An image feature extraction and recognition integrated system

Abstract

The application discloses an image feature extraction and recognition integrated system, and relates to the field of image processing, comprising: a conversion module, which is used for receiving an image to be processed and converting image data into a multi-type image data set; a setting module, which is used for setting a feature recognition window specification and configuring the feature recognition window in each image in the multi-type image data set; and an analysis module, which is used for analyzing feature values of a region image under the feature recognition window in the image; the application enhances image quality through composite processing, acquires a region image by moving a self-defined window, captures local features by combining feature value calculation, determines a feature region based on an analysis result, accurately locates a key region through multi-dimensional fusion and iterative verification, and improves the comprehensiveness of feature extraction and the accuracy of recognition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to an image feature extraction and recognition integrated system. BACKGROUND

[0002] Image feature regions refer to local regions in images that have unique properties and can be used to distinguish or describe image content, which are core concepts in the field of computer vision. These regions are usually stable and common types include corners, blobs, edges, etc. Extracting feature regions can help computers complete image matching, target recognition, scene understanding, and other tasks, and are widely used in security monitoring, autonomous driving, medical image analysis, and other fields.

[0003] In the invention patent application with application number 202310499435.4, a feature image recognition system is disclosed. The system includes: a feature quantization unit for obtaining fundus color image, establishing a mapping relationship with retinal vascular structure morphological feature parameters through an encoder-decoder method, and performing feature quantization; a data dimension reduction unit for analyzing the structure and distribution of the quantized feature data itself, performing data dimension reduction, and obtaining the most different feature space; a prediction model establishment unit for using a plurality of random projection matrices to perform spatial dimension reduction according to the feature space, obtaining corresponding low-dimensional subspaces; for each low-dimensional subspace, a machine learning method is used to establish a low-dimensional feature parameter prediction model; a model integration unit for integrating the low-dimensional feature parameter prediction models using a majority voting method to obtain an integrated model corresponding to the low-dimensional subspaces; a final model unit for performing final integration of the integrated model using a model stacking method to obtain a final prediction model; and a prediction unit for identifying selected features of the obtained fundus color image according to the final prediction model. The application aims to solve the problem of how to effectively obtain the features of fundus color images and perform fine analysis and identification.

[0004] Based on the above, it is not difficult to see that the current image feature recognition technology is widely used in various fields.

[0005] Therefore, we newly propose an image feature extraction and recognition integrated system. SUMMARY

[0006] In view of the above-mentioned shortcomings of the prior art, the present application provides an image feature extraction and recognition integrated system, which can effectively solve the problems of the prior art.

[0007] To achieve the above purpose, the present application is realized by the following technical scheme.

[0008] The present application discloses an image feature extraction and recognition integrated system, which comprises:

[0009] The conversion module is configured to receive an image to be processed and convert image data into a plurality of types of image data sets. The setting module is configured to set a feature recognition window specification and configure a feature recognition window in each image in the plurality of types of image data sets. The analysis module is configured to analyze a feature value of a region image under the feature recognition window in the image. The selection module is configured to obtain an analysis result of the region image feature value in the analysis module and select a region image based on the analysis result. The recognition module is configured to recognize whether the region image selected by the selection module has an intersection region. The output module is configured to set a feature region radius, obtain a center pixel of the intersection region recognized by the recognition module, and determine a feature region based on the center pixel and the feature region radius.

[0010] The feature region radius set in the output module is a pixel number.

[0011] Further, the conversion module is internally provided with a storage unit configured to receive a conversion result of the conversion module for the image to be processed and store each type of image data obtained by the conversion.

[0012] The conversion of the image data into the plurality of types of image data sets by the conversion module includes a grayscale image, a contour image, a relief image, and a binary image.

[0013] Before the conversion of the image to be processed into the plurality of types of image data sets by the conversion module, the image to be processed is subjected to a quality enhancement process. After the quality enhancement process of the image to be processed is completed, the conversion of the plurality of types of image data sets is performed.

[0014] The plurality of types of image data sets further include an original image of the image to be processed.

[0015] Further, the quality enhancement process formula of the image to be processed is:

[0016] ;

[0017] In the formula, is the image after the quality enhancement process; is a structure enhancement parameter; is a Gaussian blur operator of a standard deviation of the image to be processed; is a detail sharpening parameter; ; is a global denoising parameter; represents a result of further Gaussian smoothing on a low-frequency structure layer; is the image to be processed; represents a pixel-by-pixel multiplication operation;

[0018] wherein, , , respectively represent the processing of structure enhancement, detail sharpening and global denoising on the image to be processed, the structure enhancement parameter The default value range is [0.5, 0.8], the detail sharpening parameter The default value range is [0.3, 1], the global denoising parameter The default value range is [0.1, 0.3].

[0019] Further, the setting module is provided with a control unit, which is used to control the continuous movement of the feature recognition window on the image surface to obtain different area images, and the control unit is used to forward the obtained area images to the analysis module in real time for processing.

[0020] The size of the feature recognition window is customized by the user of the system end, and is not greater than one fourth of the image area and not less than one sixteenth of the image area. When the control unit controls the continuous movement of the feature recognition window on the image surface, the feature recognition window moves in a straight line horizontally or vertically, and the two area images obtained by the adjacent two movements have an intersection area, and the intersection area is 1 / 4 to 1 / 2 of the area image area.

[0021] Further, the feature value calculation logic of the area image under the feature recognition window in the gray image is:

[0022] ;

[0023] In the formula: is the feature response value of the pixel point (x, y); is the gradient amplitude; is the gradient direction change rate; is the local contrast;

[0024] Wherein, the feature response value of each pixel in the area image under the feature recognition window is calculated based on formula (1), and then summed, and the sum result is recorded as the feature response value of the area image;

[0025] The feature value calculation logic of the area image under the feature recognition window in the contour image is:

[0026] ;

[0027] In the formula: is the feature performance value of the contour point (x, y); is the weight coefficient; is the curvature of the contour point; is the distance from the contour point to the concave area in the area image; is the association strength of the contour point and the hole in the area image;

[0028] Wherein, and are non-zero positive numbers, the feature value of the region image under the feature recognition window is calculated based on formula (2) for each contour point pixel in the region image, and then summed, and the sum is recorded as the feature value of the region image.

[0029] Further, the feature value calculation logic of the region image under the feature recognition window in the relief image is:

[0030] ;

[0031] In the formula: is the feature score of the region image globally; , , is the gray scale gradient, texture entropy, and edge density of the region image; is the extreme value of the gray scale gradient of the global neighborhood of the region image; is the extreme value of the texture entropy of the global neighborhood of the region image; is the extreme value of the edge density of the global neighborhood of the region image;

[0032] The feature value calculation logic of the region image under the feature recognition window in the image to be processed of the quality enhancement processing is:

[0033] ;

[0034] In the formula: is the feature representation value of the current pixel (x, y); is the gray scale value of the current pixel (x, y); is the average gray scale value in the neighborhood N centered at (x, y); is the gray scale standard deviation in the neighborhood N; is a minimum constant; is the gradient amplitude of the current pixel (x, y); is the global maximum value of the gradient amplitude in the image; is the gray scale entropy in the neighborhood N centered at (x, y); is the theoretical maximum value of the entropy value;

[0035] Wherein, each pixel in the region image is calculated based on formula (3), and then summed, and the sum is recorded as the feature representation value of the region image.

[0036] Further, the feature value calculation logic of the region image under the feature recognition window in the binary image is:

[0037] ;

[0038] In the formula: is the feature reference value of the region image; is the number of pixels with two color values ​​in the region image; is the total number of pixels in the region image.

[0039] Furthermore, the analysis module runs the analysis of the regional image and the result of the characteristic value is recorded as:

[0040] ;

[0041] ;

[0042] ;

[0043] ;

[0044] ;

[0045] Five sets of eigenvalue calculation results are obtained, and the calculation results in the five sets are arranged in descending order;

[0046] Obtain the region image corresponding to the first calculation result in the five sets, and then use the recognition module to identify whether there is an intersection between the region images. If so, jump to the output module to run;

[0047] If it does not exist, obtain a regional image that does not intersect with other regional images, use the obtained regional image as the iteration target, and iterate the regional image corresponding to the eigenvalues ​​of the eigenvalue calculation result set corresponding to the iteration target, which are adjacently sorted.

[0048] Furthermore, the selection module and the identification module are provided with a jump unit at the lower level. The jump unit is triggered when the identification module's recognition result is no, and jumps to the selection module to run again until the identification module follows the selection module and runs again. When the recognition result is yes, the process ends.

[0049] Furthermore, the conversion module is interactively connected to a storage unit via a wireless network, the conversion module is interactively connected to the setting module via a wireless network, the lower level of the setting module is interactively connected to the control unit via a wireless network, the setting module is interactively connected to the analysis module and the selection module via a wireless network, the analysis module is interactively connected to the control unit via a wireless network, the selection module is interactively connected to the identification module and the output module via a wireless network, and the selection module and the lower level of the identification module are interactively connected to the jump unit via a wireless network.

[0050] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0051] The application provides an image feature extraction and recognition integrated system, which, in the running process, first enhances the image quality, combines the composite processing of structure enhancement, detail sharpening and global denoising, effectively improves the image quality, lays a foundation for subsequent analysis, then converts the image into a multi-type data set containing gray scale, contour, relief, binary image and original image, analyzes from different feature dimensions, continuously moves and acquires the partially overlapped area image on the image through the feature recognition window with a self-definable specification, combines the feature value calculation logic corresponding to each type of image, realizes the accurate capture of the local features of the image, and then selects the area image based on the feature value analysis result, determines the feature area by judging whether there is an intersection area, the multi-dimensional fusion analysis and iterative verification mechanism can accurately locate the key feature area of the image, effectively improves the comprehensiveness of feature extraction and the accuracy of recognition, and seeks a new image feature recognition and extraction scheme for the image analysis field. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0053] Figure 1 It is a structural schematic diagram of an image feature extraction and recognition integrated system. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0055] The present application will be further described in combination with the embodiments.

[0056] Embodiment:

[0057] An image feature extraction and recognition integrated system of the embodiment, as shown in Figure 1 It comprises:

[0058] The conversion module is used for receiving the image to be processed, and converting the image data into a multi-type image data set;

[0059] The conversion module is internally provided with a storage unit, which is used to receive the conversion result of the conversion module on the to-be-processed image and store each type of image data obtained by conversion;

[0060] The conversion module converts the image data into a plurality of types of image data sets, including a grayscale image, a contour image, a relief image and a binary image;

[0061] The conversion module performs a quality enhancement process on the to-be-processed image before converting the to-be-processed image into a plurality of types of image data sets, and performs the conversion operation of the plurality of types of image data sets after the quality enhancement process of the to-be-processed image is completed;

[0062] The plurality of types of image data sets further include an original image of the to-be-processed image;

[0063] The quality enhancement process formula of the to-be-processed image is:

[0064] ;

[0065] In the formula, is the image after the quality enhancement process; is a structure enhancement parameter; is a Gaussian blur operator of a standard deviation of the to-be-processed image; is a detail sharpening parameter; ; is a global denoising parameter; represents a result of further Gaussian smoothing on a low-frequency structure layer; is the to-be-processed image; represents a pixel-by-pixel multiplication operation;

[0066] wherein, 、 、 respectively represent structure enhancement, detail sharpening and global denoising processes on the to-be-processed image, the structure enhancement parameter has a default value range of [0.5, 0.8], the detail sharpening parameter has a default value range of [0.3, 1], and the global denoising parameter has a default value range of [0.1, 0.3];

[0067] Through the calculation of the above logical formula, the to-be-processed image is enhanced to ensure that the subsequent processing result of the system on the to-be-processed image is more accurate;

[0068] The setting module is used to set a feature recognition window specification and configure the feature recognition window in each image in the plurality of types of image data sets;

[0069] A control unit is provided at the lower level of the setting module. The control unit is used to control the feature recognition window to continuously move on the image surface to obtain images of different regions. The control unit forwards the obtained regional images to the analysis module in real time for processing by the analysis module.

[0070] The specification of the feature recognition window is customized by the system user and is no larger than one-quarter of the image area and no smaller than one-sixteenth of the image area. When the control unit controls the feature recognition window to move continuously on the image surface, the feature recognition window moves in a horizontal or vertical straight line, and the two regional images obtained by two adjacent movements have an intersection area, and the intersection area is 1 / 4 to 1 / 2 of the regional image area;

[0071] An analysis module is used to analyze the characteristic values ​​of the image in the area under the feature recognition window in the image;

[0072] The logic for calculating the eigenvalues ​​of the image in the area under the feature recognition window in the grayscale image is:

[0073] ;

[0074] Where: is the characteristic response value of the pixel point (x, y); is the gradient amplitude; is the rate of change of gradient direction; is the local contrast;

[0075] Among them, based on formula (1), each pixel in the regional image under the feature recognition window is calculated, and then summed up, and the summation result is recorded as the feature response value of the regional image;

[0076] The logic for calculating the feature value of the area image under the feature recognition window in the contour image is:

[0077] ;

[0078] Where: is the characteristic expression value of the contour point (x, y); is the weight coefficient; is the contour point curvature; is the distance from the contour point to the concave area in the regional image; is the correlation strength between the contour points and the holes in the region image;

[0079] in, The sum is 1, and they are all non-zero positive numbers. Based on formula (2), each contour point pixel in the regional image under the feature recognition window is calculated and then summed. The sum result is recorded as the feature performance value of the regional image;

[0080] The logic for calculating the feature value of the area image under the feature recognition window in the relief image is:

[0081] ;

[0082] In the formula: is the feature score of the region image globally; 、 、 is the gray gradient, texture entropy, and edge density of the region image; is the extreme value of the gray gradient of the global neighborhood of the region image; is the extreme value of the texture entropy of the global neighborhood of the region image; is the extreme value of the edge density of the global neighborhood of the region image;

[0083] The feature value calculation logic of the region image under the feature recognition window in the image to be processed of the picture quality enhancement processing is as follows:

[0084] ;

[0085] In the formula: is the feature representation value of the current pixel (x, y); is the gray value of the current pixel (x, y); is the average gray value in the neighborhood N with (x, y) as the center; is the gray standard deviation in the neighborhood N; is a minimum constant; is the gradient amplitude of the current pixel (x, y); is the global maximum value of the gradient amplitude in the image; is the gray entropy in the neighborhood N with (x, y) as the center; is the theoretical maximum value of the entropy value;

[0086] Wherein, the formula (3) is used to calculate each pixel in the region image, and then the sum is calculated, and the sum result is recorded as the feature representation value of the region image;

[0087] The feature value calculation logic of the region image under the feature recognition window in the binary image is as follows:

[0088] ;

[0089] In the formula: is the feature reference value of the region image; is the number of pixels of two color values in the region image; is the total amount of pixels in the region image;

[0090] The feature value of each type of image in the feature recognition window is calculated by the logical formula, and the operation data support is provided for the selection module and the recognition module of the system in the embodiment, so as to ensure that the system in the embodiment can stably recognize and extract the feature region in the image to be processed.

[0091] The feature value result of the analysis module running the analysis region image is recorded as:

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] ;

[0097] Five feature value calculation result sets are obtained, and each calculation result in the five sets is arranged in descending order;

[0098] The first calculation result corresponding region image in the five sets is obtained, and the recognition module is used to recognize whether there is an intersection of each region image, if there is, the jump output module is run;

[0099] If there is no intersection, the region image which has no intersection with other region images is obtained, and the obtained region image is taken as an iteration target, and the region image corresponding to the adjacent feature value in the feature value calculation result set corresponding to the iteration target is iterated;

[0100] The selection module is used to obtain the feature value analysis result of the region image in the analysis module, and the region image is selected based on the analysis result;

[0101] The recognition module is used to recognize whether there is an intersection region of the region image selected by the selection module;

[0102] The selection module and the recognition module are provided with a jump unit, the jump unit is triggered when the recognition result of the recognition module is false, and jumps to the selection module to run again, until the recognition result is true after the recognition module runs again following the selection module, and the process ends;

[0103] The output module is used to set the feature region radius, obtain the center pixel of the intersection region recognized by the recognition module, and determine the feature region based on the center pixel and the feature region radius;

[0104] The feature region radius set in the output module is the number of pixels;

[0105] The conversion module is connected with the storage unit through a wireless network, the conversion module is connected with the setting module through a wireless network, the setting module is connected with the control unit through a wireless network, the setting module is connected with the analysis module and the selection module through a wireless network, the analysis module is connected with the control unit through a wireless network, the selection module is connected with the identification module and the output module through a wireless network, and the selection module and the identification module are connected with the jump unit through a wireless network.

[0106] In the embodiment, the conversion module receives an image to be processed, converts the image data into a multi-type image data set, the storage unit synchronously receives the conversion result of the image to be processed by the conversion module, and stores each type of image data obtained by conversion, the setting module is post-operated to set the feature recognition window specification, and the feature recognition window is configured on each image in the multi-type image data set, the control unit controls the feature recognition window to continuously move on the image surface to obtain different region images, the control unit transmits the obtained region images to the analysis module in real time, and the analysis module processes the region images, analyzes the feature values of the region images under the feature recognition window in the image, the selection module further obtains the analysis result of the feature values of the region images in the analysis module, selects the region images based on the analysis result, the identification module identifies whether the region images selected by the selection module have an intersection region, the jump unit is triggered when the identification result of the identification module is no, and the jump unit jumps to the selection module to run again until the identification result of the identification module is yes after the selection module runs again, and the process ends, and finally, the output module sets the feature region radius, obtains the center pixel of the intersection region identified by the identification module, and determines the feature region based on the center pixel and the feature region radius.

[0107] Through the system operation in the above embodiment, a new feature recognition and extraction scheme suitable for a wide range of scene images is provided, and the application of the image feature recognition and extraction technology is more universal.

[0108] In summary, in the running process of the system in the above embodiment, the image quality is enhanced, the composite processing of structure enhancement, detail sharpening and global denoising is combined, the image quality is effectively improved, the foundation is laid for subsequent analysis, the image is converted into a multi-type data set containing gray scale, contour, relief, binary image and original image, the analysis is performed from different feature dimensions, the feature recognition window with a self-definable specification is continuously moved on the image and part of the overlapping region images are obtained, the feature value calculation logic corresponding to each type of image is combined, the local feature of the image is accurately captured, the region images are selected based on the feature value analysis result, and the feature region is determined by judging whether there is an intersection region. This multi-dimensional fusion analysis and iterative verification mechanism can accurately locate the key feature region of the image, effectively improve the comprehensiveness of feature extraction and the accuracy of recognition, and seek a new image feature recognition and extraction scheme for the image analysis field.

[0109] The above examples are only used to illustrate the technical solutions of the present application, but not to limit the present application; although the present application has been described in detail with reference to the foregoing examples, those ordinarily skilled in the art should understand that they can modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for some technical features therein; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An integrated system for image feature extraction and recognition, characterized in that: include: A conversion module, configured to receive an image to be processed and convert the image data into a multi-type image dataset; A setting module is used to set the feature recognition window specifications and configure the feature recognition window for each image in the multi-type image data set; An analysis module is used to analyze the characteristic values ​​of the image in the area under the feature recognition window in the image; The logic for calculating the eigenvalues ​​of the image in the area under the feature recognition window in the grayscale image is: ； Where: is the characteristic response value of the pixel point (x, y); is the gradient amplitude; is the rate of change of gradient direction; is the local contrast; Among them, based on formula (1), each pixel in the regional image under the feature recognition window is calculated, and then summed up, and the summation result is recorded as the feature response value of the regional image; The logic for calculating the feature value of the area image under the feature recognition window in the contour image is: ； Where: is the characteristic expression value of the contour point (x, y); is the weight coefficient; is the contour point curvature; is the distance from the contour point to the concave area in the regional image; is the correlation strength between the contour points and the holes in the region image; in, The sum is 1, and they are all non-zero positive numbers. Based on formula (2), each contour point pixel in the regional image under the feature recognition window is calculated and then summed. The summation result is recorded as the feature performance value of the regional image; A selection module is used to obtain the analysis results of the regional image feature values ​​in the analysis module and select the regional image based on the analysis results; A recognition module, used to identify whether the area images selected by the selection module have an intersection area; The output module is used to set the radius of the feature area, obtain the center pixel of the intersection area identified by the recognition module, and determine the feature area based on the center pixel and the radius of the feature area; Among them, the feature area radius set in the output module is the number of pixels.

2. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The conversion module is internally provided with a storage unit, and the storage unit is used to receive the conversion result of the conversion module on the image to be processed and store the various types of image data obtained by the conversion; The conversion module converts the image data into a multi-type image data set including: grayscale image, contour image, relief image, and binary image; The conversion module performs image quality enhancement processing on the image to be processed before converting it into a multi-type image dataset, and then performs the conversion operation of the multi-type image dataset after the image quality enhancement processing is completed on the image to be processed; The multi-type image dataset also includes the original image of the image to be processed.

3. The integrated system for image feature extraction and recognition according to claim 2, characterized in that: The image quality enhancement processing formula of the image to be processed is: ； Where: The image is processed for image quality enhancement; Enhance parameters for the structure; is the Gaussian blur result of the image to be processed based on the standard deviation; Detail sharpening parameters; ; is the global denoising parameter; It represents the result after further Gaussian smoothing of the low-frequency structure layer; is the image to be processed; Represents the element-by-element multiplication operation; in, 、 、 They represent the processing of structure enhancement, detail sharpening and global denoising on the processed image, and the structure enhancement parameters The default value range is [0.5, 0.8], detail sharpening parameter The default value range is [0.3, 1], the global denoising parameter The default value range is [0.1, 0.3].

4. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The setting module is provided with a control unit at a lower level, and the control unit is used to control the feature recognition window to continuously move on the image surface to obtain images of different regions. The control unit operates to forward the obtained regional images to the analysis module in real time for processing by the analysis module; The specifications of the feature recognition window are customized by the system user and are no larger than one quarter of the image area and no smaller than one sixteenth of the image area. When the control unit controls the feature recognition window to move continuously on the image surface, the feature recognition window moves in a horizontal or vertical straight line, and there is an intersection area between the two area images obtained by two adjacent movements, and the intersection area is 1 / 4 to 1 / 2 of the area of ​​the area image.

5. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The logic for calculating the feature value of the area image under the feature recognition window in the relief image is: ； Where: is the global feature score of the regional image; 、 、 is the grayscale gradient, texture entropy, and edge density of the regional image; is the gray gradient extreme value of the global neighborhood of the regional image; is the extreme value of texture entropy of the global neighborhood of the regional image; is the edge density extreme value of the global neighborhood of the regional image; The logic for calculating the characteristic value of the image in the area under the feature recognition window in the image to be processed during image quality enhancement is as follows: ； Where: is the feature representation value of the current pixel (x, y); is the grayscale value of the current pixel (x, y); is the average grayscale value in the neighborhood N centered at (x, y); is the grayscale standard deviation within the neighborhood N; is a minimum constant; is the gradient magnitude of the current pixel (x, y); is the global maximum value of the gradient amplitude in the image; is the grayscale entropy within the neighborhood N centered at (x, y); is the theoretical maximum value of entropy; Here, each pixel in the regional image is calculated based on formula (3), and then summed up. The summation result is recorded as the characteristic representation value of the regional image.

6. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The calculation logic of the feature value of the area image under the feature recognition window in the binary image is: ； Where: is the characteristic reference value of the regional image; is the number of pixels with two color values ​​in the region image; is the total number of pixels in the region image.

7. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The analysis module runs the analysis of the characteristic value results of the regional image to obtain five characteristic value calculation result sets, and the calculation results in the five sets are arranged in descending order; Obtain the region image corresponding to the first calculation result in the five sets, and then use the recognition module to identify whether there is an intersection between the region images. If so, jump to the output module to run; If it does not exist, obtain a regional image that does not intersect with other regional images, use the obtained regional image as the iteration target, and iterate the regional image corresponding to the eigenvalues ​​of the eigenvalue calculation result set corresponding to the iteration target, which are adjacently sorted.

8. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The selection module and the identification module are provided with a jump unit at the lower level. The jump unit is triggered when the identification module recognizes that the result is no, and jumps to the selection module to run again until the identification module follows the selection module and runs again. When the recognition result is yes, the process ends.

9. The integrated system for image feature extraction and recognition according to claim 1, characterized in that: The conversion module is interactively connected to a storage unit via a wireless network, the conversion module is interactively connected to the setting module via a wireless network, the setting module is interactively connected to the control unit via a wireless network, the setting module is interactively connected to the analysis module and the selection module via a wireless network, the analysis module is interactively connected to the control unit via a wireless network, the selection module is interactively connected to the identification module and the output module via a wireless network, and the selection module and the identification module are interactively connected to the jump unit via a wireless network.

Images