A special equipment worker positioning system and method using image processing

Abstract

The present invention provides a system and method for positioning special equipment practitioners by image processing. The system includes an input module, a communication module, an RFID reader, an RFID tag, a positioning module and a confirmation module; the input module is used to input the employee number of the special equipment practitioner; the communication module is used to forward the employee number to the RFID reader; the RFID reader is used to generate a feedback message; the communication module is also used to transmit the feedback message to the positioning module; the positioning module is used to determine the position coordinates of the special equipment practitioner corresponding to the employee number according to the feedback message; the confirmation module is used to confirm whether the equipment practitioner is at the position coordinates to obtain a confirmation result. The method rate corresponds to the system. The present invention can avoid the occurrence of positioning errors when the RFID tag is separated from the special equipment practitioner. It is conducive to improving the accuracy of positioning of special equipment practitioners.

Description

Technical Field

[0001] The present invention relates to the field of positioning, and in particular to a positioning system and method for personnel engaged in special equipment for image processing. Background Art

[0002] Special equipment practitioners refer to workers and their related managers of boilers, pressure vessels, pressure pipelines, elevators, lifting machinery, passenger ropeways, large amusement facilities, and special motor vehicles in the field (factory). For special equipment practitioners, because their working environment is relatively dangerous, they need to be positioned in real time. In the prior art, positioning is generally performed by setting an RFID tag and a wired RFID reader. However, this positioning method can only determine the position of the RFID tag, but the RFID tag is not necessarily carried by the special equipment practitioners, so it is easy to misposition. Summary of the invention

[0003] In view of the above problems, the object of the present invention is to provide a special equipment practitioner positioning system and method for image processing.

[0004] In one aspect, the present invention provides a special equipment practitioner positioning system for image processing, comprising an input module, a communication module, an RFID reader, an RFID tag, a positioning module, and a confirmation module;

[0005] The input module is used to input the employee number of the special equipment practitioner and transmit the employee number to the communication module;

[0006] The communication module is used to forward the employee number to the RFID reader;

[0007] The RFID reader is used to determine whether there is an RFID tag corresponding to the employee number within its communication range. If so, the RFID reader generates a feedback message and transmits the feedback message to the communication module;

[0008] The communication module is also used to transmit the feedback message to the positioning module;

[0009] The positioning module is used to determine the position coordinates of the special equipment practitioner corresponding to the employee number according to the feedback message, and send the position information to the confirmation module;

[0010] The confirmation module is used to use image processing technology to confirm whether the equipment practitioner is at the position coordinates to obtain a confirmation result.

[0011] Preferably, the communication module includes a wireless cellular communication network or an optical fiber communication network;

[0012] The wireless cellular communication network or the optical fiber communication network is used to transmit the employee number to the RFID reader;

[0013] The wireless cellular communication network or the optical fiber communication network is also used to transmit the feedback message to the positioning module.

[0014] Preferably, the RFID tag is an active tag, and the RFID tag broadcasts a preset positioning message in a fixed time period;

[0015] The positioning message includes the employee number of the special equipment practitioner corresponding to the RFID tag, the ID of the RFID tag, and the remaining power of the RFID tag.

[0016] Preferably, determining whether there is an RFID tag corresponding to the employee number within the communication range includes:

[0017] The RFID reader matches the employee number with the employee numbers contained in all positioning messages received within a preset time period. If the match is successful, it means that there is an RFID tag corresponding to the employee number within the communication range of the RFID reader. If the match fails, it means that there is no RFID tag corresponding to the employee number within the communication range of the RFID reader.

[0018] Preferably, the feedback message includes the remaining power of the RFID tag corresponding to the employee number and the ID of the RFID reader.

[0019] Preferably, the determining the position coordinates of the special equipment employee corresponding to the employee number according to the feedback message includes:

[0020] Acquire the location coordinates of the RFID from an RFID location database according to the ID of the RFID reader;

[0021] The location coordinates are used as the location coordinates of the special equipment practitioner corresponding to the employee number.

[0022] Preferably, the confirmation module includes an image acquisition unit, an image recognition unit and a result display unit;

[0023] The image acquisition unit is used to acquire facial images of all persons within a range with a radius R and a center at the position coordinate;

[0024] The image recognition unit is used to determine whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number.

[0025] If yes, then the confirmation result is that the special equipment practitioner corresponding to the employee number is at the position coordinates.

[0026] If not, the confirmation result is that the special equipment practitioner corresponding to the employee number is not at the location coordinates;

[0027] The result display unit is used to visually display the confirmation result.

[0028] On the other hand, the present invention also provides a method for positioning special equipment practitioners by image processing, comprising:

[0029] S1, inputting the employee number of the special equipment practitioner through the input module, and transmitting the employee number to the communication module;

[0030] S2, forwarding the employee number to the RFID reader through a communication module;

[0031] S3, determining, by an RFID reader, whether there is an RFID tag corresponding to the employee number within its communication range, and if so, generating a feedback message by the RFID reader, and transmitting the feedback message to the communication module;

[0032] S4, transmitting the feedback message to the positioning module through the communication module;

[0033] S5, determining the position coordinates of the special equipment employee corresponding to the employee number according to the feedback message through the positioning module, and sending the position information to the confirmation module;

[0034] S6, using image processing technology to confirm whether the equipment operator is at the position coordinates through the confirmation module to obtain a confirmation result.

[0035] The present invention first obtains the position coordinates of the special equipment practitioners through the RFID tag, and then confirms whether the special equipment practitioners are at the position coordinates through image recognition. This setting method can avoid the occurrence of positioning errors when the RFID tag is separated from the special equipment practitioners. It is conducive to improving the accuracy of positioning the special equipment practitioners. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0037] Figure 1 , which is a diagram of an exemplary embodiment of a special equipment practitioner positioning system for image processing of the present invention. DETAILED DESCRIPTION

[0038] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0039] like Figure 1 In one embodiment shown, on the one hand, the present invention provides a positioning system for special equipment practitioners in image processing, including an input module, a communication module, an RFID reader, an RFID tag, a positioning module and a confirmation module;

[0040] The input module is used to input the employee number of the special equipment practitioner and transmit the employee number to the communication module;

[0041] The communication module is used to forward the employee number to the RFID reader;

[0042] The RFID reader is used to determine whether there is an RFID tag corresponding to the employee number within its communication range. If so, the RFID reader generates a feedback message and transmits the feedback message to the communication module;

[0043] The communication module is also used to transmit the feedback message to the positioning module;

[0044] The positioning module is used to determine the position coordinates of the special equipment practitioner corresponding to the employee number according to the feedback message, and send the position information to the confirmation module;

[0045] The confirmation module is used to use image processing technology to confirm whether the equipment practitioner is at the position coordinates to obtain a confirmation result.

[0046] The present invention first obtains the position coordinates of the special equipment practitioners through the RFID tag, and then confirms whether the special equipment practitioners are at the position coordinates through image recognition. This setting method can avoid the occurrence of positioning errors when the RFID tag is separated from the special equipment practitioners. It is conducive to improving the accuracy of positioning the special equipment practitioners.

[0047] The input module may include various input devices, including a keyboard, a touch screen, a voice input device, etc. After the input device obtains the employee number, it sends the employee number to the communication module through the terminal host.

[0048] RFID readers are scattered throughout the working environment of special equipment practitioners.

[0049] Preferably, the communication module includes a wireless cellular communication network or an optical fiber communication network;

[0050] The wireless cellular communication network or the optical fiber communication network is used to transmit the employee number to the RFID reader;

[0051] The wireless cellular communication network or the optical fiber communication network is also used to transmit the feedback message to the positioning module.

[0052] The fiber optic communication network mainly transmits the employee number to the RFID reader when the terminal host uses WiFi communication or fiber optic communication;

[0053] If the terminal host uses chips including 4G communication chips, 5G communication chips, etc., the employee number can be sent to the communication base station corresponding to the wireless cellular communication network, and then the communication base station forwards the employee number to the RFID reader.

[0054] Preferably, the RFID tag is an active tag, and the RFID tag broadcasts a preset positioning message in a fixed time period;

[0055] The positioning message includes the employee number of the special equipment practitioner corresponding to the RFID tag, the ID of the RFID tag, and the remaining power of the RFID tag.

[0056] Specifically, the use of active tags can greatly increase the communication distance and effectively reduce the size of the RFID reader.

[0057] Preferably, determining whether there is an RFID tag corresponding to the employee number within the communication range includes:

[0058] The RFID reader matches the employee number with the employee numbers contained in all positioning messages received within a preset time period. If the match is successful, it means that there is an RFID tag corresponding to the employee number within the communication range of the RFID reader. If the match fails, it means that there is no RFID tag corresponding to the employee number within the communication range of the RFID reader.

[0059] Specifically, the preset time period can be determined in the following manner:

[0060] The RFID reader records the time when the employee number is received as T1, and the preset period interval as L, so the preset time period is [T1-L, T1].

[0061] Preferably, the feedback message includes the remaining power of the RFID tag corresponding to the employee number and the ID of the RFID reader.

[0062] Preferably, the determining the position coordinates of the special equipment employee corresponding to the employee number according to the feedback message includes:

[0063] Acquire the location coordinates of the RFID from an RFID location database according to the ID of the RFID reader;

[0064] The location coordinates are used as the location coordinates of the special equipment practitioner corresponding to the employee number.

[0065] Preferably, the confirmation module includes an image acquisition unit, an image recognition unit and a result display unit;

[0066] The image acquisition unit is used to acquire facial images of all persons within a range with a radius R and a center at the position coordinate;

[0067] The image recognition unit is used to determine whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number.

[0068] If yes, then the confirmation result is that the special equipment practitioner corresponding to the employee number is at the position coordinates.

[0069] If not, the confirmation result is that the special equipment practitioner corresponding to the employee number is not at the location coordinates;

[0070] The result display unit is used to visually display the confirmation result.

[0071] In one embodiment, the image acquisition unit includes an unmanned vehicle, and the unmanned vehicle is used to obtain facial images of all persons within a range of radius R with the position coordinate as the center of the circle.

[0072] In another embodiment, the image acquisition unit includes a drone, and the drone is used to acquire facial images of all persons within a range of a radius R with the position coordinate as the center.

[0073] Preferably, the determining whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number includes:

[0074] Store the facial images of all the persons into a set Q;

[0075] Obtain the first feature information of each facial image in the set Q respectively;

[0076] Obtaining a reserved facial image of a special equipment practitioner corresponding to the employee number from a database, and obtaining second feature information of the reserved facial image;

[0077] The first feature information corresponding to each facial image in the set Q is matched with the second feature information. If the match is successful, it indicates whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number.

[0078] Preferably, the step of respectively obtaining the first feature information of each facial image in the set Q comprises:

[0079] For the qth face image img in the set Q p , use the following method to obtain the first characteristic information contained therein:

[0080] Img p Convert to grayscale image;

[0081] Performing filtering on the grayscale image to obtain a filtered image;

[0082] Using the Otsu algorithm to obtain a foreground image in the filtered image;

[0083] The first feature information contained in the foreground image is obtained using a HOG feature extraction algorithm.

[0084] Preferably, filtering the grayscale image to obtain a filtered image comprises:

[0085] Perform edge detection on the grayscale image, obtain edge pixel points contained in the grayscale image, and store all edge pixel points into a set blU; store all pixel points in the grayscale image into a set N;

[0086] The filtered image is filtered using iterative filtering:

[0087] The first filtering process uses a preset filtering function to filter each edge pixel point included in blU to obtain the first filtered image img1; the edge pixel points included in blU are stored in the filtered pixel point set P;

[0088] In the second filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q1, and the pixel points in the set Q1 that do not belong to the filtered pixel point set P are stored in the to-be-filtered set R2; in the first filtered image img1, a preset filtering function is used to filter each edge pixel point included in the to-be-filtered set R2 to obtain the second filtered image img2, and the pixel points included in the to-be-filtered set R2 are stored in the filtered pixel point set P;

[0089] In the mth filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q m-1 In the above example, we will set Q m-1 The pixels that do not belong to the filtered pixel set P are stored in the to-be-filtered set R m In the m-1th filtering image img m-1 The preset filter function is used to treat the filter set R m Filter each edge pixel contained in to obtain the mth filtered image img m , the set to be filtered R m The pixels contained in are stored in the filtered pixel set P;

[0090] The condition for the iteration to end is Q m-1 is an empty set;

[0091] The preset filtering function includes:

[0092]

[0093] Where apix(c) represents the result of filtering pixel c, nuc represents the set of pixels in a window of preset size centered on c, lst(c,g) represents the length of the straight line between c and g, and zg 2 represents the variance of the straight line length between the pixel point in nuc and c,

[0094] z represents a preset constant parameter, noisdx represents the variance of the noise estimation of the image to be filtered, ue(c) and ue(g) represent the gradient amplitudes of c and g respectively, and xe(g) represents the gradient correction parameter. If ue(c)>bl(nuc), then xe(g)=ue ma -ue(g), if ue(c)<bl(nuc), then xe(g)=ue mi -ue(g), if ue(c) = bl(nuc), then xe(g) = 0, bl(nuc) represents the mean value of the gradient amplitude of the pixels contained in nuc, ue ma andue mi They respectively represent the maximum and minimum values ​​of the gradient amplitude of the pixels contained in nuc; pix(g) represents the pixel value of pixel g, nfnuc represents the number of pixels contained in nuc, gd represents the preset control parameter, and pl(nuc) represents the mean pixel value of the pixels contained in nuc.

[0095] In the above embodiment of the present invention, when filtering a grayscale image, edge pixels are first obtained, and then filtering is performed by gradually diffusing from the edge pixels to the pixels in the 8-neighborhood. This processing method can effectively improve the accuracy of the filtering result while retaining edge details. The filtering method in the prior art generally uses the same filtering function to filter all pixels in the same image, which is not conducive to maintaining the edge detail information of the image.

[0096] In the above-mentioned embodiment of the present invention, the edge pixel points obtained by edge detection do not include noise pixel points. Therefore, through the step-by-step diffusion filtering process, the information in the previous filtered image can be passed to the next filtering process. When the iteration is completed, the correct information carried by the edge pixel points is diffused to all pixel points, which is conducive to retaining the edge detail information of the image.

[0097] In the filtering process, the filtering result is obtained by weighted fusion of other pixels in the neighborhood of the pixel currently being filtered, which is further conducive to retaining the edge detail information of the image. In setting the weight, in addition to considering the straight line length, the present invention also considers the gradient amplitude and the gradient correction parameter, which can avoid the problem of local blur in the filtered image.

[0098] On the other hand, the present invention also provides a method for positioning special equipment practitioners by image processing, comprising:

[0099] S1, inputting the employee number of the special equipment practitioner through the input module, and transmitting the employee number to the communication module;

[0100] S2, forwarding the employee number to the RFID reader through a communication module;

[0101] S3, determining, by an RFID reader, whether there is an RFID tag corresponding to the employee number within its communication range, and if so, generating a feedback message by the RFID reader, and transmitting the feedback message to the communication module;

[0102] S4, transmitting the feedback message to the positioning module through the communication module;

[0103] S5, determining the position coordinates of the special equipment employee corresponding to the employee number according to the feedback message through the positioning module, and sending the position information to the confirmation module;

[0104] S6, using image processing technology to confirm whether the equipment operator is at the position coordinates through the confirmation module to obtain a confirmation result.

[0105] It should be noted that this method is implemented based on the aforementioned system, and each module in the method corresponds to the aforementioned device and can implement different implementation methods in the aforementioned device. For details, please refer to the above description of the device, which will not be described in detail here.

[0106] Although embodiments of the present invention have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A special equipment worker positioning system using image processing, characterized in that: It includes an input module, a communication module, an RFID reader, an RFID tag, a positioning module and a confirmation module; The input module is used to input the employee number of the special equipment practitioner and transmit the employee number to the communication module; The communication module is used to forward the employee number to the RFID reader; The RFID reader is used to determine whether there is an RFID tag corresponding to the employee number within its communication range. If so, the RFID reader generates a feedback message and transmits the feedback message to the communication module; The communication module is also used to transmit the feedback message to the positioning module; The positioning module is used to determine the position coordinates of the special equipment practitioner corresponding to the employee number according to the feedback message, and send the position coordinates to the confirmation module; The confirmation module is used to use image processing technology to confirm whether the equipment practitioner is at the position coordinates to obtain a confirmation result; The confirmation module includes an image acquisition unit, an image recognition unit and a result display unit; The image acquisition unit is used to acquire facial images of all persons within a range with a radius R and a center at the position coordinate; The image recognition unit is used to determine whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number. If yes, then the confirmation result is that the special equipment practitioner corresponding to the employee number is at the position coordinates. If not, the confirmation result is that the special equipment practitioner corresponding to the employee number is not at the location coordinates; The result display unit is used to visually display the confirmation result; The determining whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number includes: Store the facial images of all the persons into a set Q; Obtain the first feature information of each facial image in the set Q respectively; Obtaining a reserved facial image of a special equipment practitioner corresponding to the employee number from a database, and obtaining second feature information of the reserved facial image; Matching the first feature information corresponding to each facial image in the set Q with the second feature information, if the match is successful, it indicates whether the facial images of all the persons include the facial image of the special equipment practitioner corresponding to the employee number; The step of respectively obtaining the first feature information of each facial image in the set Q includes: For the qth face image img in the set Q p , use the following method to obtain the first characteristic information contained therein: Img p Convert to grayscale image; Performing filtering on the grayscale image to obtain a filtered image; Using the Otsu algorithm to obtain a foreground image in the filtered image; Using the HOG feature extraction algorithm to obtain first feature information contained in the foreground image; The filtering process is performed on the grayscale image to obtain a filtered image, comprising: Perform edge detection on the grayscale image, obtain edge pixel points contained in the grayscale image, and store all edge pixel points into a set blU; store all pixel points in the grayscale image into a set N; The filtered image is filtered using iterative filtering: The first filtering process uses a preset filtering function to filter each edge pixel point included in blU to obtain the first filtered image img1; the edge pixel points included in blU are stored in the filtered pixel point set P; In the second filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q1, and the pixel points in the set Q1 that do not belong to the filtered pixel point set P are stored in the to-be-filtered set R2; in the first filtered image img1, a preset filtering function is used to filter each edge pixel point included in the to-be-filtered set R2 to obtain the second filtered image img2, and the pixel points included in the to-be-filtered set R2 are stored in the filtered pixel point set P; In the mth filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q m-1 In the above example, we will set Q m-1 The pixels that do not belong to the filtered pixel set P are stored in the to-be-filtered set R m In the m-1th filtering image img m-1 The preset filter function is used to treat the filter set R m Filter each edge pixel contained in to obtain the mth filtered image img m , the set to be filtered R m The pixels contained in are stored in the filtered pixel set P; The condition for the iteration to end is Q m-1 is an empty set; The preset filtering function includes: Among them, apix(c) represents the result of filtering the pixel point c, nuc represents the set of pixel points within a preset - sized window centered on c, lst(c,g) represents the straight - line length between c and g, zg 2 represents the variance of the straight - line lengths between the pixel points in nuc and c, z represents a preset constant parameter, noisdx represents the variance of noise estimation for the image to be filtered, ue(c) and ue(g) respectively represent the gradient magnitudes of c and g, xe(g) represents the gradient correction parameter. If ue(c)>bl(nuc), then xe(g)=ue ma -ue(g); if ue(c)<bl(nuc), then xe(g)=ue mi -ue(g); if ue(c)=bl(nuc), then xe(g)=0. bl(nuc) represents the mean of the gradient magnitudes of the pixel points included in nuc, ue ma and ue mi respectively represent the maximum and minimum of the gradient magnitudes of the pixel points included in nuc; pix(g) represents the pixel value of the pixel point g, nfnuc represents the number of pixel points included in nuc, gd represents a preset control parameter, pl(nuc) represents the mean of the pixel values of the pixel points included in nuc.

2. The image processing system for positioning personnel of special equipment according to claim 1, characterized in that: The communication module includes a wireless cellular communication network or an optical fiber communication network; The wireless cellular communication network or the optical fiber communication network is used to transmit the employee number to the RFID reader; The wireless cellular communication network or the optical fiber communication network is also used to transmit the feedback message to the positioning module.

3. The image processing special equipment practitioner positioning system according to claim 2, characterized in that: The RFID tag is an active tag, and the RFID tag broadcasts a preset positioning message in a fixed time period; The positioning message includes the employee number of the special equipment practitioner corresponding to the RFID tag, the ID of the RFID tag, and the remaining power of the RFID tag.

4. The image processing system for positioning personnel of special equipment according to claim 3, characterized in that: Determining whether there is an RFID tag corresponding to the employee number within the communication range includes: The RFID reader matches the employee number with the employee numbers contained in all positioning messages received within a preset time period. If the match is successful, it means that there is an RFID tag corresponding to the employee number within the communication range of the RFID reader. If the match fails, it means that there is no RFID tag corresponding to the employee number within the communication range of the RFID reader.

5. The image processing system for positioning personnel of special equipment according to claim 1, characterized in that: The feedback message includes the remaining power of the RFID tag corresponding to the employee number and the ID of the RFID reader.

6. The image processing system for positioning personnel of special equipment according to claim 5, characterized in that: The determining, according to the feedback message, the position coordinates of the special equipment practitioner corresponding to the employee number includes: Acquire the location coordinates of the RFID reader from an RFID location database according to the ID of the RFID reader; The location coordinates are used as the location coordinates of the special equipment practitioner corresponding to the employee number.

7. A method for locating special equipment practitioners using image processing, characterized in that: include: S1, inputting the employee number of the special equipment practitioner through the input module, and transmitting the employee number to the communication module; S2, forwarding the employee number to the RFID reader through the communication module; S3, determining through the RFID reader whether there is an RFID tag corresponding to the employee number within its communication range, and if so, generating a feedback message through the RFID reader and transmitting the feedback message to the communication module; S4, transmitting the feedback message to the positioning module through the communication module; S5, determining the position coordinates of the special equipment employee corresponding to the employee number according to the feedback message through the positioning module, and sending the position coordinates to the confirmation module; S6, using an image processing technique to confirm whether the equipment operator is at the position coordinates by a confirmation module, and obtaining a confirmation result; The confirmation module includes an image acquisition unit, an image recognition unit and a result display unit; The image acquisition unit is used to acquire facial images of all persons within a range with a radius R and a center at the position coordinate; The image recognition unit is used to determine whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number. If yes, then the confirmation result is that the special equipment practitioner corresponding to the employee number is at the position coordinates. If not, the confirmation result is that the special equipment practitioner corresponding to the employee number is not at the location coordinates; The result display unit is used to visually display the confirmation result; The determining whether the facial images of all the personnel include the facial image of the special equipment practitioner corresponding to the employee number includes: Store the facial images of all the persons into a set Q; Obtain the first feature information of each facial image in the set Q respectively; Obtaining a reserved facial image of a special equipment practitioner corresponding to the employee number from a database, and obtaining second feature information of the reserved facial image; Matching the first feature information corresponding to each facial image in the set Q with the second feature information, if the match is successful, it indicates whether the facial images of all the persons include the facial image of the special equipment practitioner corresponding to the employee number; The step of respectively obtaining the first feature information of each facial image in the set Q includes: For the qth face image img in the set Q p , use the following method to obtain the first characteristic information contained therein: Img p Convert to grayscale image; Performing filtering on the grayscale image to obtain a filtered image; Using the Otsu algorithm to obtain a foreground image in the filtered image; Using the HOG feature extraction algorithm to obtain first feature information contained in the foreground image; The filtering process is performed on the grayscale image to obtain a filtered image, comprising: Perform edge detection on the grayscale image, obtain edge pixel points contained in the grayscale image, and store all edge pixel points into a set blU; store all pixel points in the grayscale image into a set N; The filtered image is filtered using iterative filtering: The first filtering process uses a preset filtering function to filter each edge pixel point included in blU to obtain the first filtered image img1; the edge pixel points included in blU are stored in the filtered pixel point set P; In the second filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q1, and the pixel points in the set Q1 that do not belong to the filtered pixel point set P are stored in the to-be-filtered set R2; in the first filtered image img1, a preset filtering function is used to filter each edge pixel point included in the to-be-filtered set R2 to obtain the second filtered image img2, and the pixel points included in the to-be-filtered set R2 are stored in the filtered pixel point set P; In the mth filtering process, the pixel points of the 8-neighborhood of each pixel point in the filtered set P are stored in the set Q m-1 In the above example, we will set Q m-1 The pixels that do not belong to the filtered pixel set P are stored in the to-be-filtered set R m In the m-1th filtering image img m-1 The preset filter function is used to treat the filter set R m Filter each edge pixel contained in to obtain the mth filtered image img m , the set to be filtered R m The pixels contained in are stored in the filtered pixel set P; The condition for the iteration to end is Q m-1 is an empty set; The preset filtering function includes: Where apix(c) represents the result of filtering pixel c, nuc represents the set of pixels in a window of preset size centered on c, lst(c,g) represents the length of the straight line between c and g, and zg 2 represents the variance of the straight line length between the pixel point in nuc and c, z represents a preset constant parameter, noisdx represents the variance of noise estimation for the image to be filtered, ue(c) and ue(g) respectively represent the gradient magnitudes of c and g, xe(g) represents the gradient correction parameter. If ue(c) > bl(nuc), then xe(g) = ue ma -ue(g). If ue(c) < bl(nuc), then xe(g) = ue mi -ue(g). If ue(c) = bl(nuc), then xe(g) = 0. bl(nuc) represents the mean of the gradient magnitudes of the pixel points included in nuc, ue ma and ue mi respectively represent the maximum and minimum of the gradient magnitudes of the pixel points included in nuc; pix(g) represents the pixel value of pixel point g, nfnuc represents the number of pixel points included in nuc, gd represents a preset control parameter, and pl(nuc) represents the mean of the pixel values of the pixel points included in nuc.

Images