Selection frame analysis method and device based on feature image recognition

Abstract

The present invention provides a method and device for parsing a selection box based on feature image recognition, which can be applied to the field of artificial intelligence, including: performing OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image; locating the selection box keyword and intercepting the local image of the area where the selection box is located from the text area; converting the local image into a binary image by a binarization method; and determining the selected selection box according to the number of black pixels in the binary image. The present application extracts the image of the selection box area through the relative spatial position of the keyword in the paper document image, and then completes the analysis of the image selection box area by integrating statistics, image morphological features and image classification counting, and finally improves the accuracy of image recognition of the selection box in the paper document.

Description

Technical Field

[0001] The present application belongs to the field of image recognition technology, and in particular, relates to a selection box parsing method and device based on feature image recognition. Background Art

[0002] In current international cross-regional business transactions, there is no unified electronic document. Therefore, in actual business processing, paper documents are still the main method. For example, when handling business such as opening a letter of credit, customers need to submit paper documents such as the letter of credit application form. In order to improve the efficiency of business processing, it is necessary to input images of these paper documents and perform optical character recognition (OCR) on the images to finally form structured data, so as to realize the automatic input and automatic document review functions of paper documents in transactions. However, the existing OCR technology can only recognize the text in paper documents, and cannot accurately recognize the selection components, because the selection boxes are non-standardized, such as ticking, blacking, etc. When encountering these selection boxes, the existing OCR technology often recognizes them as garbled or cannot recognize any content, which ultimately affects the image parsing effect. Summary of the invention

[0003] The present application provides a selection box parsing method and device based on feature image recognition, so as to at least solve the problem that the current OCR technology recognizes the selection box in the paper document as garbled characters or fails to recognize any content.

[0004] According to one aspect of the present application, a selection box parsing method based on feature image recognition is provided, comprising:

[0005] Performing OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image;

[0006] By locating the selection box keyword, a partial image of the area where the selection box is located is captured from the text area;

[0007] The local image is converted into a binary image by a binarization method;

[0008] The selected selection box is determined according to the number of black pixels in the binary image, or the obtained binary image is input into an image classification model pre-established by using deep learning image classification technology to determine the selected selection box.

[0009] In one embodiment, by locating the selection box keyword, a partial image of the area where the selection box is located is captured from the text area, including:

[0010] Searching for a preset selection box keyword from the text area and obtaining the position coordinates of the selection box keyword in the paper document image;

[0011] Based on the relative position relationship between the selection box keywords, a local image of the area where the selection box is located is captured.

[0012] In one embodiment, converting the local image into a binary image by a binarization method includes:

[0013] Extract a selection box image from the local image and convert the selection box image into a grayscale image;

[0014] The grayscale image is converted into a binary image by using the image grayscale weighted average method.

[0015] In one embodiment, determining the selected selection box according to the number of black pixels in the binary image includes:

[0016] Count the number of black pixels in the binary image;

[0017] Compare the number of black pixels in the selection box image with a preset value, if it exceeds the preset value, the selection box is selected;

[0018] Alternatively, the acquired binary image is input into an image classification model pre-established using deep learning image classification technology to determine the selected selection box.

[0019] According to another aspect of the present application, a selection frame parsing device based on feature image recognition is also provided, comprising:

[0020] An OCR recognition unit is used to perform OCR recognition on the acquired paper document image to obtain a text area in the paper document image and a position of the text area in the paper document image;

[0021] A local image capture unit, used to capture a local image of the area where the selection box is located from the text area by positioning the selection box keyword;

[0022] A binarization unit, used for converting a local image into a binary image by a binarization method;

[0023] The selection judgment unit is used to determine the selected selection box according to the number of black pixels in the binary image.

[0024] In one embodiment, the local image capture unit includes:

[0025] A keyword search module, used to search for a preset selection box keyword from the text area and obtain the position coordinates of the selection box keyword in the paper document image;

[0026] The local image capture module is used to capture the local image of the area where the selection box is located based on the relative position relationship between the selection box keywords.

[0027] In one embodiment, the binarization unit includes:

[0028] A grayscale conversion module is used to extract a selection box image from a local image and convert the selection box image into a grayscale image;

[0029] The binarization module is used to convert the grayscale image into a binary image through the image grayscale weighted average method.

[0030] In one embodiment, the selection and determination unit includes:

[0031] The pixel number counting module is used to count the number of black pixels in the binary image;

[0032] The comparison module is used to compare the number of black pixels in the selection box image with a preset value. If the number exceeds the preset value, the selection box is selected.

[0033] In one embodiment, the device further includes an image classification judgment unit, which is used to input the acquired binary image into an image classification model pre-established using deep learning image classification technology to determine a selected selection box.

[0034] This application extracts the image of the selection box area through the relative spatial position of the keywords in the paper document image, and then completes the analysis of the image selection box area by integrating statistics, image morphological features and image classification counts, ultimately improving the accuracy of image recognition of selection boxes in paper documents. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A flowchart of a selection box parsing method based on feature image recognition provided in this application.

[0037] Figure 2A This is a scanned image of a paper document in an embodiment of the present application.

[0038] Figure 2B It is a schematic diagram of the recognition results in the embodiment of the present application.

[0039] Figure 3 This is a flow chart of a method for capturing a partial image of the area where a selection box is located in an embodiment of the present application.

[0040] Figure 4AThis is a schematic diagram of keyword search in the embodiments of the present application.

[0041] Figure 4B This is a schematic diagram of locking the selection box in an embodiment of the present application.

[0042] Figure 5 This is a flow chart of a method for converting a local image into a binary image in an embodiment of the present application.

[0043] Figure 6 This is a flow chart of determining a selected selection box according to the number of black pixels in a binary image in an embodiment of the present application.

[0044] Figure 7 This is a schematic diagram of converting a grayscale image into a binary image in an embodiment of the present application.

[0045] Figure 8 This is a structural block diagram of a selection box parsing device based on feature image recognition provided in this application.

[0046] Fig. 9 4 is a structural block diagram of a local image capture unit in an embodiment of the present application.

[0047] Fig.10 2 is a structural block diagram of a binarization unit in an embodiment of the present application.

[0048] Fig.11 This is a structural block diagram of a selection judgment unit in an embodiment of the present application.

[0049] Fig.12 This is a specific implementation of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The specific embodiments of the present application can be applied to the field of artificial intelligence, and can also be applied to other technical fields besides the field of artificial intelligence, and the present application is not limited to this.

[0051] In current transactions, there is a paper document review process, which requires image entry and optical character recognition of paper documents to achieve automatic entry and automatic document review in transactions. However, existing OCR technology can only recognize text in paper documents, and cannot accurately recognize selection components, because the selection boxes are non-standardized, such as ticks, blackening, etc. When encountering these selection boxes, existing OCR technology often recognizes them as garbled or cannot recognize any content, which ultimately affects the image analysis effect.

[0052] In order to accurately identify the checkbox in the paper document, the present application provides a method for parsing the selection box based on feature image recognition to solve the above problems, such as Figure 1 As shown, including:

[0053] S101: performing OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image.

[0054] S102: Locate the selection box keyword and capture a partial image of the area where the selection box is located from the text area.

[0055] S103: Convert the local image into a binary image by a binarization method.

[0056] S104: Determine the selected selection box according to the number of black pixels in the binary image.

[0057] The paper document image to be analyzed is positioned and identified by OCR technology to obtain the recognition result text and the position information of the text area corresponding to the recognition result in the image. For the rectangular text area, it can be represented by the left vertex pixel coordinates and the length and width pixel values ​​of the area. The keywords after the selection box are screened in the rectangular text area for positioning, so as to intercept the local image of the selection box in the rectangular text area. Then, the selection box image is converted into a binary image (the binary image is composed of black and white pixels) by a binarization method, and whether the selection box is selected can be known according to the number of black pixels in the binary image. In addition to the above-mentioned determination of whether the selection box is selected by the number of black pixels in the binary image, in another embodiment of the present application, the obtained binary image can also be input into the image classification model pre-established by deep learning image classification technology to determine the selected selection box.

[0058] In a specific embodiment, if Figure 2A As shown, the image text of the paper document and the coordinate position of the text in the image are recognized. The OCR recognition engine can be selected from Tesseract, etc. The recognition results are as follows Figure 2B As shown, it includes the recognized text and the position coordinates of the corresponding area of ​​the text in the image.

[0059] In one embodiment, the selection box is positioned by keyword, and a partial image of the area where the selection box is located is captured from the text area, such as Figure 3 As shown, including:

[0060] S301: searching for a preset selection box keyword in the text area and obtaining the position coordinates of the selection box keyword in the paper document image.

[0061] S302: capturing a local image of the area where the selection box is located based on the relative position relationship between the selection box keywords.

[0062] The title or options of the selection box are searched by keywords, and the location of the selection box or options is obtained. The image of the area where the selection box is located is captured based on the found location information, that is, a partial image of the area where the selection box is located is obtained.

[0063] In a specific embodiment, Figure 4A As shown in the figure, parse the selection components in the figure, find the selection boxes corresponding to the question stems of "(42A) Draw on", "issuing bank" and "other bank (pls indicate)" through keywords, and then analyze the positions of the three keywords in the impact of paper documents, and get the following Figure 4B The coordinates of the eight points A, B, C, D, E, F, G, and H shown are used to intercept the two rectangular images ABCD and EFGH respectively, and the selection box diagrams corresponding to the two options "issuing bank" and "other bank (plsindicate)" are obtained.

[0064] In one embodiment, the local image is converted into a binary image by a binarization method, such as Figure 5 As shown, including:

[0065] S401: extracting a selection box image from a local image and converting the selection box image into a grayscale image.

[0066] S402: Convert the grayscale image into a binary image by using the image grayscale weighted average method.

[0067] The local image of the area where the selected box is located is first converted into a grayscale image. The weighted average method of image grayscale can be used. The formula is as follows, where R, G, and B represent the three components in the color image:

[0068] Gray=0.114B+0.587G+0.299R

[0069] In one embodiment, the selected selection box is determined according to the number of black pixels in the binary image, such as Figure 6 As shown, including:

[0070] S501: Count the number of black pixels in the binary image.

[0071] S502: Compare the number of black pixels in the selection box image with a preset value. If the number exceeds the preset value, the selection box is selected.

[0072] In a specific embodiment, after the grayscale image is converted into a binary image, there are only two kinds of pixels, pure white and pure black, in the binary image. Binarization techniques such as simple threshold, adaptive threshold and Otsu's binarization can be used to convert the grayscale image into a binary image, such as Figure 7 As shown. Count the number of black pixels after binarization. The number of black pixels in the selected selection box should be significantly larger than that in the unselected box. For single-selection type selection components, the selected selection box can be found by comparing the number of black pixels in the selection box to complete the analysis. Alternatively, a reasonable threshold can be set. If the number of black pixels after binarization obviously exceeds the threshold, then the selection box is considered to have been selected.

[0073] In another specific embodiment of the present application, the acquired binary image may also be input into an image classification model pre-established by using deep learning image classification technology to determine the selected selection box, as follows:

[0074] Use the collected selection boxes of different styles to train an image classification model. You can use deep learning image classification technologies such as VGG and TesNet. The model input is an image, and the output is whether the image is a selected box. The model generation unit is executed once at the beginning of the system, and is executed in a timely manner as training samples are accumulated during the operation of the system.

[0075] The output binary image is sent to the image classification model to determine whether the image is a selected selection box image, thereby completing the analysis of the selection box image in the paper document.

[0076] Based on the same inventive concept, the embodiment of the present application also provides a selection box parsing device based on feature image recognition, which can be used to implement the method described in the above embodiment, as described in the following embodiment. Since the principle of solving the problem by the selection box parsing device based on feature image recognition is similar to that of the selection box parsing method based on feature image recognition, the implementation of the selection box parsing device based on feature image recognition can refer to the implementation of the selection box parsing method based on feature image recognition, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements predetermined functions. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0077] According to another aspect of the present application, a selection frame parsing device based on feature image recognition is also provided. Figure 8 As shown, including:

[0078] The OCR recognition unit 601 is used to perform OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image;

[0079] The local image capture unit 602 is used to capture the local image of the area where the selection box is located from the text area by positioning the selection box keyword;

[0080] A binarization unit 603, used to convert the local image into a binary image by a binarization method;

[0081] The selection determination unit 604 is used to determine the selected selection box according to the number of black pixels in the binary image.

[0082] In one embodiment, if Fig. 9 As shown, the local image interception unit 602 includes:

[0083] The keyword search module 701 is used to search for a preset selection box keyword in the text area and obtain the position coordinates of the selection box keyword in the paper document image;

[0084] The local image capture module 702 is used to capture the local image of the area where the selection box is located according to the relative position relationship between the selection box keywords.

[0085] In one embodiment, if Fig.10 As shown, the binarization unit 603 includes:

[0086] A grayscale conversion module 801 is used to extract a selection frame image from a local image and convert the selection frame image into a grayscale image;

[0087] The binarization module 802 is used to convert the grayscale image into a binary image by using the image grayscale weighted average method.

[0088] In one embodiment, if Fig.11 As shown, the selection and judgment unit 604 includes:

[0089] A pixel number counting module 901 is used to count the number of black pixels in the binary image;

[0090] The comparison module 902 is used to compare the number of black pixels in the selection box image with a preset value. If the number exceeds the preset value, the selection box is selected.

[0091] In one embodiment, the device further includes an image classification judgment unit, which is used to input the acquired binary image into an image classification model pre-established using deep learning image classification technology to determine a selected selection box.

[0092] In a specific embodiment, the processing flow method provided by the present application comprises the following steps:

[0093] Step 101: Initialize the application system;

[0094] Step 102: executing the image classification model generation unit 32 of the selection state recognition device 3 to generate an initial image classification model;

[0095] Step 103: The application system starts to start;

[0096] Step 104: Acquire the image to be processed;

[0097] Step 105: Perform OCR positioning recognition on the image to obtain recognition content and location information;

[0098] Step 106: intercepting the selection frame diagram through keyword search positioning and relative position analysis;

[0099] Step 107: binarizing the intercepted selection box diagram to obtain a binarized selection box diagram;

[0100] Step 108: Determine whether the selection field to be parsed is single-select or multiple-select;

[0101] Step 110: For a single-selection type selection domain, whether it is selected can be determined by comparing the number of black pixels;

[0102] Step 111: For single-select and multiple-select selection fields, an image classification model can be used to determine whether a selection is made;

[0103] Step 112: Integrate the judgment results of step 110 and step 111. In the early stage when the image classification model is not perfect, only the processing of step 110 can be used as the main method, and sufficient training samples can be accumulated by only analyzing the selection domain of the single-choice type. After the image classification model is perfected, the judgment results of the image classification model can be used as the main method.

[0104] The embodiments of the present application also provide a specific implementation of an electronic device that can implement all the steps in the method in the above embodiments, see Fig.12 , the electronic device specifically includes the following contents:

[0105] Processor 1101, memory 1102, communication interface 1103, bus 1104 and non-volatile memory 1105;

[0106] The processor 1101, memory 1102, and communication interface 1103 communicate with each other via the bus 1104;

[0107] The processor 1101 is used to call the computer program in the memory 1102 and the non-volatile memory 1105. When the processor executes the computer program, all the steps in the method in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0108] S101: performing OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image.

[0109] S102: Locate the selection box keyword and capture a partial image of the area where the selection box is located from the text area.

[0110] S103: Convert the local image into a binary image by a binarization method.

[0111] S104: Determine the selected selection box according to the number of black pixels in the binary image.

[0112] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the method in the above embodiments. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all the steps of the method in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0113] S101: performing OCR recognition on the acquired paper document image to obtain the text area in the paper document image and the position of the text area in the paper document image.

[0114] S102: Locate the selection box keyword and capture a partial image of the area where the selection box is located from the text area.

[0115] S103: Convert the local image into a binary image by a binarization method.

[0116] S104: Determine the selected selection box according to the number of black pixels in the binary image.

[0117] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the hardware + program embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. Although the embodiment of this specification provides the method operation steps described in the embodiment or flow chart, more or less operation steps can be included based on conventional or non-creative means. The order of steps listed in the embodiment is only one way of executing the order of many steps, and does not represent the only execution order. When the device or terminal product in practice is executed, it can be executed in sequence or in parallel according to the method shown in the embodiment or the accompanying drawings (for example, a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment). The term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, product or device. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. For the convenience of description, the above device is described by dividing it into various modules according to its functions. Of course, when implementing the embodiments of this specification, the functions of each module can be implemented in the same one or more software and / or hardware, or the module implementing the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The present invention is described with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the process and / or box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process Figure 1 A process or multiple processes and / or boxes Figure 1A device for specifying functions in a box or multiple boxes. It should be understood by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts refer to the partial description of the method embodiment. In the description of this specification, the description of the reference term "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction. The above description is only an embodiment of the embodiment of this specification and is not intended to limit the embodiment of this specification. For those skilled in the art, the embodiment of this specification may have various changes and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the embodiment of this specification shall be included in the scope of the claims of the embodiment of this specification.

Claims

1. A selection box parsing method based on feature image recognition, characterized in that: include: Performing OCR recognition on the acquired paper document image to obtain a rectangular text area in the paper document image and the position of the rectangular text area in the paper document image; The rectangular text area is represented by the pixel coordinates of the left vertex and the length and width of the area in pixels; The keyword behind the selection box is selected in the rectangular text area for positioning, and a partial image of the area where the selection box is located is captured from the rectangular text area, including: Searching for a preset selection box keyword from the rectangular text area and obtaining the position coordinates of the selection box keyword in the paper document image; According to the relative position relationship between the keywords in the selection box, a local image of the area where the selection box is located is captured; Converting the local image into a binary image by a binarization method; The selected selection box is determined according to the number of black pixels in the binary image.

2. The selection box parsing method based on feature image recognition according to claim 1, characterized in that: The step of converting the local image into a binary image by a binarization method comprises: Extracting a selection frame image from the partial image and converting the selection frame image into a grayscale image; The grayscale image is converted into a binary image by using an image grayscale weighted average method.

3. The selection box parsing method based on feature image recognition according to claim 2, characterized in that: The step of determining the selected selection box according to the number of black pixels in the binary image includes: Counting the number of black pixels in the binary image; The number of black pixels in the selection frame image is compared with a preset value, and if the number exceeds the preset value, the selection frame is selected.

4. The selection box parsing method based on feature image recognition according to claim 3 is characterized in that: Also includes: The acquired binary image is input into an image classification model pre-established using deep learning image classification technology to determine the selected selection box.

5. A selection frame analysis device based on feature image recognition, characterized in that: include: The OCR recognition unit is used to perform OCR recognition on the acquired paper document image to obtain a rectangular text area in the paper document image and the position of the rectangular text area in the paper document image; the rectangular text area is represented by the left vertex pixel coordinates and the length and width pixel values ​​of the area; A local image capture unit, used for screening the keywords behind the selection box in the rectangular text area for positioning, and capturing a local image of the area where the selection box is located from the rectangular text area; The local image interception unit comprises: A keyword search module, used to search for a preset selection box keyword from the rectangular text area and obtain the position coordinates of the selection box keyword in the paper document image; A local image capture module, used to capture a local image of the area where the selection box is located according to the relative position relationship between the selection box keywords; A binarization unit, used for converting the local image into a binary image by a binarization method; The selection judgment unit is used to determine the selected selection box according to the number of black pixels in the binary image.

6. The selection frame analysis device based on feature image recognition according to claim 5, characterized in that: The binarization unit comprises: A grayscale conversion module, used for extracting a selection frame image from the local image and converting the selection frame image into a grayscale image; The binarization module is used to convert the grayscale image into a binary image by using an image grayscale weighted average method.

7. The selection frame analysis device based on feature image recognition according to claim 6, characterized in that: The selection and judgment unit comprises: A pixel number counting module, used for counting the number of black pixels in the binary image; The comparison module is used to compare the number of black pixels in the selection box image with a preset value. If the number exceeds the preset value, the selection box is selected.

8. The selection frame analysis device based on feature image recognition according to claim 7, characterized in that: Also includes: The image classification judgment unit is used to input the acquired binary image into an image classification model pre-established by using deep learning image classification technology to determine the selected selection box.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the selection box analysis method based on feature image recognition according to any one of claims 1 to 4 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the selection frame analysis method based on feature image recognition described in any one of claims 1 to 4 are implemented.

Images