IMAGE SELECTION DEVICE, IMAGE SELECTION METHOD, AND PROGRAM
The image selection device iteratively adjusts selection criteria based on user input to enhance accuracy, addressing the issue of selecting undesired images in multiple-image scenarios.
Patent Information
- Application Number
- JP2022530451
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-06-11
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-06-11
AI Technical Summary
Existing image selection methods are prone to selecting images other than the user's desired image, especially when multiple images are involved.
An image selection device and method that acquires query information, sets selection conditions, displays images, and iteratively adjusts these conditions based on user input to refine the selection process.
Improves the accuracy of image selection by allowing users to correct and re-adjust selection criteria based on displayed images, ensuring higher precision in choosing desired images.
Smart Images

Figure 0007586177000001 
Figure 0007586177000002 
Figure 0007586177000003
Abstract
Description
[Technical field]
[0001] The present invention relates to an image selection device, an image selection method, and a program. [Background technology]
[0002] In recent years, cameras are installed in various places. One of the ways to use images from such cameras is to select images that satisfy query information. In this image search, images can be used as query information. For example, Patent Document 1 describes a method of extracting feature data from each still image that constitutes a query video, and searching for similar videos using this feature data.
[0003] Patent Document 2 discloses the following technique: First, a user registers a plurality of selection key images in the system. Then, the user searches for images using a search key image.
[0004] Patent Document 3 also discloses the following image search device. First, the image search device searches for similar images based on an input query image, and obtains a first search result including a plurality of ranked images. Next, some of the images ranked highly in the first search result are specified as images to be displayed. Next, similar images are searched again based on the images ranked highly in the first search result.
[0005] Furthermore, Patent Document 4 discloses the following image retrieval device. First, the image retrieval device performs a comparative search between a query image selected by a user and N reference images stored in an image database, and extracts K similar images according to the matching rank. Next, the image retrieval device stores the top M similar images of the K similar images in a buffer, and stores the remaining (KM) similar images in a separate storage medium. Next, the image retrieval device adopts the most similar image of the top M similar images stored in the buffer as a new query image, performs matching with the M similar images stored in the buffer, and rearranges the matching rank. [Prior art documents] [Patent documents]
[0006] [Patent Document 1] JP 2007-323268 A [Patent Document 2] JP 2003-150634 A [Patent Document 3] JP 2010-25059 A [Patent Document 4] JP 2004-133947 A Summary of the Invention [Problem to be solved by the invention]
[0007] When a user selects a desired image from a plurality of images, as the number of images to be selected increases, the selection result is likely to include images other than the image desired by the user. One of the problems that the present invention aims to solve is to improve the accuracy of image selection. [Means for solving the problem]
[0008] According to the present invention, a selection unit performs a selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions; a display processing means for displaying the plurality of images on a display means; Equipped with The selection means is receiving a selection input indicating a selection of at least one of the images from the plurality of images; There is provided an image selection device that changes the selection condition using the image indicated by the selection input, and performs the selection process again using the changed selection condition.
[0009] According to the present invention, a computer A selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions. Reason and , a display process for displaying the plurality of images on a display means; Do the following: In the selection process, the computer receiving a selection input indicating a selection of at least one of the images from the plurality of images; An image selection method is provided in which the image indicated by the selection input is used to change the selection condition, and the selection process is performed again using the changed selection condition.
[0010] According to the present invention, a computer includes: a selection function that performs a selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions; a display processing function for displaying the plurality of images on a display means; Let them have The selection function is receiving a selection input indicating a selection of at least one of the images from the plurality of images; A program is provided that changes the selection condition using the image indicated by the selection input, and performs the selection process again using the changed selection condition. Effect of the Invention
[0011] According to the present invention, the accuracy of image selection is improved. [Brief description of the drawings]
[0012] The above objects, as well as other objects, features and advantages, will become more apparent from the following preferred embodiments and the accompanying drawings.
[0013] [Figure 1] FIG. 2 is a diagram illustrating an example of a functional configuration of an image selection device according to an embodiment. [Diagram 2] FIG. 4 is a diagram showing an example of selection conditions stored in a selection unit; [Diagram 3] FIG. 2 is a diagram illustrating an example of a hardware configuration of a main part of an image selection device. [Figure 4] 13 is a flowchart illustrating an example of a process performed by the image selection device. [Diagram 5] FIG. 11 is a diagram showing an example of a screen displayed on the display unit in step S40. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are given the same reference numerals and the description will be omitted as appropriate.
[0015] FIG. 1 is a diagram showing an example of a functional configuration of an image selection device 10 according to an embodiment. The image selection device 10 includes a selection unit 110 and a display processing unit 120. The selection unit 110 acquires query information, sets image selection conditions (e.g., image features) using the query information, and performs a process (hereinafter, referred to as selection process) of selecting multiple images using the selection conditions. The display processing unit 120 displays the multiple images selected by the selection unit 110 on the display unit 130, for example, a display. Here, the selection unit 110 acquires an input (hereinafter, referred to as selection input) indicating the result of selecting at least one image from the multiple images after the display unit 130 displays the multiple images. This selection input is performed by a user of the image selection device 10. The selection unit 110 then changes (e.g., modifies) the selection conditions using the image indicated by the selection input, and performs the selection process again using the changed selection conditions. Note that the selection unit 110 performs the above-mentioned choice The conditions are memorized.
[0016] The image to be selected may be a still image or a video. In the latter case, it may be the video of the initial query information. In this case, an example of the selection condition is an indication of the movement of a predetermined moving object (e.g., an animal such as a person). Here, an example of "information indicating movement" is, for example, a plurality of frame images showing different postures from each other being detected in a predetermined order. The process performed by the selection unit 110 on each frame image is the same as the process performed on a still image.
[0017] As described above, the selection unit 110 changes the selection conditions using the selected image. For example, the selection conditions may include multiple items (e.g. multiple feature amounts), and an image may be selected using the scores of these multiple items. As an example, the similarity may be calculated using the scores of these multiple items, and the condition for selecting the image may be that the similarity is equal to or greater than a reference value. In such a case, the selection unit 110 does not change the items included in the selection conditions, but changes the algorithm for selecting an image using the scores of multiple items. As an example, the selection unit 110 may calculate the similarity using the scores of multiple items. of When calculating, the weighting of these multiple items is changed. Note that the weighting coefficient may become 0.
[0018] Here, the user may select an image that he / she desires from the images displayed on display unit 130, or may select an image that he / she does not desire. In the former case, selection unit 110 changes the selection conditions so that the selected image becomes easier to select. On the other hand, in the latter case, selection unit 110 changes the selection conditions so that the selected image becomes harder to select.
[0019] When modifying the selection conditions, the selection unit 110 may delete some items from the selection conditions or add new items to the selection conditions. Even in this case, it is preferable that at least one of the items before modification remains in the selection conditions.
[0020] When an image includes a person, a user of the image selection device 10 may select an image in which the person's posture or movement is similar. In this case, at least some of the items included in the selection conditions relate to the relative positions of different parts of the human body with respect to other parts. These parts are, for example, the person's head and joints, and the relative positions are the lines connecting them. As an example, the parts are, for example, the head, neck, right shoulder, left shoulder, right elbow, left elbow, right hand, left hand, right hip, left hip, right knee, left knee, right foot, and left foot.
[0021] The query information acquired by the selection unit 110 may include an image, may include text information, or may include both of these. 。 For example, when the query information is an image, the selection unit 110 extracts a feature from the image and sets the feature as a selection condition. When the query information is text information, the selection unit 110 converts the text information into a selection condition according to a predetermined conversion rule.
[0022] In addition, when the image serving as the query information shows a human movement, the image may be generated by photographing a person performing the movement, or may be generated by computer graphics. In the latter case, the movement may be shown by a marionette instead of a person. The image selection device 10 may include a processing unit for performing the computer graphics.
[0023] The selection unit 110 and the display processing unit 120 may repeat the above-mentioned process multiple times. In this case, after the selection process is performed again by the selection unit 110, the display processing unit 120 causes the display unit 130 to display again the multiple images selected by the selection process. The selection unit 110 obtains a selection input again, and then changes the selection conditions and performs the selection process again.
[0024] In the example shown in the figure, a plurality of images that are used as a population for selecting an image are stored in the image storage unit 140. These plurality of images are images generated by, for example, a surveillance camera. In this case, there may be a plurality of surveillance cameras. In this case, the plurality of surveillance cameras may be installed in the same area, but apart from each other. Note that the image storage unit 140 may be a part of the image selection device 10, or may be located outside the image selection device 10.
[0025] 2 is a diagram showing an example of selection conditions stored in the selection unit 110. In the example shown in this figure, the selection unit 110 stores body parts and relative positions as items included in the selection conditions. The selection unit 110 also stores a weighting for each item.
[0026] 3 is a diagram showing an example of the hardware configuration of the main part of the image selection device 10. The image selection device 10 includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.
[0027] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, input / output interface 1050, and network interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to bus connection.
[0028] The processor 1020 is implemented by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.
[0029] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.
[0030] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read only memory (ROM), or the like. The storage device 1040 stores program modules that realize each function of the image selection device 10 (e.g., the selection unit 110 and the display processing unit 120). The processor 1020 loads each of these program modules onto the memory 1030 and executes them, thereby realizing each function corresponding to the program module. The storage device 1040 may also function as an image storage unit 140.
[0031] The input / output interface 1050 is an interface for connecting the main part of the image selecting device 10 to various input / output devices (for example, the display unit 130).
[0032] The network interface 1060 is an interface for connecting the image selection device 10 to a network. This network is, for example, a local area network (LAN) or a wide area network (WAN). The method for connecting the network interface 1060 to the network may be a wireless connection or a wired connection.
[0033] 4 is a flowchart showing an example of processing performed by the image selection device 10. First, a user of the image selection device 10 inputs query information to the selection unit 110 (step S10). The selection unit 110 then sets selection conditions using the input query information (step S20), and selects multiple images from the images stored in the image storage unit 140 using the selection conditions (step S30). The display processing unit 120 causes the display unit 130 to display the multiple images selected in step S30 (step S40).
[0034] If the user of the image selection device 10 is not satisfied with the selected image, the user may select an image on the display unit 130. display At least one image is selected from the multiple images displayed (step S50: Yes). An input indicating this selection result is made using an input device such as a touch panel or a mouse. Here, the user may select an image he or she desires, or may select an image he or she does not desire. Then, the selection unit 110 changes the selection conditions (step S60) and returns to step S30. The changes made here are as described with reference to FIG. 1.
[0035] The processes shown in steps S30 to S60 are repeated until the user is satisfied with the selected image (step S50: Yes).
[0036] 5 is a diagram showing an example of a screen displayed on the display unit 130 in step S40. In the example shown in this figure, the display processing unit 120 displays on the display unit 130 3The multiple images selected in step S50 are displayed selectably, and a search button 132 and a confirm button 134 are also displayed. If the user further selects the search button 132 after selecting at least one image, the selection unit 110 determines that the result is Yes in step S50. Also, if the user selects the confirm button 134, the selection unit 110 determines that the result is No in step S50.
[0037] As described above, according to this embodiment, the selection unit 110 of the image selection device 10 selects multiple images according to the query information, and then when a selection input for some of the multiple images is acquired from the user, the selection unit 110 changes the image selection conditions using the images indicated by the selection input and reselects images, thereby improving the accuracy of image selection.
[0038] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various configurations other than those described above can also be adopted.
[0039] In addition, in the flowcharts used in the above description, multiple steps (processes) are described in order, but the order of execution of the steps performed in each embodiment is not limited to the order described. In each embodiment, the order of the steps shown in the figures can be changed to the extent that the content is not affected. In addition, the above-mentioned embodiments can be combined to the extent that the content is not contradictory.
[0040] A part or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes. 1. A selection means for performing a selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions; a display processing means for displaying the plurality of images on a display means; Equipped with The selection means is receiving a selection input indicating a selection of at least one of the images from the plurality of images; An image selection device that changes the selection condition using the image indicated by the selection input, and performs the selection process again using the selection condition after the change. 2. In the image selection device according to the above item 1, After the selection process is performed again, the display processing means causes the display means to display the plurality of images selected by the selection process; Thereafter, the selection means obtains the selection input again, and thereafter changes the selection condition and performs the selection process again. 3. In the image selection device according to the above 1 or 2, The selection condition has a plurality of items, The selection means is In the selection process, a score for each of the plurality of items is calculated, and the plurality of images are selected using the plurality of scores; An image selection device that changes a weighting of each of the plurality of items by using the image indicated by the selection input in changing the selection condition. 4. In the image selection device according to claim 3, the image includes a person; An image selection device, wherein at least some of the plurality of items relate to the relative positions of different parts of a human body with respect to other parts of the human body. 5. In the image selection device according to any one of 1 to 4 above, The query information includes an image. 6. In the image selection device according to any one of 1 to 5 above, The query information includes text information. 7. In the image selection device according to any one of claims 1 to 6, An image selection device in which the images to be selected are moving images. 8. The computer a selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions; a display process for displaying the plurality of images on a display means; Do the following: In the selection process, the computer receiving a selection input indicating a selection of at least one of the images from the plurality of images; The image selection method includes changing the selection condition using the image indicated by the selection input, and performing the selection process again using the changed selection condition. 9. In the image selection method according to claim 8, In the display process, after the selection process is performed again, the computer causes the display means to display the plurality of images selected by the selection process; Thereafter, in the selection process, the computer again obtains the selection input, and then changes the selection conditions and performs the selection process again. 10. In the image selection device according to claim 8 or 9, The selection condition has a plurality of items, In the selection process, the computer calculating a score for each of the plurality of items, and selecting the plurality of images using the plurality of scores; The image selection method includes changing a weighting of each of the plurality of items using the image indicated by the selection input in changing the selection condition. 11. In the image selection method according to claim 10, the image includes a person; A method of image selection, wherein at least some of the plurality of items relate to the relative positions of different parts of a human body relative to other parts of the human body. 12. In the image selection method according to any one of claims 8 to 11, The method of image selection, wherein the query information includes an image. 13. In the image selection method according to any one of claims 8 to 12, The method for selecting images, wherein the query information includes text information. 14. In the image selection method according to any one of claims 8 to 13, The image selection method, wherein the images to be selected are moving images. 15. On the computer, a selection function that performs a selection process of acquiring query information, setting image selection conditions using the query information, and selecting a plurality of images using the selection conditions; a display processing function for displaying the plurality of images on a display means; Let them have The selection function is receiving a selection input indicating a selection of at least one of the images from the plurality of images; A program that changes the selection condition using the image indicated by the selection input, and performs the selection process again using the changed selection condition. 16. In the program according to 15 above, the display processing function displays, after the selection process is performed again, the plurality of images selected by the selection process on the display means; Thereafter, the selection function again obtains the selection input, and then the program changes the selection conditions and performs the selection process again. 17. In the program according to 15 or 16 above, The selection condition has a plurality of items, The selection function is In the selection process, a score for each of the plurality of items is calculated, and the plurality of images are selected using the plurality of scores; A program for changing the weighting of each of the plurality of items using the image indicated by the selection input when changing the selection condition. 18. In the program according to 17 above, the image includes a person; A program wherein at least some of the plurality of items relate to the relative positions of different parts of a human body relative to other parts. 19. In the program according to any one of the above 15 to 18, The query information includes an image. 20. In the program according to any one of the above 15 to 19, The query information includes text information. 21. In the program according to any one of the above 15 to 20, A program in which the image to be selected is a video. [Explanation of symbols]
[0041] 110 Selection section 120 Display processing unit 130 Display section 140 Image storage unit
Claims
1. a selection means for performing a selection process of acquiring query information, setting a selection condition for a person image using the query information, and selecting a plurality of person images using the selection condition; a display processing means for displaying the plurality of person images on a display means; Equipped with the selection condition has a plurality of items including at least one item related to relative positions of different parts of a human body; The selection means is In the selection process, a score for each of the plurality of items is calculated, and the plurality of person images are selected using the plurality of scores; obtaining a selection input indicating a result of selecting at least one of the portrait images from the plurality of portrait images; changing the selection condition using the person image indicated by the selection input, and performing the selection process again using the changed selection condition; In changing the selection condition, a weighting of each of the plurality of items is changed using the person image indicated by the selection input. Image selection device.
2. 2. The image selection device according to claim 1, After the selection process is performed again, the display processing means displays the plurality of person images selected by the selection process on the display means; Thereafter, the selection means obtains the selection input again, and thereafter changes the selection condition and performs the selection process again.
3. 3. The image selection device according to claim 1, The query information includes an image.
4. The image selection device according to any one of claims 1 to 3, The query information includes text information.
5. The image selection device according to any one of claims 1 to 4, The image selection device according to the present invention, wherein the portrait image to be selected is a moving image.
6. The computer a selection process of acquiring query information, setting a selection condition for a person image using the query information, and selecting a plurality of person images using the selection condition; a display process for displaying the plurality of person images on a display means; Do the following: the selection condition has a plurality of items including at least one item related to relative positions of different parts of a human body; In the selection process, the computer calculating a score for each of the plurality of items, and selecting the plurality of person images using the plurality of scores; obtaining a selection input indicating a result of selecting at least one of the portrait images from the plurality of portrait images; changing the selection condition using the person image indicated by the selection input, and performing the selection process again using the changed selection condition; In changing the selection condition, a weighting of each of the plurality of items is changed using the person image indicated by the selection input. Image selection method.
7. On the computer, a selection function that performs a selection process of acquiring query information, setting a selection condition for a person image using the query information, and selecting a plurality of person images using the selection condition; a display processing function for displaying the plurality of person images on a display means; Let them have the selection condition has a plurality of items including at least one item related to relative positions of different parts of a human body; The selection function is In the selection process, a score for each of the plurality of items is calculated, and the plurality of person images are selected using the plurality of scores; obtaining a selection input indicating a result of selecting at least one of the portrait images from the plurality of portrait images; changing the selection condition using the person image indicated by the selection input, and performing the selection process again using the changed selection condition; In changing the selection condition, a weighting of each of the plurality of items is changed using the person image indicated by the selection input. program.
Citation Information
Patent Citations
Multi-media retrieving and providing system, multi-media retrieving and providing method, program for multi- media retrieving and providing system, and recording medium for multi-media retrieving and providing system
JP2003150634A
Method of retrieving similarity by compatibility feedback
JP2003228581A
Image retrieving method and device using repetitive matching
JP2004133947A
Video providing device
JP2007323268A
Exhaust recirculation device for internal combustion engine
JP2010025059A
