An image processing method, a base station feeder detection method and related devices

Abstract

Embodiments of the present invention disclose an image processing method, a base station feeder detection method and related devices. The feeder images of the base stations to be detected collected by the image acquisition terminal are stored corresponding to the identifiers of the corresponding base stations to be detected. In this way, the image processing terminal can search for and obtain the target feeder image according to the detection request, and convert it into a characteristic grayscale image after performing image preprocessing operations on the target feeder image; finally, the obtained characteristic grayscale image is compared with the standard feeder characteristic grayscale image, and it is determined whether the feeder of the target detection base station is installed correctly according to the comparison result. Thus, almost automatic detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical fields of communication and image processing, and in particular, to an image processing method, a base station feeder detection method, and related devices. Background Art

[0002] The base stations of the fifth-generation mobile communication technology (5G) are the core devices of the 5G network, providing wireless coverage and realizing wireless signal transmission between the wired communication network and wireless terminals. The architecture and form of the base stations directly affect how the 5G network is deployed. In technical standards, the frequency band of 5G is much higher than that of other existing networks. At present, the 5G network mainly operates in the frequency band of 3000 - 5000 MHz. Since the higher the frequency, the greater the attenuation during signal propagation, the base station density of the 5G network will be higher. Currently, the total number of base stations has exceeded 700,000.

[0003] To ensure the normal operation of the base station, it is necessary to frequently detect the installed feeders, which are generally distinguished by the number of feeder marking points. The installation requirements of the feeders are as Figure 1 shown: the number of feeder marking points increases or decreases from left to right. When the order requirement is not met, it is an incorrect installation.

[0004] Currently, the high-density base station layout has increased the workload of base station feeder detection. Moreover, considering that base stations are often set on the top of buildings, the work of detecting whether the installation of base station feeders is accurate generally uses the method of manual climbing and visual inspection, which consumes manpower and time and has certain potential safety hazards for personnel. At the same time, the existing detection scheme relies entirely on manual visual inspection, which is too subjective, easily leading to inaccurate detection results and low detection efficiency. Summary of the Invention

[0005] An object of one or more embodiments of this specification is to provide an image processing method, a base station feeder detection method, and related devices, which can automatically detect whether the installation of base station feeders is correct by using the feeder detection algorithm set based on image processing for the images of base station feeders collected manually or by drones during installation, so as to reduce the detection cost and potential safety hazards and improve the detection accuracy and efficiency.

[0006] To solve the above technical problems, one or more embodiments of this specification are implemented as follows:

[0007] In a first aspect, an image processing method is provided, including: based on the identifier of the target detection base station in the current detection request, searching for and obtaining a matching target feeder image, where the target feeder image is obtained by an image acquisition terminal collecting the feeder image of the base station to be detected and storing the feeder image corresponding to the identifier of the current base station to be detected; performing an image preprocessing operation on the target feeder image and converting the target feeder image after the image preprocessing operation into a feature grayscale image;

[0008] comparing the feature grayscale image with the standard feeder feature grayscale image in a preset feeder standard library, and determining whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0009] In a second aspect, an image processing apparatus is proposed, including: an acquisition module for searching for and obtaining a matching target feeder image based on the identifier of the target detection base station in the current detection request, where the target feeder image is obtained by an image acquisition terminal collecting the feeder image of the base station to be detected and storing the feeder image corresponding to the identifier of the current base station to be detected; a conversion module for performing an image preprocessing operation on the target feeder image and converting the target feeder image after the image preprocessing operation into a feature grayscale image; a comparison module for comparing the feature grayscale image with the standard feeder feature grayscale image in a preset feeder standard library, and determining whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0010] In a third aspect, a base station feeder detection method is proposed, including: an image acquisition terminal collecting the feeder image of the base station to be detected and storing the feeder image corresponding to the identifier of the current base station to be detected; an image processing terminal searching for and obtaining a matching target feeder image based on the identifier of the target detection base station in the current detection request; the image processing terminal performing an image preprocessing operation on the target feeder image and converting the target feeder image after the image preprocessing operation into a feature grayscale image; the image processing terminal comparing the feature grayscale image with the standard feeder feature grayscale image in a preset feeder standard library, and determining whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0011] Fourthly, a base station feeder detection system is proposed, including: an image acquisition terminal and an image processing terminal; wherein, the image acquisition terminal acquires the feeder image of the base station to be detected, and stores the feeder image corresponding to the identifier of the current base station to be detected; the image processing terminal searches for and acquires the target feeder image based on the identifier of the target detection base station in the current detection request; the image processing terminal performs image preprocessing operations on the target feeder image, and converts the target feeder image after the image preprocessing operation into a feature grayscale image; the image processing terminal compares the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library, and determines whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0012] Fifthly, an electronic device is proposed, including: a processor; and a memory arranged to store computer-executable instructions, and the executable instructions, when executed, cause the processor to execute the image processing method described above.

[0013] Sixthly, a storage medium is proposed for computer-readable storage, and the storage medium stores one or more programs, and when the one or more programs are executed by one or more processors, the image processing method described above is implemented.

[0014] As can be seen from the technical solutions provided by one or more embodiments of this specification above, the feeder image of the base station to be detected collected by the image acquisition terminal and the identifier of the corresponding base station to be detected are stored correspondingly. In this way, the image processing terminal can search for and acquire the target feeder image according to the detection request, and convert it into a feature grayscale image after performing image preprocessing operations on the target feeder image; finally, the obtained feature grayscale image is compared with the standard feeder feature grayscale image, and it is determined whether the feeder of the target detection base station is correctly installed according to the comparison result. Thus, almost automatic detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the following will briefly introduce the drawings required to be used in the description of one or more embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a schematic diagram of the correct installation result of the feeder of the base station involved in this specification.

[0017] Figure 2It is a schematic diagram of the scenario architecture applicable to the base station feeder detection solution provided by the embodiments of the present invention.

[0018] Figure 3a and Figure 3b are schematic diagrams of the steps of two base station feeder detection methods provided by the embodiments of the present invention.

[0019] Figure 4 is a schematic diagram of the method steps for the image processing terminal to perform image preprocessing operations on the target feeder image provided by the embodiments of the present invention.

[0020] Figure 5 is a schematic diagram of scaling processing for the target feeder image provided by the embodiments of the present invention.

[0021] Figure 6 is a schematic diagram of selecting feeder marking points and corresponding HSV values provided by the embodiments of the present invention.

[0022] Figure 7 is a schematic diagram of the target feeder image in HSV image format provided by the embodiments of the present invention.

[0023] Figure 8 is a schematic diagram of the steps for the image processing terminal to convert the target feeder image after image preprocessing operations into a feature grayscale image provided by the embodiments of the present invention.

[0024] Figure 9a is a schematic diagram of dilation and erosion operations provided by the embodiments of the present invention.

[0025] Figure 9b is a schematic diagram of the feeder image after being processed by dilation and erosion operations provided by the embodiments of the present invention.

[0026] Figure 10a is a schematic diagram of the comparison process of step 308 provided by the embodiments of the present invention.

[0027] Figure 10b is a detection flow chart provided by the embodiments of the present invention.

[0028] Figure 11a and Figure 11b are respectively schematic diagrams of the structure of the base station feeder detection system 1100 provided by the embodiments of the present invention.

[0029] Figure 12 is a schematic diagram of the steps of an image processing method provided by the embodiments of the present invention.

[0030] Figure 13 is a schematic diagram of the structure of an image processing device provided by the embodiments of the present invention. Detailed implementation manners

[0031] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in one or more embodiments of this specification. Obviously, the described one or more embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on one or more embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this document.

[0032] Considering the unreliable and unsafe problems existing in the feeder detection of base stations in the prior art, the embodiments of this specification propose a new feeder detection scheme. The feeder image of the base station to be detected collected by the image acquisition terminal is sent to the cloud server, and the cloud server stores the feeder image and the identifier of the corresponding base station to be detected. In this way, the image processing terminal can search and obtain the target feeder image from the cloud server according to the detection request, and convert it into a feature grayscale image after performing image preprocessing operations on the target feeder image; finally, the obtained feature grayscale image is compared with the standard feeder feature grayscale image, and it is determined whether the feeder of the target detection base station is correctly installed according to the comparison result. Thus, almost automated detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0033] Before introducing the base station feeder detection scheme involved in the embodiments of this specification, the scenario architecture applicable to this base station feeder detection scheme will be briefly introduced first. Refer to Figure 2 As shown, the image acquisition terminal 202 is movable and is not installed near the base station to be detected. Generally, the image acquisition terminal 202 can be a handheld camera or a drone with image acquisition function. When there is a need for acquisition services or detection services, it is arranged to take pictures near the feeder of the base station to be detected. At the same time, it can also include a cloud server 204, which mainly realizes the centralized storage of the collected feeder images, is safe, efficient and can achieve large-capacity storage, facilitating access by the detection party. In addition, an image processing terminal 206 is also arranged, which is mainly used to obtain the feeder image from the cloud server and perform a series of image processing operations to realize the automatic detection of whether the feeder is correctly installed. It should be understood that the image processing terminal 206 cooperates remotely with the image acquisition terminal 202 and the cloud server 204, minimizing human participation as much as possible and reducing the detection error rate and potential safety hazards of detection. Among them, Figure 2 The shown architecture is an optimal system architecture. In specific implementation, the cloud server may not be included in the system architecture, and in this case, the storage function of the cloud server can be realized by the image processing terminal.

[0034] The following uses specific examples to elaborate on the base station feeder detection solution involved in this specification.

[0035] Embodiment 1

[0036] Referring to Figure 3a shown in the figure, it is a schematic diagram of the steps of a base station feeder detection method provided by an embodiment of this specification. The base station feeder detection method may include:

[0037] Step 302a: An image acquisition terminal acquires a feeder image of a base station to be detected, and stores the feeder image corresponding to the identifier of the current base station to be detected.

[0038] It should be understood that at this time, the image acquisition terminal sends the acquired feeder image and the identifier of the current base station to be detected to an image processing terminal or a cloud server for corresponding storage.

[0039] Step 304a: Based on the identifier of the target detection base station in the current detection request, the image processing terminal searches for and acquires a matching target feeder image.

[0040] After that, based on the identifier of the target detection base station in the current detection request, the image processing terminal searches for and acquires a matching target feeder image from the local or the cloud server.

[0041] Step 306a: The image processing terminal performs an image preprocessing operation on the target feeder image, and converts the target feeder image after the image preprocessing operation into a feature grayscale image.

[0042] Step 308a: The image processing terminal compares the feature grayscale image with the standard feeder feature grayscale image in a preset feeder standard library, and determines whether the feeder of the target detection base station is installed correctly according to the comparison result.

[0043] Furthermore, referring to Figure 3b shown in the figure, it is a schematic diagram of the steps of another base station feeder detection method provided by an embodiment of this specification, which is applied to the Figure 2 system architecture shown in the figure. The base station feeder detection method may include the following steps:

[0044] Step 302b: An image acquisition terminal acquires a feeder image of a base station to be detected, and sends the feeder image and the identifier of the current base station to be detected to a cloud server, so that the cloud server stores the feeder image corresponding to the identifier of the current base station to be detected.

[0045] The feeder image includes: a feeder picture or a feeder video. Among them, the identifier of the current base station to be detected can be the base station code generated by the image acquisition terminal according to the local positioning device as the identifier. It can also obtain the base station code as the identifier by scanning the base station barcode at the location where the feeder is located.

[0046] In the embodiments of this specification, the image acquisition terminal may be a contact-type handheld image capturing device, such as a handheld camera, a mobile phone, or a tablet computer, etc.; it may also be a non-contact image capturing device, such as a shooting drone, etc. Correspondingly, based on the specific type of the image acquisition terminal being different, the specific implementation of collecting the feeder image also varies. Specifically:

[0047] When the image acquisition terminal is a contact-type handheld image capturing device, the image acquisition terminal collects the feeder image of the base station to be detected, which can be specifically implemented as: the contact-type handheld image capturing device responds to the user's touch operation and captures the feeder image from the base station to be detected.

[0048] When the image acquisition terminal is a non-contact image capturing device, the image acquisition terminal collects the feeder image of the base station to be detected, which can be specifically implemented as: the non-contact image capturing device responds to the user's remote control and captures the feeder image from the base station to be detected.

[0049] Regardless of which method, the acquisition of the feeder image can be achieved, facilitating the subsequent automatic detection of whether the feeder is correctly installed. Preferably, when conditions permit, the feeder image should be collected by a non-contact image capturing device, which can completely avoid the subjective problems brought by manual data collection and avoid potential safety hazards. It can be seen that the method of collecting and detecting the feeder image in a non-contact manner is more objective, accurate, safe, and reliable.

[0050] Step 304b: The image processing terminal searches for and obtains a matching target feeder image from the cloud server based on the identifier of the target detection base station in the current detection request.

[0051] Among them, the current detection request may be initiated by the detecting party, and the identifier of the target detection base station is carried in this detection request. In addition, other information attached to this detection service may also be included, such as the location information and the location of origin information of the target detection base station, etc.

[0052] Step 306b: The image processing terminal performs an image preprocessing operation on the target feeder image and converts the target feeder image after the image preprocessing operation into a feature grayscale image.

[0053] In this step 306b, it can be specifically divided into two major parts: the image preprocessing operation and the grayscale image conversion operation.

[0054] -- Image preprocessing operation

[0055] The image processing terminal performs an image preprocessing operation on the target feeder image. Referring to Figure 4 as shown, specifically includes:

[0056] Step 402: Scale the target feeder image according to the detected pixel requirements.

[0057] Refer to Figure 5 As shown, after reading the feeder image obtained from the cloud server locally, without affecting the feeder detection, the image is scaled so that the feeder image meets the preset detected pixel requirements.

[0058] Step 404: Select feeder marker points from the scaled target feeder image and extract the HSV values of each feeder marker point.

[0059] Specifically refer to Figure 6 As shown, the feeder marker points can be selected from the feeder image by means of manual assisted touch or automatic marking, and the HSV values are extracted for each selected feeder marker point.

[0060] Step 406: Based on the HSV values of each extracted feeder marker point, convert the scaled target feeder image into the HSV image format.

[0061] At this time, the format of the target feeder image is still the RGB image format. Therefore, it is necessary to convert the target feeder image shown in Figure 6 into the HSV image format based on the HSV values of each feeder marker point. The converted image can be referred to Figure 7 .

[0062] After performing the above series of image preprocessing operations on the target feeder image, the specific recognition processing operation of the feeder in the obtained HSV target feeder image can be further performed.

[0063] --Grayscale image conversion operation

[0064] Refer to Figure 8 As shown, the image processing terminal converts the target feeder image after the image preprocessing operation into a feature grayscale image, which may specifically include:

[0065] Step 802: Use edge detection technology to process the converted HSV target feeder image to obtain the edge contour of the feeder marker point.

[0066] Edge detection technology is actually a kind of filtering, and different operators have different extraction effects. The common edge detection technologies at least include: Sobel edge detection operator, Laplacian edge detection operator, Canny edge detection operator. Among them,

[0067] The Sobel operator detection method has a good effect on processing images with gradual gray - scale changes and more noise. The Sobel operator is not very accurate in edge localization, and the edges of the image are more than one pixel. When the accuracy requirement is not very high, it is a relatively commonly used edge - detection method.

[0068] The Laplacian operator method is sensitive to noise, so it is rarely used to detect edges. Instead, it is used to determine whether the edge pixels belong to the bright area or the dark area of the image. The Laplacian of Gaussian operator is a second - order derivative operator that will produce a sharp zero - crossing at the edge. The Laplacian operator is isotropic and can sharpen boundaries and lines in any direction, without directionality. This is the biggest advantage of the Laplacian operator compared with other algorithms.

[0069] The Canny method is not easily affected by noise and can detect real weak edges. Its advantages are that two different thresholds are used to detect strong edges and weak edges respectively, and when the weak edges are connected to the strong edges, the weak edges are included in the output image. Its goal is to find an optimal edge. The definition of the optimal edge is as follows: good detection: the algorithm can mark as many actual edges in the image as possible; good localization: the marked edges should be as close as possible to the actual edges in the real image; minimum response: the edges in the image can only be marked once, and possible image noise should not be marked as edges.

[0070] Optionally, when the image - processing terminal processes the converted HSV target feeder image using the Canny edge - detection operator, it specifically includes:

[0071] Step 1: Use a Gaussian filter to smooth the converted HSV target feeder image and filter out noise.

[0072] In order to minimize the impact of noise on the edge - detection result as much as possible, it is necessary to filter out noise to prevent false detection caused by noise. To smooth the image, the Gaussian filter is convolved with the image, and this step will smooth the image to reduce the obvious noise impact on the edge detector.

[0073] Step 2: Calculate the gradient value and direction of each pixel point in the processed HSV target feeder image.

[0074] Step 3: Perform non - maximum suppression on each pixel point based on the calculated gradient intensity and direction.

[0075] Non-maximum suppression is an edge thinning technique, and the role of non-maximum suppression is to "thin" the edges. After calculating the gradient of the image, the edges extracted only based on the gradient values are still very blurred. Non-maximum suppression can help suppress all gradient values outside the local maximum to 0. The algorithm for performing non-maximum suppression on each pixel in the gradient image is: compare the gradient intensity of the current pixel with two pixels along the positive and negative gradient directions. If the gradient intensity of the current pixel is the largest compared to the other two pixels, then this pixel point is retained as an edge point; otherwise, this pixel point will be suppressed. Usually, for more accurate calculation, linear interpolation is used between two adjacent pixels across the gradient direction to obtain the pixel gradients to be compared.

[0076] Step 4, further optimize the edges using double-threshold edge detection: If the gradient value of an edge pixel is greater than the high threshold, then it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, then it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, then it is suppressed.

[0077] After applying non-maximum suppression, the remaining pixels can more accurately represent the actual edges in the image. However, there are still some edge pixels caused by noise and color changes. To solve these spurious responses, the edge pixels must be filtered with weak gradient values, and the edge pixels with high gradient values are retained, which can be achieved by selecting high and low thresholds. If the gradient value of an edge pixel is higher than the high threshold, then it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, then it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, then it will be suppressed. The choice of threshold depends on the content of the given input image.

[0078] Step 5, determine the edge contour of the final feeder marking points by suppressing isolated weak edge pixels.

[0079] The pixel points classified as strong edges have been determined as edges because they are extracted from the real edges in the image. However, for weak edge pixels, there will be some controversy because these pixels can be extracted from real edges or caused by noise or color changes. To obtain accurate results, the weak edges caused by the latter should be suppressed. Usually, the weak edge pixels caused by real edges will be connected to strong edge pixels, while the noise responses are not connected. To track edge connections, by looking at the weak edge pixels and their 8 neighboring pixels, as long as one of them is a strong edge pixel, then this weak edge point can be retained as a real edge.

[0080] Step 804: Perform median filtering on the HSV target feeder image after edge detection processing.

[0081] The median filtering method is a non-linear smoothing technique that sets the gray value of each pixel point to the median of the gray values of all pixel points within a certain neighborhood window of that point. Median filtering can overcome the blurring of image details caused by common linear filters such as box filters and mean filters under certain conditions. It is also very effective in filtering out pulse interference and image scanning noise. It is often used to protect edge information, and its characteristic of preserving edges makes it useful in situations where edge blurring is not desired. It is a very classic method for smoothing noise. When performing median filtering on an image, the image obtained by Canny edge detection will be further denoised to exclude smaller independent noise points in the image.

[0082] Step 806: Perform morphological operations of dilation or erosion on the edges of the feeder marker points in the HSV target feeder image after median filtering.

[0083] Among them, dilation is a morphological operation that dilates the edges of an object; erosion is a morphological operation that erodes the edges of an object. The dilation and erosion operations are as Figure 9a shown. In specific implementation, dilation or erosion needs to be performed according to the specific requirements of the HSV target feeder image. For example, some feeder line segments are dilated, and other feeder line segments are eroded. Figure 9b The figure shows the HSV target feeder image after dilation and erosion processing.

[0084] Step 308b: The image processing terminal compares the feature gray image with the standard feeder feature gray image in the preset feeder standard library, and determines whether the feeder of the target detection base station is installed correctly according to the comparison result.

[0085] Specifically, the comparison process of step 308 can refer to Figure 10a shown, and specifically includes:

[0086] Step 1002: Perform a coarse-grained comparison of the feature gray image and the standard feeder feature gray image through edge contour search and line segment counting; if the coarse-grained comparison result does not match, it is determined that the feeder is installed incorrectly; if the coarse-grained comparison result matches, step 1004 is further executed.

[0087] Through edge contour search, traverse the feature gray image obtained in the previous part, find and retain the connected line segments, count the number of line segments and record it as k, and compare it with the number of line segments k-std in the standard gray feeder image. If they are the same, continue with the next comparison. If they are different, it is determined that the feeder is installed incorrectly. Through contour search and counting comparison, the general comparison of the feeder image is completed in a coarse-grained manner as a prerequisite for the next finer-grained comparison.

[0088] Step 1004: Locate and register the feeder part in the feature grayscale image, and intercept to obtain the minimum target feeder image, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space;

[0089] Locate and register the feeder part in the feature grayscale image. Specifically, denote the pixel point number (i, j) in the feature grayscale image, and assume the image size is m, n. Assume four variables imin, imax, jmin, jmax, and their initial values are set to 0, 0, m - 1, n - 1 respectively. Starting from the point (0, 0), traverse all pixel points of the grayscale image. If the pixel point grayscale value is higher than the preset threshold ε, continue traversing; if lower, compare with the four variables. If the grayscale value is less than imin, assign the current i value to imin, and if less than jmin, assign the current j value to jmin; if the grayscale value is greater than imax, assign the current i value to imax, and if greater than jmax, assign the current j value to jmax. Image location is to prepare for the next step of target image interception, reduce unnecessary comparison areas, and improve the accuracy and performance of the algorithm.

[0090] Retain the rectangular area enclosed by imin, imax, jmin, and jmax obtained in the previous step, and transform the image size to the standard size m - std, n - std. Intercept the image according to the points located in the image to obtain the minimum target image; at the same time, perform scale change on the obtained minimum target image, so as to compare the two comparison images in the same dimension and increase the accuracy of comparison.

[0091] Step 1006: Compare the line segment length and line segment width of the line segment contained in the feeder in the minimum target feeder image with the line segment standard length and standard width of the line segment contained in the feeder in the standard feeder feature grayscale image in a fine - grained manner; if the fine - grained comparison result does not match, determine that the feeder is installed incorrectly; if the fine - grained comparison result matches, further execute Step 1008.

[0092] Specifically, when implementing, the line segments of the feeder can be numbered in sequence. Calculate the difference between the Y - coordinate value of the highest point and the Y - coordinate value of the lowest point of each line segment, that is, the line segment length L; calculate the difference between the X - coordinate value of the left - most point and the X - coordinate value of the right - most point of each line segment, that is, the line segment width W. Compare with the length L - std and width W - std of the line segment with the same number in the standard feature grayscale image. Preset a low threshold ε1 and a high threshold ε2. If L - std * ε1 < L < L - std * ε2 and W - std * ε1 < W < W - std * ε2, then pass the size feature comparison and continue to the next step; otherwise, determine that the feeder is installed incorrectly and notify the client. This step focuses on more detailed index conditions by performing line segment and fine - grained comparisons between the target image and the standard image.

[0093] Step 1008: Conduct an overall feature comparison using specific mean variance. Here, the specific mean variance is the variance determined by the mean of the gray values of all pixel points included in the line segments in the feeder of the minimum target feeder image corresponding to the line segments in the feeder of the standard feeder feature gray image. If the overall comparison result does not match, it is determined that the feeder is installed incorrectly. If the overall comparison result matches, it is determined that the feeder is installed correctly.

[0094] Specifically, the mean of the gray values of all pixel points included in the line segments of the feeder and the line segments in the standard gray map is denoted as E. The variances of the line segments of the feeder and the standard line segment from E are calculated respectively, denoted as D1 and D2. A preset variance threshold ε3 is set. If D1 + D2 < ε3, all comparisons are completed and it is determined that the feeder is installed correctly; otherwise, it is determined that the feeder is installed incorrectly. In this step, fine-grained indicators are associated and calculated through an algorithm, and each system granularity indicator is regarded as a system as a whole for judgment, thereby magnifying the differences between fine-grained indicators. The advantage of this is that magnifying the microscopic differences can more easily help us discover problems, thereby improving the overall quality of the algorithm and ensuring the accuracy of the results.

[0095] Next, through Figure 10b The entire detection step is made into a process. First, the image acquisition terminal takes feeder photos through manual shooting or drone shooting and uploads them to the corresponding storage of the cloud server. Then, the image processing terminal sends a detection request to the cloud server and obtains the corresponding feeder image. The image processing terminal performs operations such as preprocessing and image conversion on the feeder image. Then, the converted gray feature image is compared with the standard feeder image in the standard library, and it is determined whether the feeder is installed correctly according to the feature comparison result.

[0096] In the above technical solution of this specification, the feeder image of the base station to be detected collected by the image acquisition terminal is sent to the cloud server, and the cloud server stores the feeder image and the identifier of the corresponding base station to be detected correspondingly. In this way, the image processing terminal can search and obtain the target feeder image from the cloud server according to the detection request, and convert it into a feature gray image after performing image preprocessing operations on the target feeder image; finally, the obtained feature gray image is compared with the standard feeder feature gray image, and it is determined whether the feeder of the target detection base station is installed correctly according to the comparison result. Thus, through the above method, almost automated detection is achieved, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0097] Embodiment 2

[0098] Referring to Figure 11a As shown, a base station feeder detection system 1100 provided by an embodiment of the present invention includes: an image acquisition terminal 1102 and an image processing terminal 1106; where

[0099] The image acquisition terminal 1102 acquires the feeder image of the base station to be detected and stores the feeder image corresponding to the identifier of the current base station to be detected;

[0100] Based on the identifier of the target detection base station in the current detection request, the image processing terminal 1106 searches for and obtains the matching target feeder image;

[0101] The image processing terminal 1106 performs image preprocessing operations on the target feeder image and converts the target feeder image after the image preprocessing operations into a feature grayscale image;

[0102] The image processing terminal 1106 compares the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library and determines whether the feeder of the target detection base station is installed correctly according to the comparison result.

[0103] Refer to Figure 11b As shown, a base station feeder detection system 1100 provided by an embodiment of the present invention includes: an image acquisition terminal 1102, a cloud server 1104, and an image processing terminal 1106; wherein,

[0104] The image acquisition terminal 1102 acquires the feeder image of the base station to be detected and sends the feeder image and the identifier of the current base station to be detected to the cloud server 1104, so that the cloud server 1104 stores the feeder image corresponding to the identifier of the current base station to be detected;

[0105] Based on the identifier of the target detection base station in the current detection request, the image processing terminal 1106 searches for and obtains the matching target feeder image from the cloud server 1104;

[0106] The image processing terminal 1106 performs image preprocessing operations on the target feeder image and converts the target feeder image after the image preprocessing operations into a feature grayscale image;

[0107] The image processing terminal 1106 compares the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library and determines whether the feeder of the target detection base station is installed correctly according to the comparison result.

[0108] Optionally, if the image acquisition terminal 1102 is a contact handheld image capturing device, when the image acquisition terminal 1102 captures the feeder image of the base station to be detected, it is specifically configured to capture the feeder image from the base station to be detected in response to a user's touch operation; if the image acquisition terminal 1102 is a non-contact image capturing device, when the image acquisition terminal 1102 captures the feeder image of the base station to be detected, it is specifically configured to capture the feeder image from the base station to be detected in response to a user's remote control.

[0109] In an implementable solution, when the image processing terminal 1106 performs image preprocessing operations on the target feeder image, it is specifically configured to perform scaling processing on the target feeder image according to the detection pixel requirements; select feeder marker points from the scaled target feeder image, and extract the HSV values of each feeder marker point; based on the extracted HSV values of each feeder marker point, convert the scaled target feeder image into the HSV image format.

[0110] In another implementable solution, when the image processing terminal 1106 converts the target feeder image after the image preprocessing operation into a feature grayscale image, it is specifically configured to use edge detection technology to process the converted HSV target feeder image to obtain the edge contour of the feeder marker points; perform median filtering on the HSV target feeder image after edge detection processing; perform morphological operations of dilation or erosion on the edges of the feeder marker points in the HSV target feeder image after median filtering.

[0111] In another implementable solution, the edge detection technology at least includes: Sobel edge detection operator, Laplacian edge detection operator, Canny edge detection operator.

[0112] In another implementable solution, when the image processing terminal 1106 uses the Canny edge detection operator to process the converted HSV target feeder image, it is specifically configured to use a Gaussian filter to smooth the converted HSV target feeder image and filter out noise; calculate the gradient value and direction of each pixel point in the processed HSV target feeder image; perform non-maximum suppression on each pixel point based on the calculated gradient intensity and direction; use double-threshold edge detection to further optimize the edges: if the gradient value of an edge pixel is greater than the high threshold, it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, it is suppressed; determine the edge contour of the final feeder marker points by suppressing isolated weak edge pixels.

[0113] Another feasible solution is that when the image processing terminal 1106 compares the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library and determines whether the feeder of the target detection base station is correctly installed according to the comparison result, it is specifically used to perform a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; if the coarse-grained comparison result does not match, it is determined that the feeder is incorrectly installed; if the coarse-grained comparison result matches, the feeder part in the feature grayscale image is further positioned and registered, and the minimum target feeder image is intercepted, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image are respectively compared with the standard line segment length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image for fine-grained comparison; if the fine-grained comparison result does not match, it is determined that the feeder is incorrectly installed; if the fine-grained comparison result matches, a feature overall comparison is further performed using a specific mean variance, where the specific mean variance is the variance determined by the mean of the gray values of all the pixel points included in the corresponding line segments of the feeder in the minimum target feeder image and the feeder in the standard feeder feature grayscale image; if the overall comparison result does not match, it is determined that the feeder is incorrectly installed; if the overall comparison result matches, it is determined that the feeder is correctly installed.

[0114] In the above technical solution of this specification, the image acquisition terminal sends the acquired feeder image of the base station to be detected to the cloud server, and the cloud server stores the feeder image and the identifier of the corresponding base station to be detected. In this way, the image processing terminal can search for and obtain the target feeder image from the cloud server according to the detection request, and after performing image preprocessing operations on the target feeder image, convert it into a feature grayscale image; finally, compare the obtained feature grayscale image with the standard feeder feature grayscale image, and determine whether the feeder of the target detection base station is correctly installed according to the comparison result. Thus, through the above method, almost automatic detection is achieved, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0115] Embodiment III

[0116] The embodiment of this specification also provides an image processing method, specifically referring to Figure 12 as shown, this image processing method may include the following steps:

[0117] Step 1202: Based on the identifier of the target detection base station in the current detection request, search for and obtain a matching target feeder image, where the target feeder image is the feeder image of the base station to be detected acquired by the image acquisition terminal and the feeder image is stored corresponding to the identifier of the current base station to be detected.

[0118] Among them, the image acquisition terminal will send the acquired feeder image together with the identifier of the current base station to be detected to the image processing terminal or the cloud server for corresponding storage. Accordingly, the matching target feeder image can be found and obtained from the local image processing terminal or the cloud server.

[0119] Step 1204: Perform image preprocessing operations on the target feeder image, and convert the target feeder image after the image preprocessing operations into a feature grayscale image;

[0120] Step 1206: Compare the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library, and determine whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0121] Optionally, the performing image preprocessing operations on the target feeder image specifically includes: performing scaling processing on the target feeder image according to the detection pixel requirements; selecting feeder marker points from the scaled target feeder image, and extracting the HSV values of each feeder marker point; based on the extracted HSV values of each feeder marker point, converting the scaled target feeder image into the HSV image format.

[0122] A feasible solution, the converting the target feeder image after the image preprocessing operations into a feature grayscale image specifically includes: using edge detection technology to process the converted HSV target feeder image to obtain the edge contour of the feeder marker points; performing median filtering processing on the HSV target feeder image after edge detection processing; performing morphological operations of dilation or erosion on the edges of the feeder marker points in the HSV target feeder image after median filtering processing.

[0123] A feasible solution, the edge detection technology at least includes: Sobel edge detection operator, Laplacian edge detection operator, Canny edge detection operator.

[0124] Another feasible solution is that when using the Canny edge detection operator to process the converted HSV target feeder image, it specifically includes: using a Gaussian filter to smooth the converted HSV target feeder image and filter out noise; calculating the gradient value and direction of each pixel point in the processed HSV target feeder image; performing non-maximum suppression on each pixel point based on the calculated gradient intensity and direction; using double-threshold edge detection to further optimize the edges: if the gradient value of an edge pixel is greater than the high threshold, it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, it is suppressed; determining the edge contour of the final feeder marking point by suppressing isolated weak edge pixels.

[0125] Another feasible solution is that when comparing the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library and determining whether the feeder of the target detection base station is installed correctly according to the comparison result, it specifically includes: performing a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; if the coarse-grained comparison result does not match, it is determined that the feeder is installed incorrectly; if the coarse-grained comparison result matches, further perform positioning and registration on the feeder part in the feature grayscale image and intercept to obtain the minimum target feeder image, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; performing a fine-grained comparison on the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image with the standard line segment length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image respectively; if the fine-grained comparison result does not match, it is determined that the feeder is installed incorrectly; if the fine-grained comparison result matches, further perform an overall feature comparison using a specific mean variance, where the specific mean variance is the variance determined by the mean of the gray values of all pixel points included in the corresponding line segments of the line segments in the feeder of the minimum target feeder image and the feeder of the standard feeder feature grayscale image; if the overall comparison result does not match, it is determined that the feeder is installed incorrectly; if the overall comparison result matches, it is determined that the feeder is installed correctly.

[0126] According to the above technical solutions of this specification, the image processing terminal can search for and obtain the target feeder image from the cloud server according to the detection request, and convert it into a feature grayscale image after performing image preprocessing operations on the target feeder image; compare the obtained feature grayscale image with the standard feeder feature grayscale image, and determine whether the feeder of the target detection base station is installed correctly according to the comparison result. Thus, almost automated detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0127] Embodiment 4

[0128] The embodiment of this specification also provides an image processing device, specifically referring to Figure 13 As shown, the image processing device 1300 may include:

[0129] An acquisition module 1302, configured to search for and acquire a matching target feeder image based on the identifier of the target detection base station in the current detection request, where the target feeder image is an image of the feeder of the base station to be detected collected by an image acquisition terminal, and the feeder image is stored corresponding to the identifier of the current base station to be detected;

[0130] A conversion module 1304, configured to perform image preprocessing operations on the target feeder image, and convert the target feeder image after the image preprocessing operations into a feature grayscale image;

[0131] A comparison module 1306, configured to compare the feature grayscale image with the standard feeder feature grayscale image in a preset feeder standard library, and determine whether the feeder of the target detection base station is correctly installed according to the comparison result.

[0132] Optionally, when the conversion module 1304 performs image preprocessing operations on the target feeder image, it is specifically configured to perform scaling processing on the target feeder image according to the detection pixel requirements; select feeder marking points from the scaled target feeder image, and extract the HSV values of each feeder marking point; based on the extracted HSV values of each feeder marking point, perform HSV image format conversion on the scaled target feeder image.

[0133] In an implementable solution, when the conversion module 1304 converts the target feeder image after the image preprocessing operations into a feature grayscale image, it is specifically configured to use edge detection technology to process the converted HSV target feeder image to obtain the edge contour of the feeder marking points; perform median filtering processing on the HSV target feeder image after edge detection processing; perform morphological operations of dilation or erosion on the edges of the feeder marking points in the HSV target feeder image after median filtering processing.

[0134] In an implementable solution, the edge detection technology at least includes: Sobel edge detection operator, Laplacian edge detection operator, Canny edge detection operator.

[0135] Another feasible solution is that when the conversion module 1304 processes the converted HSV target feeder image using the Canny edge detection operator, it specifically uses a Gaussian filter to smooth the converted HSV target feeder image and filter out noise; calculates the gradient value and direction of each pixel point in the processed HSV target feeder image; performs non-maximum suppression on each pixel point based on the calculated gradient intensity and direction; uses double-threshold edge detection to further optimize the edges: if the gradient value of an edge pixel is greater than the high threshold, it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, it is suppressed; determines the edge contour of the final feeder marking point by suppressing isolated weak edge pixels.

[0136] Another feasible solution is that when the comparison module 1306 compares the feature grayscale image with the standard feeder feature grayscale image in the preset feeder standard library and determines whether the feeder of the target detection base station is correctly installed according to the comparison result, it specifically performs a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; if the coarse-grained comparison result does not match, it determines that the feeder is incorrectly installed; if the coarse-grained comparison result matches, it further locates and registers the feeder part in the feature grayscale image and intercepts to obtain the minimum target feeder image, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; performs a fine-grained comparison on the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image with the standard line segment length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image respectively; if the fine-grained comparison result does not match, it determines that the feeder is incorrectly installed; if the fine-grained comparison result matches, it further performs an overall feature comparison using a specific mean variance, where the specific mean variance is the variance determined by the mean of the gray values of all pixel points included in the corresponding line segments of the line segments in the feeder of the minimum target feeder image and the feeder of the standard feeder feature grayscale image; if the overall comparison result does not match, it determines that the feeder is incorrectly installed; if the overall comparison result matches, it determines that the feeder is correctly installed.

[0137] According to the above technical solutions of this specification, the image processing terminal can search for and obtain the target feeder image from the cloud server according to the detection request, and convert it into a feature grayscale image after performing image preprocessing operations on the target feeder image; compare the obtained feature grayscale image with the standard feeder feature grayscale image, and determine whether the feeder of the target detection base station is correctly installed according to the comparison result. Thus, almost automated detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0138] It should be understood that the solution details and corresponding effects involved in the above-mentioned Second Embodiment - Fourth Embodiment are all described in detail in the First Embodiment. For specific details, reference may be made to the content of the First Embodiment.

[0139] Embodiment Five

[0140] This embodiment of the present specification also provides an electronic device, including: a processor; and a memory arranged to store computer-executable instructions, and when the executable instructions are executed, the processor is caused to execute as Figure 12 the image processing method described above.

[0141] Meanwhile, a storage medium provided by an embodiment of the present invention is used for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement steps of Figure 12 the image processing method as shown.

[0142] According to the above technical solution of this specification, the image processing terminal can search for and obtain the target feeder image from the cloud server according to the detection request, and after performing image preprocessing operations on the target feeder image, convert it into a feature grayscale image; compare the obtained feature grayscale image with the standard feeder feature grayscale image, and determine whether the feeder of the target detection base station is correctly installed according to the comparison result. Thus, almost automated detection is achieved through the above method, reducing the detection cost and potential safety hazards, and improving the detection accuracy and efficiency.

[0143] In summary, the above are only the preferred embodiments of this specification and are not used to limit the protection scope of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of this specification shall be included within the protection scope of this specification.

[0144] The systems, devices, modules, or units illustrated in the above one or more embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0145] A computer-readable storage medium includes both permanent and non-permanent, removable and non-removable media and can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of the computer's storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, tape magnetic disks storage, or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0146] It should also be noted that the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity, or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity, or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity, or device comprising the element.

[0147] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the related content.

[0148] The specific embodiments of this specification have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An image processing method, comprising: acquiring a feeder image of a base station to be detected, and storing the acquired feeder image of the base station to be detected corresponding to the identifier of the current base station to be detected; based on the identifier of the target detection base station in the current detection request, searching for and obtaining a matching target feeder image; performing an image preprocessing operation on the target feeder image, and converting the target feeder image after the image preprocessing operation into a feature grayscale image; performing a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; if the result of the coarse-grained comparison does not match, determining that the feeder is incorrectly installed; if the result of the coarse-grained comparison matches, further performing positioning and registration on the feeder part in the feature grayscale image, and intercepting to obtain a minimum target feeder image, wherein the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; performing a fine-grained comparison on the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image respectively with the line segment standard length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image; if the result of the fine-grained comparison does not match, determining that the feeder is incorrectly installed; if the result of the fine-grained comparison matches, determining the mean value of the gray values of all the pixel points included in the line segment in the feeder of the minimum target feeder image and the corresponding line segment in the feeder of the standard feeder feature grayscale image, calculating the first variance of the gray values of the pixel points included in the line segment in the feeder of the minimum target feeder image and the mean value, and calculating the second variance of the gray values of the pixel points included in the corresponding line segment in the feeder of the standard feeder feature grayscale image and the mean value; if the sum value of the first variance and the second variance is greater than or equal to a preset threshold, determining that the feeder is incorrectly installed; if the sum value is less than the preset threshold, determining that the feeder is correctly installed.

2. The image processing method according to claim 1, wherein the image preprocessing operation on the target feeder image specifically comprises: performing a scaling process on the target feeder image according to the detection pixel requirement; selecting feeder marking points from the scaled target feeder image, and extracting the HSV values of each feeder marking point; based on the HSV values of the extracted feeder marking points, converting the scaled target feeder image into an HSV image format.

3. The image processing method according to claim 2, wherein converting the target feeder image after the image preprocessing operation into a feature grayscale image specifically comprises: using an edge detection technique to process the converted HSV target feeder image to obtain the edge contour of the feeder marking points; performing a median filtering process on the HSV target feeder image after the edge detection process; performing a morphological operation of dilation or erosion on the edge of the feeder marking points in the HSV target feeder image after the median filtering process.

4. The image processing method according to claim 3, wherein the edge detection technique at least comprises: Sobel edge detection operator, Laplacian edge detection operator, Canny edge detection operator; When using the Canny edge detection operator to process the converted HSV target feeder image, it specifically includes: Using a Gaussian filter to smooth the converted HSV target feeder image and filter out noise; Calculating the gradient value and direction of each pixel point in the processed HSV target feeder image; Performing non-maximum suppression on each pixel point based on the calculated gradient intensity and direction; Using double-threshold edge detection to further optimize the edge: if the gradient value of an edge pixel is greater than the high threshold, it is marked as a strong edge pixel; if the gradient value of an edge pixel is less than the high threshold and greater than the low threshold, it is marked as a weak edge pixel; if the gradient value of an edge pixel is less than the low threshold, it is suppressed; Determining the edge contour of the final feeder marking point by suppressing isolated weak edge pixels.

5. An image processing device, including: An acquisition module for collecting the feeder image of the base station to be detected and storing the collected feeder image of the base station to be detected corresponding to the identifier of the current base station to be detected; Based on the identifier of the target detection base station in the current detection request, searching for and obtaining a matching target feeder image; A conversion module for performing image preprocessing operations on the target feeder image and converting the target feeder image after the image preprocessing operations into a feature grayscale image; A comparison module for performing a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image by edge contour search and line segment counting; If the result of the coarse-grained comparison does not match, it is determined that the feeder is installed incorrectly; If the result of the coarse-grained comparison matches, further perform positioning and registration on the feeder part in the feature grayscale image and intercept to obtain the minimum target feeder image, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; Performing a fine-grained comparison on the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image with the standard line segment length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image respectively; If the result of the fine-grained comparison does not match, it is determined that the feeder is installed incorrectly; If the result of the fine-grained comparison matches, determine the mean value of the gray values of all pixel points included in the line segment in the feeder of the minimum target feeder image and the corresponding line segment in the feeder of the standard feeder feature grayscale image, calculate the first variance of the gray values of the pixel points included in the line segment in the feeder of the minimum target feeder image and the mean value, and calculate the second variance of the gray values of the pixel points included in the corresponding line segment in the feeder of the standard feeder feature grayscale image and the mean value; If the sum value of the first variance and the second variance is greater than or equal to a preset threshold, it is determined that the feeder is installed incorrectly; If the sum value is less than the preset threshold, it is determined that the feeder is installed correctly.

6. A base station feeder detection method, including: An image acquisition terminal collects the feeder image of the base station to be detected and sends the collected feeder image of the base station to be detected and the identifier of the current base station to be detected to an image processing terminal or a cloud server for corresponding storage; The image processing terminal looks up and obtains a matching target feeder image based on the identifier of the target detection base station in the current detection request; The image processing terminal performs image preprocessing operations on the target feeder image and converts the target feeder image after the image preprocessing operations into a feature grayscale image; The image processing terminal performs a coarse-grained comparison between the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; If the result of the coarse-grained comparison does not match, it is determined that the feeder is incorrectly installed; If the result of the coarse-grained comparison matches, the feeder part in the feature grayscale image is further positioned and registered, and the minimum target feeder image is intercepted, where the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; The line segment lengths and line segment widths of the line segments included in the feeder in the minimum target feeder image are respectively compared with the line segment standard lengths and standard widths of the line segments included in the feeder in the standard feeder feature grayscale image for fine-grained comparison; If the result of the fine-grained comparison does not match, it is determined that the feeder is incorrectly installed; If the result of the fine-grained comparison matches, determine the mean value of the grayscale values of all the pixel points included in the line segment in the feeder of the minimum target feeder image and the corresponding line segment in the feeder of the standard feeder feature grayscale image, calculate the first variance of the grayscale values of the pixel points included in the line segment in the feeder of the minimum target feeder image and the mean value, and calculate the second variance of the grayscale values of the pixel points included in the corresponding line segment in the feeder of the standard feeder feature grayscale image and the mean value; If the sum value of the first variance and the second variance is greater than or equal to a preset threshold, it is determined that the feeder is incorrectly installed; If the sum value is less than the preset threshold, it is determined that the feeder is correctly installed.

7. The base station feeder detection method according to claim 6, wherein if the image acquisition terminal is a contact-type handheld image capturing device, the image acquisition terminal captures the feeder image of the base station to be detected, specifically including: The contact-type handheld image capturing device captures a feeder image from the base station to be detected in response to a user's touch operation; If the image acquisition terminal is a non-contact image capturing device, the image acquisition terminal captures the feeder image of the base station to be detected, specifically including: The non-contact image capturing device captures a feeder image from the base station to be detected in response to a user's remote control.

8. The base station feeder detection method according to claim 6 or 7, wherein the image acquisition terminal stores the captured feeder image of the base station to be detected corresponding to the identifier of the current base station to be detected, specifically including: Sending the captured feeder image of the base station to be detected and the identifier of the current base station to be detected to the cloud server, so that the cloud server stores the captured feeder image of the base station to be detected corresponding to the identifier of the current base station to be detected; The image processing terminal looks up and obtains a matching target feeder image based on the identifier of the target detection base station in the current detection request, specifically including: Looking up and obtaining a matching target feeder image from the cloud server based on the identifier of the target detection base station in the current detection request.

9. A base station feeder detection system, including: An image acquisition terminal and an image processing terminal; wherein, the image acquisition terminal acquires the feeder image of the base station to be detected, and stores the acquired feeder image of the base station to be detected corresponding to the identifier of the current base station to be detected; the image processing terminal searches for and obtains the matching target feeder image based on the identifier of the target detection base station in the current detection request; the image processing terminal performs image preprocessing operations on the target feeder image, and converts the target feeder image after the image preprocessing operations into a feature grayscale image; the image processing terminal performs a coarse-grained comparison on the feature grayscale image and the standard feeder feature grayscale image through edge contour search and line segment counting; if the result of the coarse-grained comparison does not match, it is determined that the feeder is installed incorrectly; if the result of the coarse-grained comparison matches, the feeder part in the feature grayscale image is further positioned and registered, and the minimum target feeder image is intercepted, wherein the minimum target feeder image and the standard feeder feature grayscale image are in the same dimensional space; the line segment length and line segment width of the line segments included in the feeder in the minimum target feeder image are respectively compared with the line segment standard length and standard width of the line segments included in the feeder in the standard feeder feature grayscale image for fine-grained comparison; if the result of the fine-grained comparison does not match, it is determined that the feeder is installed incorrectly; if the result of the fine-grained comparison matches, the mean value of the gray values of all the pixel points included in the line segments in the feeder of the minimum target feeder image and the corresponding line segments in the feeder of the standard feeder feature grayscale image is determined, the first variance of the gray values of the pixel points included in the line segments in the feeder of the minimum target feeder image and the mean value is calculated, and the second variance of the gray values of the pixel points included in the corresponding line segments in the feeder of the standard feeder feature grayscale image and the mean value is calculated; if the sum value of the first variance and the second variance is greater than or equal to a preset threshold, it is determined that the feeder is installed incorrectly; if the sum value is less than the preset threshold, it is determined that the feeder is installed correctly.

10. An electronic device, comprising: a processor; and a memory storing computer-executable instructions, the executable instructions, when executed, cause the processor to execute the image processing method according to any one of claims 1-4, or the executable instructions, when executed, cause the processor to execute the base station feeder detection method according to any one of claims 6-8.

11. A computer-readable storage medium, the storage medium stores one or more programs, the one or more programs, when executed by one or more processors, implement the image processing method according to any one of claims 1-4, or the one or more programs, when executed by one or more processors, implement the base station feeder detection method according to any one of claims 6-8.

Images