CN114463395A - A kind of monitoring equipment offset detection method, equipment and medium - Google Patents

Abstract

The application discloses a method, a device and a medium for detecting offset of monitoring equipment, wherein the method comprises the following steps: acquiring a monitoring image of equipment to be inspected, which is shot by monitoring equipment, within a preset time interval; searching a template image with the highest similarity to the monitored image in a pre-constructed template set; the template set comprises a plurality of template images corresponding to the equipment to be inspected under different illumination conditions; respectively extracting the feature points of the monitoring image and the template image; performing feature matching on the feature points of the monitored image and the feature points of the template image, and determining the offset distance of the feature points of the monitored image relative to the feature points of the template image; and analyzing the offset distance according to a preset rule so as to perform offset detection on the monitoring equipment. The deviation of the monitoring equipment is qualitatively and quantitatively analyzed, so that the deviation detection of the monitoring equipment can be more efficiently carried out.

Description

A kind of monitoring equipment offset detection method, equipment and medium

technical field

The present application relates to the field of electric power technology, and in particular, to a method, device and medium for detecting the offset of a monitoring device.

Background technique

As the scale of the grid gradually expands, so does the deployment of surveillance equipment. If the installation position of the monitoring equipment is loose, or the engineers overhaul, etc., the position of the monitoring equipment will be shifted. Based on this, the change of the camera angle of the monitoring equipment, on the one hand, will make the installation of the monitoring equipment out of specification, affecting the user experience of intelligent inspection, on the other hand, the change of the camera angle will lead to the skewed imaging angle. Even if the target area cannot be photographed, the recognition accuracy of the algorithm will be affected, and events that rely on human-marked features will produce errors, resulting in the failure of intelligent inspection.

At present, a solution based on a video stream is usually adopted, and the offset of the monitoring device is judged by means of feature matching or optical flow between frames. However, under the premise of not affecting the insulation and safety of power equipment at all, because the power industry adopts active wireless or passive wireless methods, there are extremely high requirements for power consumption, and the method based on video streaming is difficult to meet the power consumption requirements. , the monitoring equipment has been in a dormant state for a long time, which makes it difficult to use a long-term video stream to monitor whether the monitoring equipment is offset in real time, resulting in low monitoring equipment offset detection efficiency.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a method, device, and medium for detecting an offset of a monitoring device, which are used to solve the problem of low efficiency in detecting the offset of a monitoring device.

The embodiment of the present application adopts the following technical solutions:

On the one hand, an embodiment of the present application provides a method for detecting the offset of a monitoring device, the method includes: within a preset time interval, acquiring a monitoring image related to the device to be inspected shot by the monitoring device; In the template set, retrieve the template image with the highest similarity with the surveillance image; wherein, the template set includes a plurality of template images corresponding to the equipment to be inspected under different lighting conditions; extract the surveillance image and the surveillance image respectively. feature points of the template image; feature matching of the feature points of the surveillance image with the feature points of the template image to determine the offset of the feature points of the surveillance image relative to the feature points of the template image distance; analyze the offset distance according to a preset rule, so as to perform offset detection on the monitoring device.

In an example, retrieving the template image with the highest similarity to the surveillance image from the pre-built template set specifically includes: converting the surveillance image and the multiple template images from RGB space to HSV respectively. space; according to the HSV space, extract the first V component of the surveillance image, and extract the second V component corresponding to the plurality of template images respectively; extract the first histogram feature of the first V component , and the second histogram feature of the second V component; the correlation analysis is performed on the first histogram feature and the second histogram feature to determine the surveillance image and the multiple template images. The similarity degrees corresponding to each other respectively; according to the similarity degrees, determine the template image with the highest similarity with the surveillance image.

In an example, the extracting the feature points of the surveillance image and the template image respectively includes: acquiring an image watermark of the surveillance image; wherein the template image does not include the image watermark; the The image watermark is the identification information of the manufacturer to which the surveillance equipment belongs; according to the resolution of the surveillance image, the pixel area occupied by the image watermark in the surveillance image is determined; A designated area that does not contain an image watermark is extracted from the surveillance image; the designated area of the surveillance image and the feature points of the template image are extracted respectively through a fast feature point extraction model ORB.

In an example, extracting the feature points of the designated area of the surveillance image and the template image respectively by using the fast feature point extraction model ORB, specifically includes: constructing scale images for the designated area and the template image respectively. Pyramid; perform grid processing on the scale image pyramid, and perform feature point detection on each small grid in each layer of pyramid images through the FAST feature point detection algorithm, and extract the feature points of the designated area and the template image respectively. .

In one example, the feature matching of the feature points of the surveillance image and the feature points of the template image is performed, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined, Specifically, it includes: performing feature matching on the feature points of the surveillance image and the feature points of the template image through a fast nearest search and matching algorithm to obtain multiple sets of feature points with matching relationships; filtering the multiple sets of feature points to determine multiple sets of matching feature points with correct matching relationships; among the multiple sets of matching feature points with correct matching relationships, calculate each set of matching feature points according to the coordinate parameters of each matching feature point offset distance between the surveillance image and the template image, respectively.

In an example, the analyzing the offset distance according to a preset rule to perform offset detection on the monitoring device specifically includes: calculating an average distance corresponding to a plurality of the offset distances; if the If the average distance is less than the preset offset distance threshold, then judge whether the similarity between the surveillance image and the template image is less than the preset similarity threshold; if so, send a detection notification to the user; obtain feedback from the user information, and whether the monitoring device is offset is determined through the feedback information.

In one example, the method further includes: if the average distance is greater than a preset offset distance threshold, determining that the monitoring device is offset; in the multiple sets of matching feature points with correct matching relationships, according to The coordinate parameters of each matching feature point are used to generate a coordinate transformation matrix; through the coordinate transformation matrix, the surveillance image is corrected; and through the coordinate transformation matrix, the horizontal offset of the surveillance device, The vertical direction offset component and the offset rotation angle enable the user to correct the monitoring device.

In one example, the surveillance image includes several devices to be inspected, and the method further includes: determining, according to coordinate information of the several devices to be inspected in the surveillance image, whether the several devices to be inspected are in the surveillance image. The corresponding areas in the monitoring images; the equipment to be inspected with the smallest area is offset multiple times, so that the user can judge the distance that the equipment to be inspected with the smallest area can tolerate the offset, and The offset tolerance distance is used as the preset offset distance.

On the other hand, an embodiment of the present application provides a monitoring device offset detection device, the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores There are instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can: within a preset time interval, acquire the monitoring images of the device to be inspected image; from a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images of the equipment to be inspected under different lighting conditions; feature points of the surveillance image and the template image; feature matching is performed between the feature points of the surveillance image and the template image, and the feature points of the surveillance image are determined relative to the template image. The offset distance of the feature point; the offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device.

On the other hand, an embodiment of the present application provides a non-volatile computer storage medium for monitoring device offset detection, which stores computer-executable instructions, where the computer-executable instructions are set to: within a preset time interval, obtain Surveillance images of the equipment to be inspected; from a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes multiple images of the equipment to be inspected under different lighting conditions. a template image; extract the feature points of the surveillance image and the template image respectively; perform feature matching between the feature points of the surveillance image and the template image to determine the feature points of the surveillance image The offset distance relative to the feature points of the template image; the offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device.

The above-mentioned at least one technical solution adopted in the embodiments of the present application can achieve the following beneficial effects:

In the embodiment of the present application, the surveillance images of the equipment to be inspected are acquired at preset time intervals, and there is no need to acquire the video stream in real time, which reduces the power consumption of the surveillance equipment. The template set corresponding to the template image under lighting conditions, retrieve the template image with the highest similarity with the monitoring image, which can improve the detection accuracy. Finally, by analyzing the offset distance of the monitoring image relative to the template image, the monitoring equipment Offset detection is performed, thereby completing the qualitative and quantitative analysis of the offset of the monitoring equipment, and realizing the more efficient offset detection of the monitoring equipment.

Description of drawings

In order to illustrate the technical solutions of the present application more clearly, some embodiments of the present application will be described in detail below with reference to the accompanying drawings. In the accompanying drawings:

FIG. 1 is a schematic flowchart of a method for detecting an offset of a monitoring device provided by an embodiment of the present application;

FIG. 2 is a schematic structural diagram of a monitoring device offset detection device according to an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be described clearly and completely below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

Some embodiments of the present application will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic flowchart of a method for detecting an offset of a monitoring device provided by an embodiment of the present application. Certain input parameters or intermediate results in the process allow for manual intervention adjustments to help improve accuracy.

The implementation of the analysis method involved in the embodiments of the present application may be a terminal device or a server, which is not particularly limited in the present application. For the convenience of understanding and description, the following embodiments take a server as an example for detailed description.

The flow in Figure 1 can include the following steps:

S101: Acquire, within a preset time interval, a surveillance image related to the device to be inspected shot by the surveillance device.

Wherein, the device to be inspected may be an electric device. For example, at certain time intervals, such as one hour, one day, and three days, surveillance images of the electronic device are collected.

S102: From a pre-built template set, retrieve a template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images corresponding to the device to be inspected under different lighting conditions.

It should be noted that the template image is used as the reference image of the surveillance image, that is, when the surveillance device does not shift, the device to be inspected is photographed to obtain the template image.

In addition, when the monitoring equipment shoots the equipment to be inspected under different lighting conditions, the obtained monitoring images are different. However, the lighting conditions of the monitoring images will affect the accuracy of feature point extraction. Therefore, when the monitoring equipment does not offset, it shoots two monitoring images with different lighting conditions. But in fact, the feature points of the two surveillance images are matched.

Based on this, in order to make the monitoring image match the template image with the highest similarity as much as possible when detecting the offset of the monitoring equipment, when collecting the template image, the corresponding multiple images of the equipment to be inspected under different lighting conditions are collected. A template image can be used to reduce the influence of lighting conditions on feature point extraction, thereby improving the accuracy of monitoring equipment offset detection.

Further, the template set may include template images captured by multiple surveillance devices, that is, each surveillance device shoots its corresponding device to be inspected under different lighting conditions, so as to obtain the corresponding images of the device to be inspected. Multiple template images.

For example, multiple surveillance devices in different power scenarios capture surveillance images of the respective corresponding power devices under different lighting conditions, thereby obtaining multiple template images corresponding to the respective corresponding power devices.

S103: Extract the feature points of the surveillance image and the template image respectively.

Based on this, the feature points of surveillance images are automatically extracted and the configuration of artificial feature points is reduced.

S104: Perform feature matching between the feature points of the surveillance image and the feature points of the template image, and determine the offset distance of the feature points of the surveillance image relative to the feature points of the template image.

The matching relationship of the feature points in the two images can be determined through feature matching, and the offset distance relative to the feature points of the template image can be determined through the matching relationship.

It should be noted that since the number of feature points is not only one, the number of offset distances may be multiple.

S105: Analyze the offset distance according to a preset rule, so as to perform offset detection on the monitoring device.

Wherein, if the number of offset distances is multiple, the offset distances corresponding to the multiple feature points can be combined and analyzed to determine whether the monitoring device is offset.

Through the method shown in Figure 1, the monitoring images of the equipment to be inspected are obtained at preset time intervals, and there is no need to obtain the video stream in real time, which reduces the power consumption of the monitoring equipment. In the template set corresponding to the template image under different lighting conditions, the template image with the highest similarity to the surveillance image can be retrieved, which can improve the detection accuracy. Finally, by analyzing the offset distance between the surveillance image and the template image, the monitoring The offset detection of the camera equipment is completed, thus the qualitative and quantitative analysis of the offset of the monitoring equipment is completed, and the offset detection of the monitoring equipment can be realized more efficiently, which is especially suitable for the monitoring equipment of the non-video stream in the power scene. Offset detection method.

It should be noted that, although steps S101 to S105 are sequentially introduced and described in this embodiment of the present application with reference to FIG. 1 , this does not mean that steps S101 to S105 must be performed in strict order. The reason why the embodiments of the present application describe steps S101 to S105 in sequence according to the sequence shown in FIG. 1 is to facilitate those skilled in the art to understand the technical solutions of the embodiments of the present application. In other words, in this embodiment of the present application, the sequence between steps S101 to S105 may be appropriately adjusted according to actual needs.

Based on the method in FIG. 1 , some specific implementations and expansion schemes of the method are also provided in the embodiments of the present application, and the description will be continued below.

In some embodiments of the present application, in order to reduce misjudgments in the case of illumination changes and noise, and improve the adaptability and robustness of the algorithm, the template image with the highest similarity to the current surveillance image is selected from the template set, then, the server It is necessary to confirm the similarity between the surveillance image and each template image in the template set, and then select the template image with the highest similarity with the current surveillance image.

Specifically, the server converts the surveillance image and the multiple template images from the RGB space to the HSV space respectively. Then, according to the HSV space, extract the first V component of the surveillance image, extract the second V component corresponding to the multiple template images respectively, extract the first histogram feature of the first V component, and extract the first V component of the second V component Two histogram features. Then, a correlation analysis is performed on the first histogram feature and the second histogram feature to determine the corresponding degrees of similarity between the surveillance image and the plurality of template images respectively. Finally, according to the similarity, the template image with the highest similarity with the surveillance image is determined.

It should be noted that the corresponding similarity between the surveillance image and the multiple template images is that the correlation analysis result may be a negative number. Therefore, the absolute value of the correlation analysis result is used as the similarity of the two images.

In some implementations of the present application, since the surveillance images usually carry the identification information of the manufacturer to which the surveillance equipment belongs, that is, the image watermark, in order to filter the interference of the image watermark, the server removes the image watermark area, and only performs the image watermark in the image watermark area. Extract feature points.

It should be noted that the image watermark area in the template image has been removed. Of course, the template image may also include the image watermark area. When removing the image watermark area from the surveillance image, the image watermark area of the template image is removed at the same time.

Specifically, the server obtains the image watermark of the surveillance image, determines the pixel area occupied by the image watermark in the surveillance image according to the resolution of the surveillance image, and removes the pixel area to extract the image watermark that does not contain the image watermark in the surveillance image. designated area.

After the specified area is extracted, the fast feature point extraction model ORB is used to extract the specified area of the surveillance image and the feature points of the template image respectively.

Among them, ORB feature extraction is a fast feature point extraction and description algorithm. It is divided into two parts, namely feature point extraction and feature point description. Feature extraction is developed by FAST algorithm, and feature point description is improved according to Brief feature description algorithm. The ORB feature combines the detection method of FAST feature points with the Brief feature descriptor, and improves and optimizes them on the basis of their original ones. Compared with SIFT and SURF, it has a faster running speed.

Further, when extracting the feature points of the designated area of the surveillance image and the template image respectively, the server builds a scale image pyramid for the designated area and the template image respectively, performs grid processing on the scale image pyramid, and uses the FAST feature point detection algorithm to detect each image. Feature point detection is performed on each small grid in the layer pyramid image, and the feature points of the specified area and the template image are extracted respectively.

In some implementations of the present application, when the feature points of the surveillance image are matched with the feature points of the template image, at present, the brute force search matching method BF is usually used to match the feature points, but the efficiency of this method is very low. When the data dimension is high, the efficiency of matching drops sharply. Since ORB features are binary features, the fast nearest neighbor search and matching algorithm FLANN can be used for feature matching.

Based on this, the server performs feature matching between the feature points of the surveillance image and the feature points of the template image through the fast nearest search matching algorithm, and obtains multiple sets of feature points with matching relationships.

Due to the influence of various factors, the matching results may include incorrect matching relationships. Therefore, in order to effectively ensure the accuracy of feature point matching, it is necessary to filter the incorrect matching relationships to realize the separation of the corresponding relationships between correct matching and incorrect matching. The relevant information of the matching relationship is used to judge the offset of the monitoring device and to correct the monitoring image.

Specifically, multiple sets of feature points are filtered through the grid motion statistics GMS filtering algorithm to determine multiple sets of matching feature points with correct matching relationships.

In the multiple sets of matching feature points with correct matching relationship, according to the coordinate parameters of each matching feature point, the offset distance of each group of matching feature points between the surveillance image and the template image is calculated.

Further, after obtaining the offset distance of the surveillance image relative to the template image, when analyzing the offset distance, the server may preset the offset distance threshold, and compare the offset distance with the offset distance threshold. The number of offset distances is multiple. Therefore, the average distance corresponding to the multiple offset distances can be calculated, and the average distance can be compared with the preset offset distance threshold. If the average distance is greater than the preset offset distance threshold, it means that the monitoring The shooting device is shifted.

If the average distance is less than the preset offset distance threshold, it means that the monitoring device has not shifted. However, in order to verify the confidence of the current detection result, considering that the template images in the template set are limited, even in the template set, the same as the The template image with the highest similarity to the surveillance image, in fact, may have a low similarity between the template image and the surveillance image. Therefore, the template image with the highest similarity to the surveillance image is monitored.

When the highest similarity is less than the preset similarity threshold, it means that the confidence of the result of the current template image participating in the monitoring device offset detection is not high.

Based on this, if the average distance is less than the preset offset distance threshold, it is determined whether the similarity between the surveillance image and the template image is less than the preset similarity threshold, and if so, a detection notification is sent to the user to obtain the user's feedback information, and determine whether the monitoring equipment is offset through feedback information.

If the similarity between the monitoring image and the template image is not less than the preset similarity threshold, it is determined that the monitoring device does not shift.

It should be noted that the server outputs the template images whose similarity is less than the preset similarity threshold to the template set update queue, so that the user can monitor the template set update queue and continuously update the template set.

At the same time, the non-shifted surveillance images are saved to the template set to update the template set.

In some embodiments of the present application, when the preset offset distance is determined, a fixed value is usually adopted at present, which is difficult to satisfy the tolerance of devices of different sizes to the offset of the monitoring device.

Therefore, considering that the surveillance image includes several devices to be inspected, that is to say, it may include multiple devices to be inspected, therefore, using the inspection device with the smallest area in the current surveillance scene as the benchmark, the offset is improved. The accuracy of the detection results can meet the tolerance of different sizes of equipment to be inspected for the offset of the monitoring equipment by adopting an adaptive threshold, and improve the sensitivity of the offset detection of the monitoring equipment.

Specifically, the server determines, according to the coordinate information of several devices to be inspected in the monitoring image, the corresponding areas of several devices to be inspected in the monitoring image, and then performs multiple offsets on the device to be inspected with the smallest area. so that the user can judge the distance that the device to be inspected with the smallest area can tolerate the offset, and take the distance that can tolerate the offset as the preset offset distance. That is, based on the user's operation, the device to be inspected with the smallest area is shifted multiple times, and the distance that the device to be inspected with the smallest area can tolerate the offset is determined.

In some embodiments of the present application, after it is determined that the monitoring device is shifted, the monitoring device and the monitoring image need to be corrected.

Specifically, the server generates a coordinate transformation matrix according to the coordinate parameters of each matching feature point among the multiple sets of matching feature points with correct matching relationships.

The expression of the coordinate transformation matrix is as follows:

Among them, d _x is the displacement component in the horizontal direction, _dy is the displacement component in the vertical direction, θ is the offset rotation angle, and m is the coordinate transformation matrix.

Correct the monitoring image through the coordinate transformation matrix; and obtain the horizontal offset, vertical offset component and offset rotation angle of the monitoring device through the coordinate transformation matrix, so that the user can correct the monitoring device .

Based on the same idea, some embodiments of the present application further provide a device and a non-volatile computer storage medium corresponding to the above method.

FIG. 2 is a schematic structural diagram of a monitoring device offset detection device provided by an embodiment of the present application, and the device includes:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

Obtain surveillance images of the equipment to be inspected within a preset time interval;

From a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images of the device to be inspected under different lighting conditions;

Extracting the feature points of the surveillance image and the template image respectively;

Feature matching is performed between the feature points of the surveillance image and the feature points of the template image, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined;

The offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device.

Some embodiments of the present application provide a non-volatile computer storage medium for monitoring device offset detection, which stores computer-executable instructions, and the computer-executable instructions are set to:

Obtain surveillance images of the equipment to be inspected within a preset time interval;

From a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images of the device to be inspected under different lighting conditions;

Extracting the feature points of the surveillance image and the template image respectively;

Feature matching is performed between the feature points of the surveillance image and the feature points of the template image, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined;

The offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device.

Each embodiment in this application is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the apparatus and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for related parts.

The devices, media and methods provided in the embodiments of the present application are in one-to-one correspondence. Therefore, the devices and media also have similar beneficial technical effects to their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, therefore, The beneficial technical effects of the device and the medium will not be repeated here.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-persistent memory in computer readable media, random access memory (RAM) and/or non-volatile memory in the form of, for example, read only memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media includes both persistent and non-permanent, removable and non-removable media, and storage of information may be implemented by any method or technology. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer 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 technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission 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.

It should also be noted that the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a series of elements includes not only those elements, but also Other elements not expressly listed, or which are inherent to such a process, method, article of manufacture, or apparatus are also included. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article of manufacture, or device that includes the element.

The above descriptions are merely examples of the present application, and are not intended to limit the present application. Various modifications and variations of this application are possible for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the technical principles of this application shall fall within the protection scope of this application.

Claims (10)

1. a monitoring equipment offset detection method, is characterized in that, described method comprises: Obtain the surveillance images of the equipment to be inspected from the surveillance equipment within a preset time interval; In a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images corresponding to the equipment to be inspected under different lighting conditions; Extracting the feature points of the surveillance image and the template image respectively; Feature matching is performed between the feature points of the surveillance image and the feature points of the template image, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined; The offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device. 2. The method according to claim 1, wherein, in the pre-built template set, retrieving the template image with the highest similarity with the surveillance image specifically includes: respectively converting the surveillance image and the multiple template images from RGB space to HSV space; According to the HSV space, extracting the first V component of the surveillance image, and extracting the second V component corresponding to the plurality of template images respectively; extracting a first histogram feature of the first V component, and a second histogram feature of the second V component; performing a correlation analysis on the first histogram feature and the second histogram feature, and determining the respective degrees of similarity corresponding to the surveillance image and the plurality of template images; According to the similarity, the template image with the highest similarity with the surveillance image is determined. 3. The method according to claim 1, wherein the extracting the feature points of the surveillance image and the template image respectively comprises: Obtain the image watermark of the surveillance image; wherein, the template image does not include the image watermark; the image watermark is the identification information of the manufacturer to which the surveillance device belongs; According to the resolution of the surveillance image, determine the pixel area occupied by the image watermark in the surveillance image; The pixel area is culled to extract a designated area that does not contain an image watermark in the surveillance image; Through the fast feature point extraction model ORB, the designated area of the surveillance image and the feature points of the template image are respectively extracted. 4. The method according to claim 3, wherein the feature points of the designated area of the surveillance image and the template image are respectively extracted by the fast feature point extraction model ORB, specifically comprising: Building scale image pyramids for the designated area and the template image respectively; Perform grid processing on the scale image pyramid, and perform feature point detection on each small grid in each layer of the pyramid image through the FAST feature point detection algorithm, and extract the feature points of the designated area and the template image respectively. 5 . The method according to claim 1 , wherein the feature matching of the feature points of the surveillance image and the feature points of the template image is performed to determine that the feature points of the surveillance image are relative to the The offset distance of the feature points of the template image, specifically including: Feature matching is performed on the feature points of the surveillance image and the feature points of the template image through a fast nearest search and matching algorithm to obtain multiple sets of feature points with matching relationships; Filter the multiple groups of feature points through the grid motion statistics GMS filtering algorithm to determine multiple groups of matching feature points with correct matching relationships; In the multiple sets of matching feature points with correct matching relationship, according to the coordinate parameters of each matching feature point, the offset distance of each group of matching feature points between the surveillance image and the template image is calculated. 6 . The method according to claim 5 , wherein the analyzing the offset distance according to a preset rule, so as to perform offset detection on the monitoring device, specifically comprising: 6 . calculating an average distance corresponding to a plurality of the offset distances; If the average distance is less than a preset offset distance threshold, then determine whether the similarity between the surveillance image and the template image is less than a preset similarity threshold; If so, send a detection notification to the user; Acquire feedback information of the user, and determine whether the monitoring device is offset according to the feedback information. 7. The method according to claim 6, wherein the method further comprises: If the average distance is greater than a preset offset distance threshold, it is determined that the monitoring device is offset; In the multiple sets of matching feature points with the correct matching relationship, a coordinate transformation matrix is generated according to the coordinate parameters of each matching feature point; Correcting the surveillance image through the coordinate transformation matrix; and acquiring the horizontal offset, vertical offset component and offset rotation angle of the surveillance device through the coordinate transformation matrix, to obtain causing the user to make corrections to the monitoring device. 8. The method according to claim 6, wherein the monitoring image includes several devices to be inspected, and the method further comprises: According to the coordinate information of the several devices to be inspected in the surveillance image, determine the respective areas of the several devices to be inspected in the surveillance image; Offset the equipment to be inspected with the smallest area for multiple times, so that the user can judge the distance that the equipment to be inspected with the smallest area can tolerate the offset, and use the distance tolerable offset as the preset distance. Set the offset distance. 9. A monitoring device offset detection device, wherein the device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Obtain surveillance images of the equipment to be inspected within a preset time interval; In a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images of the equipment to be inspected under different lighting conditions; Extracting the feature points of the surveillance image and the template image respectively; Feature matching is performed between the feature points of the surveillance image and the feature points of the template image, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined; The offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device. 10. A monitoring device offset detection non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are set to: Obtain surveillance images of the equipment to be inspected within a preset time interval; In a pre-built template set, retrieve the template image with the highest similarity to the surveillance image; wherein the template set includes a plurality of template images of the equipment to be inspected under different lighting conditions; Extracting the feature points of the surveillance image and the template image respectively; Feature matching is performed between the feature points of the surveillance image and the feature points of the template image, and the offset distance of the feature points of the surveillance image relative to the feature points of the template image is determined; The offset distance is analyzed according to a preset rule, so as to perform offset detection on the monitoring device.

Images