Information processing device, information processing method, and information processing program

Abstract

Provided is an information processing device comprising: a feature / item extraction unit that extracts features / items, which are features and items included in file data to be retrieved; and a search condition generation unit that generates search conditions on the basis of the extracted features / items. Also provided is a program for causing the information processing device to execute: a feature / item extraction step for extracting features / items, which are features and items included in file data to be retrieved; and a search condition generation step for generating retrieval conditions on the basis of the extracted features / items.

Description

Information Processing Apparatus, Information Processing Method, and Information Processing Program

[0006]

[0001] The present invention relates to an information processing apparatus that searches for predetermined file data.

[0002] In recent years, the use of large language models (hereinafter sometimes abbreviated as LLMs) has been spreading. LLMs can execute various natural language processing tasks such as text classification, information extraction, image search, text summarization, text generation, and question answering. For example, Japanese Unexamined Patent Application Publication No. 2018-165926 discloses a similar image search apparatus that searches for an image similar to an input image by inputting the input image into a learned model.

[0003] In the case of image search like the technology of Patent Document 1, when the search result is not what the user intended, it is impossible to modify the search conditions so that the intended result can be obtained. This can be said to be because the search conditions are unclear. Also, since similarity is determined only by appearance, a search based on the characteristics of the object shown in the image is not performed. The prior art mentioned above has such problems.

[0004] Therefore, in order to solve the above problems, the present invention provides the following information processing apparatus and the like. That is, an information processing apparatus having a feature item extraction unit that extracts feature items that are features and items included in the search target file data, and a search condition generation unit that generates search conditions based on the extracted feature items is provided.

[0005] Also, an information processing apparatus further having a search condition setting unit that sets the generated search conditions on a UI screen in addition to the above features is provided.

[0006] Also, an information processing apparatus further having a corresponding location display unit that displays the corresponding location in the search target file data for the set search conditions in addition to the above features is provided.

[0007] Also, an information processing apparatus further having a cursor movement unit that moves a cursor to a region on the UI screen where the search conditions corresponding to the corresponding location are set by an instruction operation on the corresponding location to be displayed in addition to the above features is provided.

[0008] Furthermore, the present invention provides an information processing method that, in addition to the features described above, further includes a prompt setting unit that sets a prompt to be input to the trained AI model along with the search target file data to instruct the extraction of the feature items.

[0009] In addition to the features described above, the prompt setting unit provides an information processing device that sets a prompt instructing the generation of the search conditions.

[0010] Furthermore, in addition to the above features, the present invention provides an information processing device that further includes a search unit that performs a search based on the generated search conditions.

[0011] In addition to the features described above, the system further includes a database generation unit that generates a database by structuring the file data group by extracting the feature items from the file data group that will be used as the search source, and the search unit provides an information processing device that searches the database generated by the database generation unit.

[0012] In addition to the features described above, the search condition generation unit provides an information processing device that generates search conditions based on the feature items included in the database generated by the database generation unit.

[0013] In addition to the features described above, the feature item extraction unit provides an information processing device that extracts the feature items contained in the search target file data based on the feature items contained in the database that serves as the search source.

[0014] In addition to the features described above, the search condition generation unit provides an information processing device that generates search conditions based on the feature items included in the database that serves as the search source.

[0015] In addition to the features described above, the feature item extraction unit provides an information processing device that extracts the feature items included in the search target file data using the extraction method used by the database generation unit to extract the feature items from the file data group.

[0016] In addition to the features described above, the database generation unit provides an information processing device that structures the file data group based on the feature items extracted by the feature item extraction unit and generates a database.

[0017] In addition to the features described above, the feature item extraction unit provides an information processing device that extracts the feature items by inputting the search target file data into a trained AI model.

[0018] Furthermore, the present invention provides an information processing method executed by an information processing device, comprising: a feature item extraction step for extracting feature items that are characteristics or items contained in the file data to be searched; and a search condition generation step for generating search conditions based on the extracted feature items.

[0019] Furthermore, the present invention provides a program that causes an information processing device to execute a feature item extraction step, which extracts feature items that are characteristics or items contained in the file data to be searched, and a search condition generation step, which generates search conditions based on the extracted feature items.

[0020] This invention makes it possible to modify the search criteria. Furthermore, it ensures the accuracy of the search criteria. Additionally, it allows searching for images that have the same characteristics, even if they look different.

[0021] Block diagram showing an example of the functional configuration of the information processing device of Embodiment 1 Conceptual diagram showing an example of displaying image data, which is the search target file data, and search conditions on the UI screen Conceptual diagram for explaining the relevant location display unit Conceptual diagram for explaining the cursor movement unit Conceptual diagram showing an example of the hardware configuration for realizing the information processing device of Embodiment 1 Flow diagram showing a simplified example of the processing flow of the information processing device of Embodiment 1 Block diagram showing an example of the functional configuration of the information processing device of Embodiment 2 Conceptual diagram illustrating a prompt that indirectly instructs the extraction of feature items

[0022] Embodiments of the present invention will be described below with reference to the accompanying drawings. However, the present invention is not limited in any way to these embodiments, and can be implemented in various ways without departing from its essence.

[0023] <Example 1> <Summary> The present invention is characterized by extracting features and items of the data to be searched when performing searches on various data, including image data, and generating individual search conditions corresponding to the extracted features and items.

[0024] The following describes the functions and processing flow of the information processing device, as well as the hardware components. The functional blocks of this system described below can be implemented as a combination of hardware and software. Specifically, if a computer is used, these may include hardware components such as a CPU (Central Processing Unit), main memory, a bus, or secondary storage devices (hard disk drives, non-volatile memory, storage media such as CDs and DVDs, and their readers), input devices used for information input, printing equipment, display devices, and other external peripheral devices, as well as interfaces for these external peripheral devices, communication interfaces, driver programs and other application programs for controlling the hardware, and user interface applications. The CPU's arithmetic processing, based on programs deployed in main memory, processes and stores data input from input devices and other interfaces, and stored in memory and on the hard disk, and generates instructions for controlling the aforementioned hardware and software. Alternatively, the functional blocks of this system may be implemented using dedicated hardware.

[0025] Furthermore, this invention can be realized not only as a system but also as a method. Moreover, a part of such an invention can be configured as software. In addition, programs used to cause a computer to execute such software, and recording media on which such programs are fixed, are naturally included within the technical scope of this invention (as is the case throughout this specification).

[0026] <Functional Configuration> Figure 1 is a block diagram showing an example of the functional configuration of the information processing device of this embodiment. As shown in Figure 1, the information processing device 100 includes a feature item extraction unit 101, a search condition generation unit 102, a search condition setting unit 103, a corresponding location display unit 104, and a cursor movement unit 105.

[0027] In this embodiment, the trained AI model is a machine learning model that has been pre-trained and is capable of handling general-purpose tasks. A typical example is a large-scale language model, but it may also be a small-scale language model or a multimodal language model that can handle images or other formats.

[0028] <Feature Item Extraction Unit> The feature item extraction unit 101 has the function of extracting feature items that are features or items contained in the search target file data. The search target file data is the file data that will be searched (query), and includes, for example, Word files, PDF (Portable Document Format) files, Excel files, image files, CSV (Comma Separated Values) files, text files, BIM (Building Information Modeling) files, 2D CAD (Computer Aided Design) files, 3D model files, etc.

[0029] Furthermore, feature items indicate properties or attributes that distinguish what is expressed or described in the file data from other things. The feature items to be extracted include an item name, which is a word that abstractly expresses those properties or attributes, and a meaning that includes its specific content. For example, in image data that depicts "a white two-story house with a full moon in the background," if we represent the feature items as "item name = its content," then feature items such as "background = full moon night sky," "house color = white," and "house structure = two-story house" may be extracted. Alternatively, instead of "background = full moon night sky," feature items such as "background = night sky" and "background = moon" may be extracted. The number of feature items is arbitrary, and a sufficient number of feature items to distinguish what is expressed in the target file data from other things will be extracted. Also, feature items may be cropped parts of the image, or they may be vector information contained in the searched file data.

[0030] Feature item extraction is primarily performed by feeding the target file data to a trained AI model, but it may also be performed using extraction algorithms or image matching techniques instead. Furthermore, feature item extraction may be performed by receiving an instruction to select the target file data, triggered by uploading new target file data, or triggered by selecting any of the already uploaded file data. In addition to selecting the target file data, feature item extraction may also be performed by accepting predetermined operations from the user.

[0031] Furthermore, a portion of the file data may be extracted and used as the target file for the search. This extraction process can be performed by a pre-trained AI model, or by executing an extraction script generated by the pre-trained AI model. The extracted feature items may be used for output such as display, or they may be configured to be used for the search condition generation unit described later without being displayed.

[0032] <Search Condition Generation Unit> The search condition generation unit 102 has the function of generating search conditions based on the extracted feature items. Search conditions are the conditions specified when performing a search. Search conditions based on feature items are, for example, when searching for files similar to the target file data, the feature items extracted from the target file data become the search conditions generated based on the extracted items.

[0033] In the example above, if the target file data is image data representing "a white two-story house against a night sky with a full moon," then the search conditions for searching for similar image data are generated using the feature items extracted from the target file data: "background = night sky with a full moon," "house color = white," and "house structure = two-story house."

[0034] <Search Condition Setting Unit> The search condition setting unit 103 has the function of setting the generated search conditions on the UI screen.Specific examples will be described later, but setting search conditions means accepting user modifications to the generated search conditions, accepting operations for the user to confirm and decide on the generated search conditions, and accepting operations for instructing the regeneration of search conditions.

[0035] Figure 2 is a conceptual diagram showing an example of displaying image data, which is the file data to be searched, and search conditions on the UI screen. As shown in Figure 2(a), the UI (User Interface) screen 201 displays image data 202 of a photograph of a room as the file data to be searched. Below the image data 202, a confirmation button 203 is displayed.

[0036] Here, by accepting an operation such as a click on the confirmation button 203, the displayed image data is determined to be the search target file data, and the feature items of this image data are generated by the feature item generation unit. Then, as shown in Figure 2(b), the generated feature items are displayed in the search condition field 204 on the left side of the UI screen 201. Here, the feature items "Content = Interior", "Location = Living Room", "Material = Wood", and "Furniture = Sofa" are displayed as search conditions.

[0037] Furthermore, by accepting operations such as clicking the edit button 205, the displayed search conditions can be modified. Modifications can be made by the user themselves by modifying the content of the displayed search conditions (the content of each item such as "indoors" or "living room"), or by instructing the user to add more search conditions, which will then generate new feature items and search conditions based on them. Then, by accepting operations such as clicking the search button 206, the search process is performed based on the set search conditions (search unit).

[0038] Additionally, the system may allow users to delete generated search criteria. For example, if 10 search criteria are generated, users can delete 3 of them, leaving 7.

[0039] In this way, the feature items included in the search target file data are displayed on the UI screen as search conditions, allowing users to modify the generated search conditions. In particular, as shown in Figure 2(b), displaying the search conditions generated from the search target file data along with the search conditions on the UI screen makes it clear how the search conditions are reflected from the search target file data, making it easier to modify the search conditions.

[0040] <Relevant Location Display Unit> The relevant location display unit 104 has the function of displaying the relevant location in the search target file data for the set search conditions. This clarifies the correspondence between each search condition and the relevant location in the search instruction.

[0041] FIG. 3 is a conceptual diagram for explaining the relevant location display part. As shown in FIG. 3(a), the periphery 302 of the image data which is the search target file data on the UI screen 301 is highlighted, and by surrounding the search conditions of "indoor" and "living room" with a frame 303, it becomes clear that the search conditions of "content = indoor" and "location = living room" are generated from the entire image data. That is, the correspondence between the search conditions and the relevant locations in the image data is shown. Note that the relevant location display is not limited to the display in such a manner that the correspondence with the search conditions is made clear. In the search target file data, it is only necessary to display the location used to generate the search conditions.

[0042] Also, in FIG. 3(b), the "sofa" in the search conditions is surrounded by a frame 304 and displayed, and the sofa shown in the image data 302 of the search target file data is indicated by an arrow 305. Thereby, the search condition of "furniture = sofa" and the relevant location shown in the image data become clear. Such a display of the relevant location is not limited to the above example, and variously emphasized displays or displays of text explaining the relevant location may be used.

[0043] Also, it can be configured such that by indicating any of the search conditions with a cursor or the like, the relevant location in the search target file data is highlighted. Furthermore, it may be configured such that the relevant location is highlighted by indicating the search conditions by voice input.

[0044] <Cursor movement part> The cursor movement part 105 has a function of moving the cursor to the area on the UI screen where the search condition corresponding to the relevant location is set by an instruction operation on the displayed relevant location.

[0045] Figure 4 is a conceptual diagram illustrating the cursor movement section, following Figure 3(b). As shown, the UI screen 401 displays image data, which is the file data to be searched, and the sofa shown therein is indicated by the arrow 402. When the finger cursor 403 for operation is pointing to the sofa, performing an operation such as a left click to select or execute will cause a character cursor to appear immediately after the word in the sofa in the search condition, resulting in an active input state 404. Alternatively, performing an operation on the finger cursor 403 may cause this finger cursor to move to the area where the sofa in the search condition is displayed and be displayed there.

[0046] By configuring the system to allow this type of cursor movement, it becomes easy and quick to switch between the search conditions and the corresponding locations in the target file data used to generate those search conditions, allowing for rapid and reliable review and modification of the search conditions.

[0047] <Hardware Configuration> Figure 5 is a conceptual diagram showing an example of the hardware configuration for realizing the information processing device according to Embodiment 1. As shown in the figure, the information processing device 500 includes a CPU 501 for performing various calculations, a RAM 502 which is a volatile recording medium, a storage device 503 such as a flash memory or HDD which is a non-volatile storage medium, a communication interface 504, and an input / output interface 505. The RAM 502 reads programs that perform various calculations for the CPU 501 to execute and provides a work area (work area) for those programs. In addition, multiple addresses are assigned to the RAM 502, and programs executed by the CPU 501 can exchange data with each other and perform processing by identifying and accessing these addresses (this is the same throughout this specification).

[0048] Here, each function of the feature item extraction unit 101 and the search condition generation unit 102 of the information processing apparatus 100 in FIG. 1 is mainly realized by the CPU 501 and the RAM 502 in FIG. 5. Further, the functions of the search condition setting unit 103, the corresponding location display unit 104, and the cursor movement unit 105 are mainly realized by the CPU 501, the RAM 502, and the input / output interface 505 in FIG. 5. Further, when the feature item extraction unit 101 or the search condition generation unit 102 uses a pre-trained AI model existing externally, each function is realized by mutually exchanging signals and information via the communication interface 504 and the input / output interface 505.

[0049] <Process flow>FIG. 6 is a flowchart briefly showing an example of the process flow of the information processing apparatus in the first embodiment. First, feature items are extracted as features and items included in the search target file data. (S601: Feature item extraction step). Then, the set search instruction is input to the pre-trained AI model to generate corresponding individual search conditions (S602: Search condition generation step). Then, the generated search conditions are set on the UI screen (S603: Search condition setting step). Then, for the set search conditions, the corresponding locations in the search target file data are displayed (S604: Corresponding location display step). Then, by an instruction operation on the displayed corresponding location, the cursor is moved to the area on the UI screen where the search conditions corresponding to the corresponding location are set (S605: Cursor movement step).

[0050] <Effect>According to the information processing apparatus of the present embodiment, a search focusing on the features included in the search target file data can be performed.

[0051] <Second embodiment>This embodiment is an information processing apparatus or the like in which functions are further added based on the first embodiment.

[0052] <Functional Configuration> Figure 7 is a block diagram showing an example of the functional configuration of the information processing device of Embodiment 2. As shown in Figure 7, the information processing device 700 includes a feature item extraction unit 701, a search condition generation unit 702, a search condition setting unit 703, a corresponding location display unit 704, a cursor movement unit 705, a prompt setting unit 706, a search unit 707, and a database generation unit 708. The feature item extraction unit 701, the search condition generation unit 702, the search condition setting unit 703, the corresponding location display unit 704, and the cursor movement unit 705 are the same as those in Embodiment 1, so their explanation is omitted here.

[0053] <Prompt Setting Unit> The prompt setting unit 706 has the function of setting prompts that are input to the trained AI model along with the search target file data to instruct the extraction of the feature items.

[0054] This prompt is an instruction given to a trained AI model. This prompt may directly instruct the model to extract feature items from the target file data, or it may indirectly instruct the model to extract feature items through a prompt that instructs the model to perform a search based on the target file data. For example, an example of the former type of prompt would be, "Please extract the features from the image data I am about to provide."

[0055] Figure 8 is a conceptual diagram illustrating a prompt that indirectly instructs the extraction of feature items. As shown in the figure, on the UI screen 801, a field 803 for setting the prompt is displayed in the area to the right of the image data 802, which is the search target file data and shows a room. The prompt entered in this field is, "Please search for a photograph of a dining room using the same flooring material as this image." Although this prompt directly instructs a search, it also indirectly instructs feature extraction because it extracts the feature item "flooring material" contained in the image data, which is the search target file data, as a prerequisite for the search.

[0056] Furthermore, such prompts may be pre-configured. For example, various sample prompts for extraction and searching can be pre-configured, and the user can select one of them and then process or edit the selected sample prompt. In this way, the prompt setting unit has the function of accepting prompt input, selection, and editing by the user, as well as pre-configured prompts.

[0057] Furthermore, the prompt setting unit can be configured to set prompts that instruct the generation of search conditions. For example, there may be a prompt that instructs the user to add feature items that are not included in the search conditions when the generated feature items are displayed. More specifically, a prompt such as "In addition to the generated feature items, please also add 'floor color = yellow' to the search conditions."

[0058] <Search Unit> The search unit 707 has the function of performing a search based on the generated search conditions. For example, a search is performed by inputting the target file data, the search conditions generated based on this data, and a prompt to instruct the search into the trained AI model. This makes it possible to perform a multifaceted search by combining individual search conditions. It is also possible to perform a search based on search conditions set by the search condition setting unit. For example, as shown in Figure 2, an image search is performed based on the set search conditions.

[0059] Furthermore, the search source to be searched may be a group of file data stored by this information processing device, or it may be an external database. Alternatively, the system may be configured to search a database generated by the database generation unit described later. It is also possible to configure the system to directly provide the search unit with the search conditions generated by the search condition generation unit, without requiring the search conditions to be set in the search condition setting unit.

[0060] Furthermore, the prompt setting section described above can be configured to accept user input to set search conditions and prompts that instruct the search, and then perform a search by providing these prompts to a trained AI model.

[0061] Since searches can be performed using search conditions based on feature items extracted from the target file data, especially when searching image data, it is possible to perform searches that focus on feature items, such as "search for bags of the same color as these shoes" or "search for rooms that have the same desk as the desk in this room."

[0062] <Database Generation Unit> The database generation unit 708 has the function of structuring the file data group 709, which is the source of the search, by extracting the characteristic items from the file data group 709, and generating a database 710.

[0063] The source file data set is a group of files that are searched, for example, similar images, based on the target file data. For each file data that makes up this file data set, feature items contained in the file data are extracted as described in the feature item extraction unit of Example 1. Then, the extracted items and their contents are structured as "columns and rows" or "JSON format" to generate a database. By structuring in this way, searches based on the target file data can be performed efficiently and with good accuracy.

[0064] Furthermore, the database generation unit can be configured to generate a database by structuring the file data group based on the feature items extracted by the feature item extraction unit.

[0065] In the search unit described above, by using the database generated by the database generation unit as the search source, it becomes possible to perform searches that focus on specific features. For example, conventional similar image searches could only search based on visual similarity with the target image. However, by structuring the search using feature items, it becomes possible to perform searches based on commonalities in features even if the target image looks different.

[0066] Furthermore, since it is possible to set search conditions based on characteristic items, by changing some of these conditions, it is possible to search for items that are only different in color or those that have been replaced with other materials, enabling flexible and accurate searches.

[0067] Furthermore, the search condition generation unit described in Example 1 can be configured to generate search conditions based on feature items included in the database generated by the database generation unit. For example, feature items extracted from the target file data that are common with the feature items included in the database can be used as search conditions. Alternatively, the unit may be configured to refer to the feature items included in the database when extracting feature items from the target file data, such as extracting as many feature items included in the database as possible.

[0068] Furthermore, when using an existing external database as the search source without the involvement of the database generation unit, the search condition generation unit can be configured to generate search conditions based on the feature items contained in the database being searched. Since existing databases are already structured, feature items can be extracted from them. Then, as explained in the case where a database generated by the database unit is used as the search source, search conditions are generated based on the extracted feature items.

[0069] Furthermore, if a database generated by the database generation unit already exists, the feature item extraction unit can be configured to extract feature items included in the search target file data using the extraction method that the database generation unit used to extract feature items from the file data group. This makes it easier for the feature items of the search target and the search source to match, which can contribute to improved search accuracy.

[0070] Furthermore, the feature item extraction unit can be configured to extract the feature items contained in the search target file data based on the feature items contained in the database that serves as the search source. For example, feature items similar to those contained in an existing database can be extracted from the search target file data.

[0071] <Hardware Configuration> The information processing device of Example 2 can be realized with the hardware configuration shown in Figure 5, in accordance with Example 1. The function of the prompt setting unit 706, which is a configuration unique to the information processing device of Example 2 shown in Figure 7, is mainly realized by the CPU 501, RAM 502, and input / output interface 505 shown in Figure 5. In addition, the functions of the search unit 707 and the database generation unit 708 are mainly realized by the CPU 501 and RAM 502 shown in Figure 5.

[0072] <Processing Flow> The processing flow of the information processing device in Example 2 is basically the same as that of the information processing device in Example 1. Furthermore, the information processing device in Example 2 includes a prompt setting step in which prompts are set to be input to a trained AI model along with the file data to be searched to instruct the extraction of feature items; a search step in which a search is performed based on the generated search conditions; and a database generation step in which a database is generated by structuring the file data group by extracting feature items from the file data group that will be the source of the search.

[0073] <Effects> According to the information processing device of this embodiment, it is possible to perform a search that focuses on the features contained in the file data to be searched.

[0074] 100, 500, 700: Information processing unit 101, 701: Feature item extraction unit 102, 702: Search condition generation unit 103, 703: Search condition setting unit 104, 704: Corresponding location display unit 105, 705: Cursor movement unit 706: Prompt setting unit 707: Search unit 708: Database generation unit 709: File data group 710: Database 501: CPU 502: RAM 503: Storage 504: Communication interface 505: Input / output interface

Claims

1. An information processing device comprising: a feature item extraction unit that extracts feature items that are characteristics or items contained in the file data to be searched; and a search condition generation unit that generates search conditions based on the extracted feature items.

2. The information processing apparatus according to claim 1, further comprising a search condition setting unit for setting the generated search conditions on a UI screen.

3. The information processing apparatus according to claim 2, further comprising a corresponding location display unit that displays the corresponding location in the search target file data for the generated search conditions.

4. The information processing apparatus according to claim 3, further comprising a cursor movement unit that moves the cursor to an area on the UI screen where search conditions corresponding to the displayed area are set, based on an instruction operation on the displayed area.

5. The information processing apparatus according to claim 2, further comprising a prompt setting unit that sets a prompt to be input to the trained AI model along with the search target file data to instruct the extraction of the feature items.

6. The information processing apparatus according to claim 5, wherein the prompt setting unit sets a prompt that instructs the generation of the search conditions.

7. The information processing apparatus according to claim 1 or 2, further comprising a search unit that performs a search based on the set search conditions.

8. The information processing apparatus according to claim 7, further comprising a database generation unit that generates a database by structuring a group of file data to be searched by extracting the characteristic items from the group of file data, wherein the search unit searches the database generated by the database generation unit.

9. The information processing apparatus according to claim 8, wherein the search condition generation unit generates search conditions based on the feature items included in the database generated by the database generation unit.

10. The information processing apparatus according to claim 1 or 2, wherein the feature item extraction unit extracts the feature items contained in the search target file data based on the feature items contained in the database that serves as the search source.

11. The information processing apparatus according to claim 1 or 2, wherein the search condition generation unit generates search conditions based on the feature items included in the database that serves as the source of the search.

12. The information processing apparatus according to claim 8, wherein the feature item extraction unit extracts the feature items included in the search target file data by an extraction method used by the database generation unit to extract the feature items from the file data group.

13. The information processing apparatus according to claim 8, wherein the database generation unit structures the file data group based on the feature items extracted by the feature item extraction unit and generates a database.

14. The information processing apparatus according to claim 1 or 2, wherein the feature item extraction unit inputs the search target file data into a trained AI model to extract the feature items.

15. An information processing method executed by an information processing device, comprising: a feature item extraction step of extracting feature items that are features or items contained in the file data to be searched; and a search condition generation step of generating search conditions based on the extracted feature items.

16. A program that causes an information processing device to execute a feature item extraction step for extracting feature items that are characteristics or items contained in the file data to be searched, and a search condition generation step for generating search conditions based on the extracted feature items.

Images