Warehouse-out intelligent identification method for electric power safety tool

Abstract

The invention provides a warehouse-out intelligent identification method for an electric power safety tool, and the method comprises the steps: collecting an image of the electric power safety tool, and carrying out the preprocessing of the image; performing shape recognition on the electric power safety tool according to the preprocessed image, and matching a shape recognition result with shape configuration in a preset template library to obtain a matching result; and according to the corresponding matching result, carrying out statistics on a warehouse-out result of the corresponding electric power safety tool. According to the invention, the automatic and intelligent management of the warehouse-out process of the safety tools is realized, so that the registration deviation is eliminated, the management efficiency is improved, and the electric power operation safety is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power system automation, and in particular to an intelligent identification method for power safety tools when leaving a warehouse. Background Art

[0002] In the daily operation of the power industry, the standardized management of safety tools is a key factor in ensuring the safety and efficiency of power operations. However, the traditional management method of the use of safety tools mostly relies on manual self-registration, which exposes many deficiencies and hidden dangers in actual operation. First, there is an obvious registration deviation problem in manual registration. Due to the interference of human factors, such as negligence and memory errors, the use records of safety tools often do not match the actual use. This deviation not only affects the management of safety tools, but is also likely to cause safety hazards due to inaccurate information, endangering the safe operation of power equipment. Secondly, the manual registration method also has obvious shortcomings in efficiency. With the continuous expansion of the scale of the power industry and the increase in the frequency of operations, the number of safety tools is also increasing, and the frequency of use is also increasing. The traditional manual registration method can no longer meet the needs of efficient and real-time management, resulting in increased management costs and difficult to ensure management results. In addition, manual registration also faces the problem of difficult information traceability. Once safety tools are lost, damaged or misused, it is often difficult to quickly trace and identify the cause due to incomplete or inaccurate recorded information, thus delaying the handling and resolution of the problem. Summary of the invention

[0003] The purpose of the present invention is to propose an intelligent identification method for power safety tools to solve the technical problems of how to eliminate registration deviations, improve management efficiency and ensure the safety of power operations.

[0004] On the one hand, a method for intelligent identification of power safety tools is provided, comprising:

[0005] Collect images of power safety tools and pre-process the images;

[0006] Performing shape recognition on the power safety tool according to the preprocessed image, and matching the shape recognition result with the shape configuration in the preset template library to obtain a matching result;

[0007] The corresponding power safety tool outbound results are counted according to the corresponding matching results.

[0008] Preferably, the preprocessing of the image includes sequentially performing image graying, filtering, binarization, contour extraction, background separation and image correction on the image of the power safety tool.

[0009] Preferably, it further includes grayscale processing the image of the electric power safety tool according to the following formula: Gray = 0.299R + 0.587G + 0.114B

[0010] where Gray represents the grayscale value, and R, G, and B respectively represent the values of the red, green, and blue channels of the pixel points in the image of the electric power safety tool;

[0011] Filter the image of the electric power safety tool according to the following formula

[0012]

[0013] where x and y represent the two-dimensional coordinates on the image plane of the electric power safety tool, σ is the standard deviation of the Gaussian distribution, and π is the pi;

[0014] Perform binarization processing on the image of the electric power safety tool according to the following formula

[0015]

[0016] where src(x, y) represents the grayscale value of the pixel points in the image of the electric power safety tool, dst(x, y) represents the value of the corresponding pixel points in the binarized image of the electric power safety tool, and T represents the threshold;

[0017] Perform contour extraction processing on the image of the electric power safety tool according to the following formula

[0018]

[0019] where x and y represent the centroid coordinates, μ ij represents the second-order central moment, (x - x c ) and (y - y c ) represent the coordinate offsets of the points on the contour relative to the centroid;

[0020] Perform background separation processing on the image of the electric power safety tool according to the following formula

[0021]

[0022] where X t represents the observed value of the pixel point at time, K is the number of Gaussian distributions, w k is the weight of the k-th Gaussian distribution, μ k is the mean vector of the k-th Gaussian distribution, and ∑ k is the covariance matrix of the k-th Gaussian distribution;

[0023] Perform image correction processing on the image of the electric power safety tool according to the following formula

[0024]

[0025] Among them, (x, y) are the pixel coordinates in the original image, and (x 1 , y 1 ) are the pixel coordinates in the target image after affine transformation, and a and b respectively represent the image correction parameters.

[0026] Preferably, the shape recognition of the electrical safety tool based on the preprocessed image includes

[0027] calculating the contour area and perimeter of the image of the electrical safety tool, and calculating the area enclosed by the contour, the circumscribed circle of the contour, the circumscribed rectangle of the contour, and the minimum area rectangle of the contour to obtain various data parameters of the contour;

[0028] performing shape recognition based on various data parameters of the contour and performing barcode recognition.

[0029] Preferably, it further includes calculating the contour area according to the following formula

[0030]

[0031] where (x i , y i ) represent the coordinates of the polygon vertices, n represents the number of polygon vertices, A represents the calculated result as the contour area, and i represents the ordinal number of the coordinate points.

[0032] Preferably, it further includes calculating the contour perimeter according to the following formula

[0033]

[0034] where (x i , y i ) and (x i + 1, y i + 1) represent the coordinates of two adjacent vertices on the contour, n is the number of vertices of the contour, P represents the calculated result as the contour perimeter, and i represents the ordinal number of the coordinate points.

[0035] Preferably, it further includes calculating the area enclosed by the contour according to the following formula

[0036]

[0037]

[0038] where A is the area enclosed by the contour, (x i , y i ) represent the coordinates of the polygon vertices, △A i represents the point (x i, y i ) The related small - area elements, where i represents the ordinal number of the coordinate point.

[0039] Preferably, it further includes determining the maximum and minimum values of the contour points in the x and y directions, and calculating the corresponding circum - circle of the contour, circum - rectangle of the contour, and minimum - area rectangle of the contour according to the maximum and minimum values.

[0040] Preferably, it further includes determining the maximum and minimum values of the contour points in the x and y directions according to the following formula:

[0041] w = x max -x min , h = y max -y min

[0042] x min = min(x i ), x max = max(x i )

[0043] y min = min(y i ), y max = max(y i )

[0044] Among them, (x min , y min ) represents the coordinates of the upper - left vertex of the circum - rectangle, (x max , y max ) represents the coordinates of the lower - right vertex of the circum - rectangle, w represents the width of the rectangle, and h represents the height of the rectangle.

[0045] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0046] The intelligent identification method for the outbound of power safety tools provided by the present invention realizes the automated and intelligent management of the outbound process of safety tools, thereby eliminating registration deviations, improving management efficiency, and ensuring the safety of power operations. Description of the Drawings

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, obtaining other drawings without creative efforts still belongs to the scope of the present invention.

[0048] Figure 1 It is a schematic diagram of the main process of an intelligent identification method for the outbound of power safety tools in an embodiment of the present invention. Specific Embodiment

[0049] To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] As Figure 1 shown, it is a schematic diagram of an embodiment of an intelligent recognition method for the outbound of electric power safety tools provided by the present invention. In this embodiment, the method includes the following steps:

[0051] Step S1, collect images of electric power safety tools and preprocess the images; it is understandable that the user issues instructions through the touch screen of the user terminal. The user terminal issues corresponding instruction operations to the camera, and the camera collects images according to the instructions and transmits them back to the user terminal. The user terminal receives the image data transmitted by the camera and performs analysis and processing. It should be noted that this method can be implemented based on a corresponding host, and the host may include a touch screen, a camera, and a user terminal. Among them, the host is used to deploy the communication between the user terminal and the camera to realize the automated and intelligent management of the outbound process of electric power safety tools. The touch screen is used for the interaction between the user and the user terminal. The camera collects images of the wall where the electric power safety tools are stored according to the instructions issued by the user terminal. The user terminal provides a convenient operation interface, allowing the user to quickly issue corresponding instructions through the interface.

[0052] In one embodiment, the preprocessing of the images includes sequentially performing image grayscale conversion, filtering processing, binarization processing, contour extraction processing, background separation processing, and image correction processing on the images of the electric power safety tools. The image grayscale conversion of the images of the electric power safety tools is performed according to the following formula,

[0053] Gray = 0.299R + 0.587G + 0.114B

[0054] where Gray represents the grayscale value, which represents the brightness of the corresponding pixel point in the converted grayscale image. R, G, and B respectively represent the values of the red, green, and blue channels of the pixel points in the images of the electric power safety tools; the weighting coefficients (0.299, 0.587, 0.114), and these coefficients are determined according to the sensitivity of the human eye to different colors.

[0055] The filtering processing of the images of the electric power safety tools is performed according to the following formula,

[0056]

[0057] Among them, x and y represent two-dimensional coordinates on the image plane of the electrical safety tool, σ is the standard deviation of the Gaussian distribution, which determines the shape and distribution width of the Gaussian function. π is the ratio of the circumference of a circle to its diameter, playing a fixed mathematical ratio role in the calculation of the Gaussian function to ensure the normalization of the function and the correct probability distribution characteristics, and e represents the natural constant, approximately equal to 2.71828;

[0058] Perform binarization processing on the image of the electrical safety tool according to the following formula,

[0059]

[0060] Among them, src(x, y) represents the grayscale value of the pixel point in the image of the electrical safety tool, dst(x, y) represents the value of the corresponding pixel point in the binarized image of the electrical safety tool, and T represents the threshold, which is a key parameter in the binarization process;

[0061] Perform contour extraction processing on the image of the electrical safety tool according to the following formula,

[0062]

[0063] Among them, x and y represent the centroid coordinates, μ ij represents the second-order central moment, which is used to calculate the Hu moment, (x - x c ) and (y - y c ) represent the coordinate offsets of the points on the contour relative to the centroid;

[0064] Perform background separation processing on the image of the electrical safety tool according to the following formula,

[0065]

[0066] Among them, X t represents the observed value of the pixel point at time, and in the image background subtraction scenario, it is usually the color information of the pixel point. K is the number of Gaussian distributions, which is used to describe the complexity of the color distribution of each pixel point. w k is the weight of the k-th Gaussian distribution, μ k is the mean vector of the k-th Gaussian distribution, and ∑ k is the covariance matrix of the k-th Gaussian distribution;

[0067] Perform image correction processing on the image of the electrical safety tool according to the following formula,

[0068]

[0069] Among them, (x, y) are the pixel point coordinates in the original image, (x 1 , y 1) are the pixel coordinates in the target image after affine transformation, where a and b represent the image correction parameters respectively.

[0070] Step S2: Perform shape recognition on the power safety tool according to the preprocessed image, and match the shape recognition result with the shape configuration in the preset template library to obtain a matching result. It can be understood that the analyzed image data is used for contour recognition to locate the item type, and the analyzed image data is used for barcode location recognition.

[0071] It should be noted that barcode recognition first performs edge detection on the preprocessed image. The essence of problems such as inaccurate barcode recognition and failure to recognize is the problem of edge detection. In traditional algorithms, barcode edges are generally detected based on the zero crossing of the second derivative. Due to a certain degree of blurring at the barcode image edge during image acquisition and image filtering. Also, when denoising the barcode image, the classic median filtering algorithm is mostly used. This algorithm has obvious denoising effects on salt-and-pepper noise, but poor denoising effects on Gaussian noise. In actual applications, the blurring generated during image acquisition is basically Gaussian noise, and the median filtering method cannot effectively remove this noise. Detecting barcode edges using the zero crossing of the second derivative in the case of edge blurring will result in a deviation of 1 - 2 pixels, which will affect the determination of the bar width of the barcode and thus affect recognition. Therefore, edge blurring is an important factor affecting barcode recognition.

[0072] For a pixel point (x, y), the approximate calculation formula for the gradient in the horizontal direction (x - axis direction):

[0073] G x (x,y) = f(x + 1,y - 1)+2f(x + 1,y)+f(x + 1,y + 1)-f(x - 1,y - 1)-2f(x - 1,y) Formula One

[0074] The approximate calculation formula for the gradient in the vertical direction (y - axis direction):

[0075] G y (x, y) = f(x - 1,y + 1)+2f(x,y + 1)+f(x + 1,y + 1)-f(x - 1,y - 1)-2f(x, y - 1) Formula Two

[0076] Where G x (x,y) and G y (x,y) are the gradient values of the pixel point (x, y) in the horizontal and vertical directions respectively. These gradient values are used to measure the degree of gray - level change of the image at this point. f(x, y) is the gray - level value of the pixel point in the original image. It is the basic data for calculating the gradient.

[0077] To solve the problem of edge blurring, it is necessary to analyze the noise. Smooth images are mainly concentrated in the mid- and low-frequency parts, while noise information or edge information is mainly concentrated in the high-frequency band. Usually, filtering methods indiscriminately filter out high-frequency information, inevitably filtering out edge information as well, resulting in blurred image edges. Utilizing the good time-frequency characteristics of wavelet transform, as long as the local maximum points of the modulus of wavelet transform coefficients are detected along the gradient direction, the edge points of the image can be obtained. However, the modulus maximum points of wavelet coefficients may also correspond to noise points. Therefore, after detecting the maximum values of wavelet coefficients, it is also necessary to remove the wavelet coefficients corresponding to noise points. The Lipschitz exponent of noise is negative, while that of the signal is positive. As the scale increases, the signal coefficients gradually become larger, while the noise coefficients gradually become smaller. According to this characteristic, the false edges caused by noise can be removed, and the modulus maximum points of non-noise points are connected to obtain the edge of the image. This method can not only achieve the purpose of denoising but also preserve the edges of the image, solving the problem of image edge blurring caused by denoising.

[0078] After the edge detection is processed, barcode recognition is performed. Fourier transform is used to assist in the recognition of barcodes in a complex image environment.

[0079]

[0080] Where M and N are the dimensions of the image (number of rows and columns), and i is the imaginary unit.

[0081] In one embodiment, the shape recognition of the electrical safety tool based on the preprocessed image includes calculating the contour area and perimeter of the image of the electrical safety tool, and calculating the area enclosed by the contour, the circumscribed circle of the contour, the circumscribed rectangle of the contour, and the minimum area rectangle of the contour to obtain various data parameters of the contour; performing shape recognition based on the various data parameters of the contour and performing barcode recognition.

[0082] Specifically, the contour area is calculated according to the following formula

[0083]

[0084] Where (x i , y i ) represents the coordinates of the vertices of the polygon, n represents the number of vertices of the polygon, A represents the calculated result which is the contour area, and i represents the ordinal number of the coordinate points.

[0085] The contour perimeter is calculated according to the following formula

[0086]

[0087] Where (x i , y i ) and (xi +1, y i +1) represents the coordinates of two adjacent vertices on the contour, n is the number of vertices of the contour, P represents that the calculated result is the perimeter of the contour, and i represents the ordinal number of the coordinate point.

[0088] After calculating the contour area and perimeter, calculate the convex hull. Assume that the point set P = {p 1 , p 2 , …, p n} is given on the plane, where p i = (x i , y i ). Then, calculate the centroid of the contour.

[0089] Calculate the area enclosed by the contour according to the following formula:

[0090]

[0091] where A is the area enclosed by the contour, (x i , y i ) represents the coordinates of the polygon vertices, △A i represents the small area element related to the point (x i , y i ), and i represents the ordinal number of the coordinate point.

[0092] For an embodiment, then calculate the circumcircle of the contour, the circumscribed rectangle of the contour, and the minimum area rectangle of the contour. Calculation method: Similarly, find the maximum and minimum values of the contour points in the and directions. Determine the maximum and minimum values of the contour points in the and directions, and calculate the corresponding circumcircle of the contour, the circumscribed rectangle of the contour, and the minimum area rectangle of the contour according to the maximum and minimum values. Among them, determine the maximum and minimum values of the contour points in the and directions according to the following formula:

[0093] w = x max - x min , h = y max - y min

[0094] x min = min(x i ), x max = max(x i )

[0095] y min = min(y i ), y max = max(y i )

[0096] where (x min , y min) represents the coordinates of the upper left vertex of the circumscribed rectangle, (x max , y max ) represents the coordinates of the lower right vertex of the circumscribed rectangle, w represents the width of the rectangle, and h represents the height of the rectangle.

[0097] Finally, shape recognition is performed based on the various data of the contour calculated in the previous steps, and the recognized result is matched with the shapes in the pre-set template library.

[0098]

[0099] Among them, Z(n) is the coefficient after discrete Fourier transform, which represents the characteristics of the contour in the frequency domain. (x(k), y(k)) are the coordinates of the contour points, and z(k) represents the coordinates in complex form for convenient Fourier transform.

[0100] Step S3, count the outbound results of the corresponding electrical safety tools according to the corresponding matching results. It can be understood that the contour recognition result and the barcode recognition result are used for double verification; if the verification passes, the outbound record and outbound information are confirmed; if the verification fails, an alarm prompt is issued for manual review.

[0101] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0102] The intelligent recognition method for outbound of electrical safety tools provided by the present invention realizes the automated and intelligent management of the outbound process of safety tools, thereby eliminating registration deviations, improving management efficiency, and ensuring the safety of electrical operations.

[0103] The above-disclosed are only the preferred embodiments of the present invention. Of course, the scope of the rights of the present invention cannot be limited by this. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A method for intelligent identification of power safety tools before they leave the warehouse, characterized in that: include: Collect images of power safety tools and pre-process the images; Performing shape recognition on the power safety tool according to the preprocessed image, and matching the shape recognition result with the shape configuration in the preset template library to obtain a matching result; The corresponding power safety tool outbound results are counted according to the corresponding matching results.

2. The method according to claim 1, characterized in that The preprocessing of the image includes sequentially performing image graying, filtering, binarization, contour extraction, background separation and image correction on the image of the power safety tool.

3. The method according to claim 2, characterized in that It also includes graying the image of the power safety tool according to the following formula: Gray=0.299R+0.587G+0.114B Among them, Gray represents the gray value, R, G, and B represent the values ​​of the red, green, and blue channels of the pixel points in the image of the power safety tool, respectively; The image of the power safety tool is filtered according to the following formula: Wherein, x and y represent the two-dimensional coordinates on the image plane of the power safety tool, σ is the standard deviation of the Gaussian distribution, π is the circumference of a circle, and e represents a natural constant; The image of the power safety tool is binarized according to the following formula: Wherein, src(x,y) represents the grayscale value of the pixel in the image of the power safety tool, dst(x,y) represents the value of the corresponding pixel in the image of the power safety tool after binarization, and T represents the threshold value; The image of the power safety tool is processed for contour extraction according to the following formula: Where x and y represent the coordinates of the center of mass, μ ij represents the second-order central moment, (xx c ) and (yy c ) represents the coordinate offset of the point on the contour relative to the centroid; The image of the power safety tool is processed for background separation according to the following formula: Among them, X t represents the observed value of the pixel at time, K is the number of Gaussian distributions, and w k is the weight of the kth Gaussian distribution, μ k is the mean vector of the kth Gaussian distribution, ∑ k is the covariance matrix of the kth Gaussian distribution; The image of the power safety tool is corrected according to the following formula: Among them, (x, y) is the pixel coordinates in the original image, (x 1 ,y 1 ) is the pixel coordinate in the target image after affine transformation, and a and b represent the image correction parameters respectively.

4. The method according to claim 1, characterized in that The performing shape recognition on the power safety tool according to the preprocessed image comprises: Calculate the contour area and perimeter of the image of the power safety tool, and calculate the area enclosed by the contour, the circumscribed circle of the contour, the circumscribed rectangle of the contour and the rectangle with the minimum area of ​​the contour to obtain various data parameters of the contour; Shape recognition is performed based on various data parameters of the contour, and barcode recognition is performed.

5. The method according to claim 4, characterized in that It also includes calculating the contour area according to the following formula, Among them, (x i ,y i ) represents the coordinates of the polygon vertices, n represents the number of vertices of the polygon, A represents that the calculated result is the area of ​​the contour, and i represents the ordinal number of the coordinate point.

6. The method according to claim 4, characterized in that It also includes calculating the contour perimeter according to the following formula, Among them, (x i ,y i ) and (x i +1,y i +1) represents the coordinates of two adjacent vertices on the contour, n is the number of vertices of the contour, P indicates that the calculated result is the perimeter of the contour, and i represents the ordinal number of the coordinate point.

7. The method according to claim 4, characterized in that It also includes calculating the area enclosed by the contour according to the following formula, Where A is the area enclosed by the contour, (x i ,y i ) represents the coordinates of the polygon vertices, △A i Represents the point (x i ,y i ) related small area elements, i represents the ordinal number of the coordinate point.

8. The method according to claim 4, characterized in that The method also includes determining the maximum and minimum values ​​of the contour points in the and directions, and calculating the corresponding contour circumscribed circle, contour circumscribed rectangle and contour minimum area rectangle according to the maximum and minimum values.

9. The method according to claim 8, characterized in that It also includes determining the maximum and minimum values ​​of the contour points in the and directions according to the following formulas: w=x max -x min ,h=y max -y min x min =min(x i ),x max =max(x i ) and min =min(and i ),and max =max(y i ) Among them, (x min ,y min ) represents the coordinates of the upper left corner of the circumscribed rectangle, (x max ,y max ) represents the coordinates of the lower right corner of the circumscribed rectangle, w represents the width of the rectangle, and h represents the height of the rectangle.

Images