it is possible to correctly calculate the evaluation value.

[0047] For example, the expected size may be calculated from the external dimensions using a conversion table. For example, the conversion table is a table that converts the external dimensions into size on an image (how it appears on an image). Also, for example, the expected size may be calculated from an ID that identifies an object using a conversion table that associates IDs with expected sizes.

[0048] Although an example in which an estimated size is used as object information has been described, the object information is not limited to the estimated size. For example, the number of objects may be used as object information. In this case, the second evaluation unit 72 evaluates the extraction result by comparing the number specified by the object information with the number of extracted regions extracted from the image. Furthermore, for example, if the object is a product, the product package image may be used as object information. In this case, the second evaluation unit 72 compares the similarity between the package image registered as object information and the extracted region extracted from the image, and evaluates the extraction result based on the similarity.

[0049] [Examples of information processing methods] 6 is a flowchart showing an example of an information processing method according to the embodiment. First, the image acquisition unit 1 acquires one or more images including an object to be evaluated (in the embodiment, a product) (step S1).

[0050] Next, the extension unit 2 performs data extension processing including at least one predetermined conversion processing on the image acquired in step S1, thereby generating at least one data-extended image (step S2).

[0051] Next, the extraction unit 3 extracts an image region including the object (step S3). Specifically, the extraction unit 3 extracts an object region including the object (first object region) from the image acquired in step S1. The extraction unit 3 also extracts an object region including the object (second object region) from each data-augmented image generated in step S2.

[0052] Next, the object information acquisition unit 5 acquires object information (product information in this embodiment) relating to the object to be evaluated (step S4).

[0053] Next, the evaluation unit 7 calculates an evaluation value for evaluating the validity of the first object region based on the first evaluation criterion and a second evaluation criterion different from the first evaluation criterion (step S5). For example, the first evaluation criterion is the degree of coincidence between the first object region and at least one second object region. Furthermore, the second evaluation criterion is the degree of similarity between the object features indicated by the first object region and the object information indicating the object features.

[0054] Finally, the display unit 8 displays the evaluation information based on the evaluation value calculated in step S5 (step S6).

[0055] As described above, in information processing device 100 of the embodiment, extraction unit 3 extracts an object region (first object region) including an object from an image. Evaluation unit 7 calculates an evaluation value for evaluating the validity of the first object region based on a first evaluation criterion and a second evaluation criterion different from the first evaluation criterion.

[0056] As a result, the information processing device 100 according to the embodiment can evaluate the validity of the object region extracted from the image with higher accuracy. Specifically, by calculating the evaluation value according to two different evaluation criteria, it is expected that an error that occurs when evaluation is performed according to a single evaluation criterion (one of the evaluation criteria) can be corrected by evaluation according to the other evaluation criterion.

[0057] The information processing device 100 according to the embodiment can be applied to visual inspection of products in factories, etc. For example, when an object region is extracted in a state where the object to be processed is known (a state where object information is obtained), the validity of the extraction result can be evaluated more correctly.

[0058] In the above embodiment, an example in which an evaluation value is calculated for one image has been described, but the evaluation unit 7 may calculate an evaluation value based on two or more images. In this case, the evaluation unit 7 calculates multiple evaluation values ​​from multiple images and calculates statistics of the multiple evaluation values. The statistics may be an average value, a mode value, a maximum value, a minimum value, or a median value. Calculating statistics based on evaluation values ​​of multiple images is expected to have the effect of obtaining more stable evaluation results. For example, this has the effect of increasing robustness against errors in evaluation values ​​in some images.

[0059] (Modification of the embodiment) Next, a modified example of the embodiment will be described. In the description of the modified example, the same description as in the embodiment will be omitted, and only the differences from the embodiment will be described.

[0060] 7 is a diagram showing an example of the functional configuration of an evaluation unit 7-2 according to a modification of the embodiment. The evaluation unit 7-2 according to the modification includes a first evaluation unit 71-2 and a calculation unit 73.

[0061] In a modified example, the evaluation unit 7-2 identifies the internal area in which the object to be evaluated is stored based on position information indicating the internal area in which the object to be evaluated is stored. The second evaluation criterion is whether the first object area and the second object area are within the range of the internal area in which the object to be evaluated is stored. The evaluation unit 7-2 calculates an evaluation value that evaluates the validity of the first object area, limited to the range of the identified internal area.

[0062] Specifically, in the modified example, the first evaluation unit 71-2 calculates the evaluation value based on, for example, object information further including location information where a specific object (object to be evaluated) is stored.

[0063] Fig. 8 is a diagram for explaining an example of object information according to a modified example of the embodiment. Fig. 8 shows an example in which one container is divided into multiple compartments and different objects (e.g., products) are stored in the compartments. In this example, a case will be described in which it is evaluated whether a specific object (e.g., an object in the lower left compartment (3)) has been correctly recognized.

[0064] The image in Figure 8 also shows objects other than the object being evaluated (sections (1), (2), and (4)). Conventionally, there has been a problem in that the evaluation value for the object being evaluated cannot be calculated correctly due to the influence of the objects in sections (1), (2), and (4).

[0065] In this example, the object information further includes compartment information of the container (e.g., how the container is divided and coordinates indicating the partition locations), an object ID that identifies the object, and location information that indicates the location of the internal area in which the object identified by the object ID is stored.

[0066] The first evaluation unit 71-2 calculates the evaluation value by taking into account only the section (3) in the evaluation value calculation shown in the example of Fig. 4. This makes it possible to more accurately calculate the evaluation value for a specific object.

[0067] The format of the container compartment information is not limited to position information such as coordinates, etc. For example, the compartment information may be given by a mask image in which only the area to be evaluated is enabled.

[0068] As described above, in the modified example, objects are stored in a plurality of internal regions within a container according to the object type. The object information further includes position information for each of the plurality of internal regions. The evaluation unit 7-2 then identifies the internal region in which the object to be evaluated is stored from the position information, and calculates an evaluation value for evaluating the validity of the object region (first object region) by limiting the range to the specified internal region.

[0069] As a result, according to the modified example, when multiple types of products are stored in a container in a categorized manner, it becomes possible to focus on and evaluate only a specific product.

[0070] Finally, an example of the hardware configuration of the information processing apparatus 100 according to the embodiment will be described.

[0071] [Example of hardware configuration] 9 is a diagram illustrating an example of a hardware configuration of the information processing device 100 according to the embodiment. The information processing device 100 according to the embodiment includes a processor 201, a main storage device 202, an auxiliary storage device 203, a display device 204, an input device 205, and a communication device 206. The processor 201, the main storage device 202, the auxiliary storage device 203, the display device 204, the input device 205, and the communication device 206 are connected via a bus 210.

[0072] Note that some of the above configurations may not be provided in the information processing device 100. For example, if the information processing device 100 can use the input function and display function of an external device, the information processing device 100 may not be provided with the display device 204 and the input device 205.

[0073] The processor 201 executes a program read from the auxiliary storage device 203 to the main storage device 202. The main storage device 202 is a memory such as a ROM and a RAM. The auxiliary storage device 203 is a HDD, a memory card, or the like.

[0074] The display device 204 is, for example, a liquid crystal display. The input device 205 is an interface for operating the information processing device 100. The display device 204 and the input device 205 may be realized by a touch panel or the like having a display function and an input function. The communication device 206 is an interface for communicating with other devices.

[0075] For example, the program executed by the information processing device 100 is provided as a computer program product in the form of a file in an installable or executable format, recorded on a computer-readable storage medium such as a memory card, hard disk, CD-RW, CD-ROM, CD-R, DVD-RAM, and DVD-R.

[0076] Furthermore, for example, the program executed by the information processing device 100 may be stored on a computer connected to a network such as the Internet, and may be provided by being downloaded via the network.

[0077] Furthermore, for example, the program executed by the information processing device 100 may be provided via a network such as the Internet without being downloaded. Specifically, the information processing may be performed by a so-called ASP (Application Service Provider) type service that realizes processing functions by issuing an execution instruction and obtaining the results from a server computer without transferring the program.

[0078] Furthermore, for example, the program for the information processing device 100 may be provided by being pre-installed in a ROM or the like.

[0079] The program executed by the information processing device 100 has a modular configuration that includes functions that can be realized by the program among the above-mentioned functional configurations. As for each function, the processor 201 reads the program from a storage medium and executes it, and the above-mentioned functional blocks are loaded onto the main storage device 202, in terms of actual hardware. In other words, the above-mentioned functional blocks are generated on the main storage device 202.

[0080] Note that some or all of the above-described functions may be realized by hardware such as an integrated circuit (IC) rather than by software.

[0081] Furthermore, each function may be realized using a plurality of processors 201, in which case each processor 201 may realize one of the functions, or may realize two or more of the functions.

[0082] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0083] 1 Image acquisition unit 2 Extension 3 Extraction part 4 Input section 5 Object information acquisition section 6 Object information DB 7 Evaluation Section 8 Display 71 First Evaluation Section 72 Second Evaluation Section 73 Calculation Unit 100 Information processing device 201 processor 202 Main storage 203 Auxiliary storage device 204 Display device 205 Input Device 206 Communication Equipment 210 Bus

Claims

1. an extraction unit that extracts a first object region including an object from an image; an evaluation unit that calculates an evaluation value that evaluates the validity of the first object region based on a first evaluation criterion and a second evaluation criterion different from the first evaluation criterion; An information processing device comprising:

2. an extension unit that generates at least one data-extended image by performing a data extension process including at least one predetermined conversion process on the image; the extractor further extracts a second object region including the object from the at least one data-augmented image; the first evaluation criterion is a degree of coincidence between the first object region and the at least one second object region; The evaluation unit calculates a higher evaluation value for the first object region as the degree of match increases. The information processing device according to claim 1 .

3. the second evaluation criterion is a similarity between a feature of the object indicated by the first object region and object information indicating the feature of the object; the evaluation unit calculates a higher evaluation value for the first object region as the degree of similarity between the feature of the object indicated by the first object region and the feature of the object indicated by the object information increases; 3. The information processing device according to claim 1.

4. the object information includes at least one of identification information for identifying the object, a size of the object, a shape of the object, the number of the objects, and a texture of the object; The information processing device according to claim 3 .

5. an input unit for inputting input information indicating the object to be evaluated; an acquisition unit that acquires the object information based on the input information; a display unit that displays evaluation information based on the evaluation value; The information processing device according to claim 3 , further comprising:

6. the objects are stored in a plurality of internal regions included in the container, each region being divided into different regions for each type of object; the evaluation unit identifies an internal area in which the object to be evaluated is stored based on position information indicating the internal area in which the object to be evaluated is stored; the second evaluation criterion is whether or not the first object region and the second object region are within a range of an internal region in which the object to be evaluated is stored; the evaluation unit calculates an evaluation value for evaluating the validity of the first object region within a range of the identified internal region. The information processing device according to claim 2 .

7. the evaluation unit calculates a plurality of the evaluation values ​​from a plurality of the images and calculates statistics of the plurality of the evaluation values; 3. The information processing device according to claim 1.

8. The object is an object to be picked by a picking robot.

3. The information processing device according to claim 1.

9. An information processing device extracts a first object region including an object from an image; a step in which the information processing device calculates an evaluation value that evaluates the validity of the first object region based on a first evaluation criterion and a second evaluation criterion different from the first evaluation criterion; An information processing method including:

10. Computer, an extraction unit that extracts a first object region including an object from an image; an evaluation unit that calculates an evaluation value that evaluates the validity of the first object region based on a first evaluation criterion and a second evaluation criterion that is different from the first evaluation criterion; A program to function as a

Images