Image data extraction method and system and medium

Abstract

The invention discloses an image data extraction method and system and a medium, and the method comprises the steps: dividing image data, dividing the image data into a plurality of transverse regions in the column direction, and enabling each transverse region to be composed of a plurality of rows of pixels; counting a transverse area with the largest number of covering line center points in each column direction so as to confirm a line center point concentrated area of each column; and judging whether the frequency of the line center point concentration area in each transverse area is greater than a preset threshold, if so, outputting a line center point close to a preset reference position of the corresponding transverse area in each column direction, and if not, judging whether the line center point concentration area of each column is continuous. According to the method, parallel processing can be rapidly carried out on the multiple line center points extracted from each column, the extraction efficiency is effectively improved, continuous features of the laser rays can be effectively combined, features of the line center points between the columns are considered, the extraction expectation of the laser rays is met on the whole, and the extraction efficiency is improved. And screening and processing of any number of line center points in each column are accurately and efficiently realized.

Description

Technical Field

[0001] The present invention belongs to the field of image data processing, and in particular relates to an image data extraction method, system and medium. Background Art

[0002] With the development of machine vision technology, image data processing has gradually become a key step to significantly improve image quality. For the 3D imaging technology of line laser 3D cameras, accurately obtaining the position of the laser line has become the core of 3D imaging. There are many methods for laser line extraction, such as thinning method, Steger algorithm based on Hessian matrix, Gaussian fitting method, etc. After obtaining the complete image, the complete image data is transferred to the PC for processing. For example, the directional template method, sliding window method, etc., use each column of local data to extract the center point of the line, which can be implemented in FPGA.

[0003] Patent document CN118587272A discloses a line center extraction method, system and medium suitable for FPGA parallel computing. By finding the column maximum value in each column direction and combining the parallel processing mechanism of FPGA, the line center can be efficiently and accurately obtained through the adjacent pixel values ​​and indexes of the column maximum value. Usually, only a single line center point can be extracted. Even if multiple line center points can be found through the operating logic of the sliding window in the document, multiple line center points cannot be screened and processed, and the information between columns cannot be taken into account, resulting in the extracted laser line only partially meeting expectations.

[0004] Therefore, in order to solve the above-mentioned problems, the present invention provides an image data extraction method, system and medium. Summary of the invention

[0005] The purpose of the present invention is to overcome the above problems existing in the prior art and to provide an image data extraction method, system and medium.

[0006] In order to achieve the above technical objectives and the above technical effects, the present invention is implemented through the following technical solutions: An image data extraction method collects all line center point data in each column direction of the image data and extracts the target line center point data to obtain laser line information in the image data. The extraction method includes: Divide the image data into a number of horizontal regions along the column direction, each horizontal region is composed of a number of rows of pixels, so as to count the distribution of the line center points in each horizontal region; Count the horizontal area with the largest number of line center points in each column direction to confirm the concentrated area of ​​line center points in each column; Determine whether the frequency of the concentrated area of ​​the line center points in each horizontal area is greater than the preset threshold. If so, output the line center point close to the preset reference position of the corresponding horizontal area in each column direction as the corresponding target line center point. If not, then: Determine whether the line center point concentration areas of each column are continuous. If so, output the line center point in the line center point concentration area of ​​the current column that is close to the target line center point of the previous column as the target line center point of the current column. If not, output the line center point in the preset position within the corresponding range of the line center point concentration area of ​​the current column as the target line center point of the current column.

[0007] Furthermore, a number of lateral regions intersect with each column of pixels to form a number of sub-regions, and the sub-region covering the largest number of center points is the line center point concentration region.

[0008] Furthermore, whether the line center point concentration area of ​​each column is continuous is determined by judging whether the line center point concentration area of ​​the previous column is located in the same horizontal area or in an adjacent horizontal area.

[0009] Furthermore, if the frequency of the concentrated area of ​​line center points in the top horizontal area is greater than a preset threshold, the line center points close to the first row of pixels at the top in each column direction are output as the corresponding target line center points.

[0010] Furthermore, if the frequency of the concentrated area of ​​line center points in the bottom horizontal area is greater than a preset threshold, the line center points close to the bottom last row of pixels in each column direction are output as the corresponding target line center points.

[0011] Furthermore, if the frequency of the concentrated area of ​​line center points in the horizontal area in the center of the image is greater than a preset threshold, the middle point of all the line center points in each column direction is output as the corresponding target line center point.

[0012] Furthermore, if the concentrated area of ​​line center points in the current column is judged to be continuous, the image height differences between the line center points of the current column and the target line center points of the previous column are calculated in turn to select the line center point corresponding to the minimum image height difference as the target line center point of the current column.

[0013] Furthermore, if the line center point concentration area of ​​the current column is determined to be discontinuous, the middle point of all the line center points in the line center point concentration area of ​​the current column is output as the target line center point of the current column.

[0014] The present invention also provides an image data extraction system, comprising: A region division module is used to divide the image data into a number of horizontal regions along the column direction, each of which is composed of a number of rows of pixels, so as to count the distribution of the line center points in each horizontal region; The regional statistics module is used to count the horizontal areas with the largest number of line center points in each column direction to confirm the concentrated areas of the line center points in each column; The strategy arbitration module is used to determine whether the frequency of the concentrated area of ​​the line center points in each horizontal area is greater than the preset threshold. If so, the line center point close to the preset reference position of the corresponding horizontal area in each column direction is output as the corresponding target line center point. If not, then: It is used to determine whether the line center point concentration areas of each column are continuous. If so, the line center point in the line center point concentration area of ​​the current column that is close to the target line center point of the previous column is output as the target line center point of the current column. If not, the line center point in the preset position within the corresponding range of the line center point concentration area of ​​the current column is output as the target line center point of the current column.

[0015] The present invention also provides a computer-readable storage medium, comprising a computer program, wherein the computer program implements the above extraction method when executed by a processor.

[0016] The beneficial effects of the present invention are: (1) The present invention can greatly optimize the processing strategy by dividing the image data into several horizontal areas, combining the concentrated areas of the line center points of each column and judging whether the frequency of the concentrated areas of the line center points in each horizontal area is greater than a preset threshold. When most of the line center points are concentrated in one horizontal area, the multiple line center points extracted from each column can be quickly processed in parallel, effectively improving the extraction efficiency; by judging whether the concentrated areas of the line center points of each column are continuous, it can effectively combine the continuous characteristics of the laser line, and make reasonable choices for laser line interruptions in complex situations to confirm the target line center point. The present invention takes into account the characteristics of the line center points between columns, meets the extraction expectations of the laser line as a whole, and accurately and efficiently realizes the screening and processing of any number of line center points in each column.

[0017] (2) The present invention determines whether the frequency of the line center point concentration area in each horizontal area is greater than a preset threshold and determines whether the line center point concentration areas in each column are continuous. In the face of different situations in various complex environments, different strategies can be selected according to different scenarios, and the specific methods in each strategy are combined to achieve rapid confirmation of the target line center point.

[0018] (3) The target line center points extracted from each column by the present invention have a strong correlation as a whole, which can effectively avoid the interference caused by errors of other line center points or the actual environment, and greatly improve the overall accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a flow chart of the extraction method in the present invention; Figure 2 It is a schematic diagram of the center points of each column line in the present invention; Figure 3 It is a schematic diagram of the horizontal area division in the present invention; Figure 4 It is a structural block diagram of the extraction system in the present invention. DETAILED DESCRIPTION

[0020] 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 of 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.

[0021] like Figure 1 As shown, this embodiment first provides an image data extraction method, which collects all line center point data in each column direction of the image data and extracts and forms target line center point data to obtain laser line information in the image data.

[0022] After acquiring image data, the line laser 3D camera will perform preliminary preprocessing, which is mainly used to filter out high-frequency noise in the image and reduce the misjudgment caused by high-frequency noise in the subsequent line center extraction. In the prior art, the processed image data is extracted according to the brightness weight method, and the area that first increases and then decreases is extracted to calculate the position of the line center point. There is no correlation between columns during the extraction process.

[0023] For different object surfaces, in this way, a column may sometimes extract multiple line center points. Therefore, after the line laser 3D camera completes the line center point extraction, all the laser lines extracted from each column can be stored. The storage uses the block RAM resources in the FPGA, and the storage format is as follows: Figure 2 As shown, assuming that there are N columns in total, and each column extracts at most M line center points, the line center points are finally stored in a two-dimensional manner as shown in the figure, where each row represents the line center point data of one column.

[0024] P (0,0) ~P (0,M-1) Represents the M laser lines extracted from column 0, and so on. Due to different sensor output data formats, some sensors scan from top to bottom, in which case the order of the line center points extracted from each column is increasing; some sensors scan from top to bottom at the same time or alternately scan from alternate rows, in which case the order of the line center points stored in each column is out of order.

[0025] Therefore, in order to reasonably select all the line center points of each column and extract the target line center point data, the extraction method specifically includes the following steps: Step 1: Divide the image data into a number of horizontal regions along the column direction, each of which is composed of a number of rows of pixels, so as to count the distribution of the line center points in each horizontal region.

[0026] like Figure 3 As shown in the figure, assuming that the image height is H, the image is divided into Q horizontal regions, and the height range corresponding to each horizontal region is 0~H1, H1~H2...H Q-1 ~H.

[0027] Step 2: Count the horizontal areas with the largest number of line center points in each column direction to confirm the concentrated area of ​​line center points in each column.

[0028] As a specific implementation of the present invention, a number of lateral regions intersect with each column of pixels to form a number of sub-regions, and the sub-region covering the largest number of center points is the line center point concentration region.

[0029] like Figure 3 As shown in the figure, each horizontal region and each column of pixels will intersect to form a sub-region, and according to the statistical results of each sub-region, the sub-region with the most line center points in each column can be identified, and thus used as the line center point concentration region. For example, the number of line center points in the sub-region corresponding to the horizontal region 0~H1 is recorded as S0, the number of line center points in the sub-region corresponding to the horizontal region H1~H2 is recorded as S1, and the number of line center points in the sub-region corresponding to the horizontal region H Q-1 The number of line center points in the sub-region corresponding to ~H is recorded as S Q-1 , if S0 is the largest, then the sub-region corresponding to S0 is used as the line center point concentration region.

[0030] Step 3: Determine whether the frequency of the concentrated area of ​​the line center points in each horizontal area is greater than the preset threshold. If so, output the line center point close to the preset reference position of the corresponding horizontal area in each column direction as the corresponding target line center point. If not, then: Determine whether the line center point concentration areas of each column are continuous. If so, output the line center point in the line center point concentration area of ​​the current column that is close to the target line center point of the previous column as the target line center point of the current column. If not, output the line center point in the preset position within the corresponding range of the line center point concentration area of ​​the current column as the target line center point of the current column.

[0031] The frequency of line center point concentration areas in the horizontal area, that is, the number of times the sub-area corresponding to the horizontal area is the line center point concentration area. For example, a laser line is distributed in the horizontal area 0~H1. At this time, according to the statistical results, it can be found that the sub-area corresponding to S0 is the line center point concentration area the most times, and it is much larger than the preset threshold. According to the reference position preset in advance in the horizontal area, such as the first row of pixel positions, the line center point closest to the reference position in each column can be found and output.

[0032] Determine whether the line center point concentration areas of each column are continuous, that is, determine whether the line center point concentration areas of the current column are discrete compared with the line center point concentration areas of the previous column. Specifically, it can be determined whether the line center point concentration areas of each column are located in the same horizontal area or adjacent horizontal areas as the line center point concentration areas of the previous column to determine whether they are continuous. In the specific calculation, the sorting values ​​of all sub-areas of each column can be set as 1, 2...M in sequence, and the difference between the sorting value of the sub-area corresponding to the line center point concentration area of ​​the current column and the sorting value of the sub-area corresponding to the line center point concentration area of ​​the previous column is calculated to obtain the discreteness ΔS. If ΔS≤0, the image continuity strategy is selected, that is, the line center point in the line center point concentration area of ​​the current column close to the target line center point of the previous column is output as the target line center point of the current column. If ΔS>1, the fixed range strategy is selected, that is, the line center point of the preset position in the corresponding range of the line center point concentration area of ​​the current column is output as the target line center point of the current column. The preset position can be preset in advance or calculated by the line center point actually extracted from the line center point concentration area.

[0033] As can be seen from the above, if the frequency of the concentrated area of ​​the line center points in the horizontal area is greater than the preset threshold, the line center points close to the preset reference position of the corresponding horizontal area in the column direction are output, which can usually include the following three situations: Case 1: If the frequency of the concentrated area of ​​line center points in the top horizontal area is greater than the preset threshold, that is, most of the line center points are concentrated in the top horizontal area, as a preferred implementation, the line center points close to the top first row of pixels in each column direction are output as the corresponding target line center points, as follows: Step 11: Assuming the image height is H, for each column of laser lines, taking column 0 as an example, initially set a maximum line center point position R0=H+1, and then read the first line center point P of column 0 in sequence (0,0) , if P (0,0) < N0, then replace R0, so R0=P (0,0) , otherwise, keep R0 unchanged.

[0034] Step 12: Repeat the steps in step 11 until all the line center points P extracted in column 0 are (0,0) ~P (0,M-1) All processing is completed. At this time, regardless of P(0,0) ~P (0,M-1) Whether the order of the line center points is disordered, P (0,0) ~P (0,M-1) The final R0 is the center point of the target line in the 0th column.

[0035] Step 13: Repeat the steps in step 11-step 12 for each column, and finally output the laser line closest to the 0th row of the entire image.

[0036] Case 2: If the frequency of the line center point concentration area in the bottom horizontal area is greater than the preset threshold, that is, most of the line center points are concentrated in the bottom horizontal area, as a preferred implementation, the line center point close to the bottom last row of pixels in each column direction is output as the corresponding target line center point, as follows: Step 21: Assuming the image height is H, for each column of laser lines, take column 0 as an example, initially set R0=0, and then read the first line center point P of column 0 in sequence (0,0) , if P (0,0) > N0, then replace R0, so R0=P (0,0) , otherwise, keep R0 unchanged.

[0037] Step 22: Repeat the steps in step 21 until all the line center points P extracted in column 0 are (0,0) ~P (0,M-1) All processing is completed. At this time, regardless of P (0,0) ~P (0,M-1) Whether the order of the line center points is disordered, P (0,0) ~P (0,M-1) The final R0 is the center point of the target line in column 0.

[0038] Step 23: Repeat the steps in step 21 and step 22 for each column, and finally output the laser line closest to the last row of the entire image.

[0039] Case three: If the frequency of the concentrated area of ​​line center points in the horizontal area in the center of the image is greater than the preset threshold, that is, most of the line center points are concentrated in the horizontal area in the center of the image, as a preferred implementation, the middle point of all the line center points in each column direction is output as the corresponding target line center point.

[0040] Step 31: Assuming the image height is H, for each column of laser lines, taking column 0 as an example, all the line center points P (0,0) ~P (0,M-1) Arrange them in ascending order, and then take the middle value as the center point R0 of the target line in column 0.

[0041] Step 32: Repeat the steps in step 31 for each column, and finally output the laser line closest to the last row of the entire image.

[0042] From the above, we can see that when most of the line center points are not concentrated in a certain horizontal area, it is determined whether the line center point concentration areas of each column are continuous, that is, whether the line center point concentration areas of each column are located in the same horizontal area or adjacent horizontal areas as the line center point concentration areas of the previous column to determine whether they are continuous. Specifically, there are the following two situations: Case 1: If the concentrated area of ​​the line center points of the current column is judged to be continuous, the image height difference between the line center points of the current column and the target line center point of the previous column is calculated in sequence to select the line center point corresponding to the minimum image height difference as the target line center point of the current column, as follows: Step 41: Assuming the image height is H, for each column of laser lines, taking column 0 as an example, all the line center points P (0,0) ~P (0,M-1) Arrange them in order from smallest to largest.

[0043] Step 42: Select the laser line R0 in the 0th column. For example, R0 can be selected as P (0,0) ~P (0,M-1) Then all laser lines P in column 1 are (1,0) ~P (1,M-1) Compare with R0 and calculate the height difference L with R0 (1,j) =|P (1,j) -R0|, thus comparing P (1,0) ~P (1,M-1) , confirm the smallest height difference L (1,k) , then the center point of the first column target line is R1=P (1,k) .

[0044] Step 43: Repeat the steps in step 41 and step 42 for the remaining columns, and finally output the most continuous laser line from left to right in the entire image.

[0045] Case 2: If the line center point concentration area of ​​the current column is determined to be discontinuous, as a preferred implementation, the middle point of all line center points in the line center point concentration area of ​​the current column is output as the target line center point of the current column, as follows: Step 51: Assuming the image height is H, for each column of laser lines, taking the 0th column as an example, all the line center points P (0,0) ~P (0,M-1) Arrange them in order from smallest to largest.

[0046] Step 52: Select the laser line P in the given range of column 0 (0,u) ~P (0,v), if u = v, that is, there is only one laser line in the given range, then select it as the center point of the target line in the 0th column; if u < v, select the midpoint from P (0,u) ~ P (0,v) as the center point of the target line in the 0th column.

[0047] Step 53: Repeat the steps in Step 51 - Step 52 for the remaining columns in sequence, and finally output the most continuous laser line from left to right in the entire image.

[0048] As Figure 4 shown, the second aspect of the present invention also provides an image data extraction system, including: A region division module, configured to divide image data into several horizontal regions along the column direction, and each horizontal region consists of several rows of pixels to count the distribution of line center points in each horizontal region; A region statistics module, configured to count the horizontal region with the largest number of line center points covered in each column direction to confirm the line center point concentration region of each column; A strategy arbitration module, configured to determine whether the frequency of the line center point concentration region in each horizontal region is greater than a preset threshold. If so, output the line center point near the preset reference position of the corresponding horizontal region in each column direction as the corresponding target line center point. If not, then: Determine whether the line center point concentration regions of each column are continuous. If so, output the line center point near the target line center point of the previous column in the line center point concentration region of the current column as the target line center point of the current column. If not, output the line center point at the preset position within the corresponding range of the line center point concentration region of the current column as the target line center point of the current column.

[0049] The third aspect of the present invention also provides a computer-readable storage medium, including a computer program, and when the computer program is executed by a processor, the above extraction method is implemented.

[0050] In practical applications, the computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device.

[0051] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0052] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0053] Computer program code for performing the operation of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0054] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. 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 present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0055] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An image data extraction method, which collects all line center point data in each column direction of the image data and extracts the target line center point data to obtain the laser line information in the image data, characterized in that: Extraction methods include: Divide the image data into a number of horizontal regions along the column direction, each horizontal region is composed of a number of rows of pixels, so as to count the distribution of the line center points in each horizontal region; Count the horizontal area with the largest number of line center points in each column direction to confirm the concentrated area of ​​line center points in each column; Determine whether the frequency of the concentrated area of ​​the line center points in each horizontal area is greater than the preset threshold. If so, output the line center point close to the preset reference position of the corresponding horizontal area in each column direction as the corresponding target line center point. If not, then: Determine whether the line center point concentration areas of each column are continuous. If so, output the line center point in the line center point concentration area of ​​the current column that is close to the target line center point of the previous column as the target line center point of the current column. If not, output the line center point at the preset position within the corresponding range of the line center point concentration area of ​​the current column as the target line center point of the current column.

2. The image data extraction method according to claim 1, characterized in that: Several lateral regions intersect with each column of pixels to form several sub-regions, and the sub-region covering the largest number of center points is the line center point concentration region.

3. The image data extraction method according to claim 1, characterized in that: Whether the line center point concentration area of ​​each column is located in the same horizontal area or adjacent horizontal areas as the line center point concentration area of ​​the previous column is determined to determine whether it is continuous.

4. The image data extraction method according to claim 1, characterized in that: If the frequency of the concentrated area of ​​line center points in the top horizontal area is greater than a preset threshold, the line center points close to the top first row of pixels in each column direction are output as the corresponding target line center points.

5. The image data extraction method according to claim 1, characterized in that: If the frequency of the concentrated area of ​​line center points in the bottom horizontal area is greater than a preset threshold, the line center points close to the bottom last row of pixels in each column direction are output as the corresponding target line center points.

6. The image data extraction method according to claim 1, characterized in that: If the frequency of the concentrated area of ​​line center points in the horizontal area in the center of the image is greater than a preset threshold, the middle point of all the line center points in each column direction is output as the corresponding target line center point.

7. The image data extraction method according to claim 1, characterized in that: If the concentrated area of ​​the line center points of the current column is judged to be continuous, the image height differences between the line center points of the current column and the target line center points of the previous column are calculated in sequence to select the line center point corresponding to the minimum image height difference as the target line center point of the current column.

8. The image data extraction method according to claim 1, characterized in that: If the line center point concentration area of ​​the current column is determined to be discontinuous, the middle point of all the line center points in the line center point concentration area of ​​the current column is output as the target line center point of the current column.

9. An image data extraction system, characterized in that: include: A region division module is used to divide the image data into a number of horizontal regions along the column direction, each of which is composed of a number of rows of pixels, so as to count the distribution of the line center points in each horizontal region; The regional statistics module is used to count the horizontal areas with the largest number of line center points in each column direction to confirm the concentrated areas of the line center points in each column; The strategy arbitration module is used to determine whether the frequency of the concentrated area of ​​the line center points in each horizontal area is greater than the preset threshold. If so, the line center point close to the preset reference position of the corresponding horizontal area in each column direction is output as the corresponding target line center point. If not, then: Determine whether the line center point concentration areas of each column are continuous. If so, output the line center point in the line center point concentration area of ​​the current column that is close to the target line center point of the previous column as the target line center point of the current column. If not, output the line center point at the preset position within the corresponding range of the line center point concentration area of ​​the current column as the target line center point of the current column.

10. A computer-readable storage medium comprising a computer program, characterized in that: When the computer program is executed by a processor, the extraction method according to any one of claims 1 to 8 is implemented.

Images