Abnormal photovoltaic panel detection method and system and storage medium

Abstract

The invention discloses an abnormal photovoltaic panel detection method and system and a storage medium, and the method is characterized in that the method comprises the steps: obtaining a to-be-detected infrared image set and a to-be-detected visible light image set of a photovoltaic panel, enabling the to-be-detected infrared image set and the to-be-detected visible light image set to be in one-to-one correspondence, and enabling the to-be-detected infrared image set and the to-be-detected visible light image set to be in one-to-one correspondence; every two pictures in one-to-one correspondence are obtained at the same position; preprocessing the to-be-detected infrared image set and the to-be-detected visible light image set; inputting the to-be-detected infrared image set into a pre-trained infrared image photovoltaic panel detection model, inputting the to-be-detected visible light image set into a pre-trained visible light photovoltaic panel detection model, detecting an abnormal photovoltaic panel, and obtaining position information of the abnormal photovoltaic panel in the image; and the position information is shared in the to-be-detected infrared image set and the to-be-detected visible light image set, so that the identification efficiency of the abnormal photovoltaic panel is improved.

Description

Technical Field

[0001] The present invention relates to the field of electric power, and in particular to an abnormal photovoltaic panel detection method, system and storage medium. Background Art

[0002] Photovoltaic panels, also known as solar panels, are devices that use sunlight to generate electricity. They convert solar energy into direct current electricity through the photovoltaic effect. Photovoltaic panels are primarily made of semiconductor materials, usually silicon, which can be either monocrystalline or polycrystalline. Photovoltaic panels are widely used in residential, commercial, and industrial applications to generate electricity and provide supplemental energy to the grid.

[0003] If the abnormal situation of photovoltaic panels is not discovered in time before it develops to a serious level, measures will not be taken to repair or replace them, causing further damage to the photovoltaic panels and reducing the life of the photovoltaic panels; hot spots, local temperature increases, etc. may affect the working efficiency and power generation capacity of the photovoltaic panels.

[0004] Currently, the detection of abnormal conditions of photovoltaic panels usually relies on manual inspection and confirmation, which increases the maintenance time and the workload and cost of maintenance personnel. Summary of the invention

[0005] To achieve the above-mentioned purpose and other related purposes, the present invention discloses a method for detecting abnormal photovoltaic panels, comprising: Acquire an infrared image set to be detected and a visible light image set to be detected of the photovoltaic panel, wherein the photos in the infrared image set to be detected and the visible light image set to be detected correspond to each other, and every two photos in the one-to-one correspondence are acquired at the same position; Preprocessing the infrared image set to be detected and the visible light image set to be detected; Input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, input the visible light image set to be detected into a pre-trained visible light photovoltaic panel detection model, detect abnormal photovoltaic panels, and obtain the location information of the abnormal photovoltaic panels in the image; The position information is shared in the infrared image set to be detected and the visible light image set to be detected.

[0006] Furthermore, the establishment of the infrared image photovoltaic panel detection model includes: Obtain a set of infrared images to be trained; Preprocess the infrared image set to be trained; Annotate abnormal photovoltaic panels in the preprocessed infrared image set to be trained; The resnet18 classification model is used to train the labeled infrared image set to be trained.

[0007] Furthermore, when obtaining the infrared image set to be detected and the visible light image set to be detected, the following steps are included: Get the length of the photographed photovoltaic panel; If the length of the photovoltaic panel is greater than a preset length threshold, photographing the photovoltaic panel in different directions; Otherwise, the photovoltaic panel is photographed once to fully present the photovoltaic panel in the image.

[0008] Furthermore, the preprocessing of the infrared image set to be detected and the visible light image set to be detected includes: Pre-crop the infrared image set to be detected; Pre-cropping the visible light image set to be detected according to the pre-cropped infrared image set to be detected; Performing target detection on the pre-cropped infrared image set to be detected and the visible light image set to be detected, and obtaining the positions of the photovoltaic panels in the infrared image set to be detected and the visible light image set to be detected; According to the position of the photovoltaic panel, the photovoltaic panel is intercepted from the infrared image set to be detected and the visible light image set to be detected; The captured image set is segmented to obtain the outline of the photovoltaic panel.

[0009] Furthermore, the pre-cropping of the infrared image set to be detected includes: Obtain the ratios Y1 and Y2 of the edge length of the photovoltaic panel and the corresponding edge of the infrared image to be detected, where: Y1=y1 / y, where Y1 is the ratio of the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected to the side length of the infrared image to be detected, y1 is the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected, and y is the side length of the infrared image to be detected; Y2=y2 / y, y2 is the distance from the lower edge of the photovoltaic panel to the upper edge of the infrared image to be detected; The portion of the infrared image to be detected that is within the range from Y1 to Y2 is retained.

[0010] Further, pre-cropping the visible light image set to be detected according to the pre-cropped infrared image set to be detected includes: According to each pre-cropped position of the infrared image to be detected, the same position of the corresponding visible light image to be detected is cropped.

[0011] Furthermore, the step of performing image segmentation on the captured image set to obtain the photovoltaic panel contour includes: The image set was segmented using DeeplabV3 and processed using morphology.

[0012] Furthermore, the performing target detection on the pre-cropped infrared image set to be detected and the visible light image set to be detected includes: The yolov7 algorithm is used to perform target detection on the pre-cropped infrared image set and visible light image set to be detected.

[0013] In a second aspect, the present invention further provides an abnormal photovoltaic panel detection system, comprising: An image acquisition module is used to acquire an infrared image set and a visible light image set of the photovoltaic panel to be detected, wherein the photos in the infrared image set and the visible light image set correspond to each other, and every two photos in the one-to-one correspondence are acquired at the same position; An image preprocessing module is used to preprocess the infrared image set to be detected and the visible light image set to be detected; The abnormal target confirmation module is used to input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, and input the visible light image set to be detected into a pre-trained visible light photovoltaic panel detection model, detect abnormal photovoltaic panels, and obtain the location information of the abnormal photovoltaic panels in the image; share the location information in the infrared image set to be detected and the visible light image set to be detected.

[0014] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above method when executed by a processor.

[0015] By adopting the above technical solution, when it is necessary to identify abnormal photovoltaic panels, first obtain the infrared image set to be detected and the visible light image set to be detected, then preprocess the infrared image set to be detected and the visible light image set to be detected, and detect them through the trained model to detect the location of the abnormal photovoltaic panel, and mark the location information in the infrared image to be detected and the visible light image to be detected, so that the staff can replace the abnormal photovoltaic panel according to the location, thereby improving efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which:

[0017] Figure 1 The present invention is a flow chart of a method for detecting abnormal photovoltaic panels. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0019] Reference Figure 1 The embodiment of the present invention provides a method for detecting abnormal photovoltaic panels, comprising the following steps: S1: Acquire an infrared image set to be detected and a visible light image set to be detected of a photovoltaic panel, wherein the photos in the infrared image set to be detected and the visible light image set to be detected correspond to each other one by one, and every two one-to-one corresponding photos are acquired at the same position.

[0020] Specifically, when obtaining the infrared image set and visible light image set to be detected, the unmanned heterogeneous system is used to automatically build the flight route and distribute the task to the DJI M30T drone series to perform the flight mission. The flight mission is mainly to collect photovoltaic panels in the factory area and take infrared images to be detected through the onboard infrared camera. The data format is jpg, and one photo is about 4M. The drone takes infrared images and visible light images at the same location at the same time, and the infrared images and visible light images are matched one by one, which is convenient for the subsequent comparison of the infrared images and visible light images of the same photovoltaic panel.

[0021] Wherein, when obtaining the infrared image set to be detected and the visible light image set to be detected, the following steps are included: Get the length of the photographed photovoltaic panel; If the length of the photovoltaic panel is greater than a preset length threshold, photographing the photovoltaic panel in different directions; Otherwise, the photovoltaic panel is photographed once to fully present the photovoltaic panel in the image.

[0022] Specifically, the specific value of the length threshold is set by those skilled in the art and will not be described in detail here. In this embodiment, when the length of the photovoltaic panel is greater than the length threshold, images of the photovoltaic panel are acquired from both sides to ensure that the entire appearance of the photovoltaic panel can be photographed, thereby improving the detection accuracy.

[0023] S2: Preprocess the infrared image set to be detected and the visible light image set to be detected.

[0024] The preprocessing of the infrared image set to be detected and the visible light image set to be detected includes: Pre-crop the infrared image set to be detected; Pre-cropping the visible light image set to be detected according to the pre-cropped infrared image set to be detected; Specifically, when cropping the visible light image set to be detected, according to each position where the infrared image to be detected is pre-cropped, the same position of the corresponding visible light image to be detected is cropped.

[0025] Performing target detection on the pre-cropped infrared image set to be detected and the visible light image set to be detected, and obtaining the positions of the photovoltaic panels in the infrared image set to be detected and the visible light image set to be detected; Specifically, the yolov7 algorithm is used to perform target detection on the pre-cropped infrared image set and the visible light image set to be detected, and the photovoltaic panels located in the image set are identified, further reducing the amount of data processing during subsequent detection.

[0026] According to the position of the photovoltaic panel, the photovoltaic panel is intercepted from the infrared image set to be detected and the visible light image set to be detected; The captured image set is segmented to obtain the outline of the photovoltaic panel.

[0027] Specifically, DeeplabV3 is used to segment the image set, and morphology is used for processing to improve the image segmentation effect.

[0028] Wherein, the pre-cropping of the infrared image set to be detected includes: Obtain the ratios Y1 and Y2 of the edge length of the photovoltaic panel and the corresponding edge of the infrared image to be detected, where: Y1=y1 / y, where Y1 is the ratio of the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected to the side length of the infrared image to be detected, y1 is the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected, and y is the side length of the infrared image to be detected; Y2=y2 / y, y2 is the distance from the lower edge of the photovoltaic panel to the upper edge of the infrared image to be detected; The portion of the infrared image to be detected that is within the range from Y1 to Y2 is retained.

[0029] As described above, the position of the photovoltaic panel in the infrared image to be detected is marked and cropped through Y1 and Y2, and the photovoltaic panel is pre-cropped, which reduces the influence of other environments in the image on subsequent recognition and improves the accuracy of recognition.

[0030] S3: Input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, input the visible light image set to be detected into a pre-trained visible light photovoltaic panel detection model, detect abnormal photovoltaic panels, and obtain the location information of the abnormal photovoltaic panels in the image.

[0031] The establishment of the infrared image photovoltaic panel detection model includes: Obtain a set of infrared images to be trained; Preprocess the infrared image set to be trained; Annotate abnormal photovoltaic panels in the preprocessed infrared image set to be trained; The resnet18 classification model is used to train the labeled infrared image set to be trained.

[0032] Specifically, the infrared image set to be trained is historical data, and the abnormal photovoltaic panels manually annotated in history are used as historical data; the method for preprocessing the infrared image set to be trained is the same as the method for processing the infrared image set to be detected, which will not be repeated here.

[0033] Wherein, the establishment of the visible light photovoltaic panel detection model includes: Obtain a visible light image set to be trained; Preprocessing the visible light image set to be trained; Annotate abnormal photovoltaic panels in the preprocessed infrared image set to be trained; The resnet18 classification model is used to train the labeled infrared image set to be trained.

[0034] Among them, the visible light image set to be trained is historical data, and the abnormal photovoltaic panels in the visible light images manually annotated in the history are taken as historical data; the method for preprocessing the visible light image set to be trained is the same as the above-mentioned method for processing the visible light image set to be detected, and will not be repeated here.

[0035] As mentioned above, the types of abnormalities of photovoltaic panels that can be detected by infrared images and visible light images are different. Infrared images can better show the bright spots and dark spots of photovoltaic panels during use, and visible light images can better distinguish the dirt on photovoltaic panels. Therefore, the combination of the two image types can more comprehensively detect abnormal photovoltaic panels.

[0036] S4: Sharing the position information in the infrared image set to be detected and the visible light image set to be detected.

[0037] As mentioned above, the position information in this embodiment is the block number where the abnormal photovoltaic panel is located from left to right in the image. Other position information can also be used to locate the abnormal photovoltaic panel, which will not be repeated here.

[0038] An embodiment of the present invention also provides an abnormal photovoltaic panel detection system. When it is necessary to identify an abnormal photovoltaic panel, firstly, a set of infrared images to be detected and a set of visible light images to be detected are obtained, and then the set of infrared images to be detected and the set of visible light images to be detected are preprocessed, and then detected through a trained model to detect the location of the abnormal photovoltaic panel, and the location information is marked in the infrared image to be detected and the visible light image to be detected, so that the staff can replace the abnormal photovoltaic panel according to the location, thereby improving efficiency.

[0039] An image acquisition module is used to acquire an infrared image set and a visible light image set of the photovoltaic panel to be detected, wherein the photos in the infrared image set and the visible light image set correspond to each other, and every two photos in the one-to-one correspondence are acquired at the same position; An image preprocessing module is used to preprocess the infrared image set to be detected and the visible light image set to be detected; The abnormal target confirmation module is used to input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, detect abnormal photovoltaic panels, obtain the location information of the abnormal photovoltaic panels in the infrared image, and mark the location information in the infrared image and the visible light image corresponding to the infrared image.

[0040] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above method is implemented.

[0041] For the convenience of description, the above devices are described in terms of functions and are divided into various units and modules. Of course, when implementing the present application, the functions of each unit and module can be implemented in the same or multiple software and / or hardware.

[0042] Those skilled in the art will appreciate that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined.

[0043] For the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0044] It can be known from the description of the above implementation modes that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute the methods described in the various implementation modes of the present application or certain parts of the implementation modes.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting abnormal photovoltaic panels, characterized in that: include: Acquire an infrared image set to be detected and a visible light image set to be detected of the photovoltaic panel, wherein the photos in the infrared image set to be detected and the visible light image set to be detected correspond to each other, and every two photos in the one-to-one correspondence are acquired at the same position; Preprocessing the infrared image set to be detected and the visible light image set to be detected; Input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, input the visible light image set to be detected into a pre-trained visible light photovoltaic panel detection model, detect abnormal photovoltaic panels, and obtain the location information of the abnormal photovoltaic panels in the image; The position information is shared in the infrared image set to be detected and the visible light image set to be detected.

2. The method according to claim 1, characterized in that The establishment of the infrared image photovoltaic panel detection model includes: Obtain a set of infrared images to be trained; Preprocess the infrared image set to be trained; Annotate abnormal photovoltaic panels in the preprocessed infrared image set to be trained; The resnet18 classification model is used to train the labeled infrared image set to be trained.

3. The method according to claim 1, characterized in that When acquiring the infrared image set to be detected and the visible light image set to be detected, the following steps are included: Get the length of the photographed photovoltaic panel; If the length of the photovoltaic panel is greater than a preset length threshold, photographing the photovoltaic panel in different directions; Otherwise, the photovoltaic panel is photographed once to fully present the photovoltaic panel in the image.

4. The method according to claim 1, characterized in that: The preprocessing of the infrared image set to be detected and the visible light image set to be detected comprises: Pre-crop the infrared image set to be detected; Pre-cropping the visible light image set to be detected according to the pre-cropped infrared image set to be detected; Performing target detection on the pre-cropped infrared image set to be detected and the visible light image set to be detected, and obtaining the positions of the photovoltaic panels in the infrared image set to be detected and the visible light image set to be detected; According to the position of the photovoltaic panel, the photovoltaic panel is intercepted from the infrared image set to be detected and the visible light image set to be detected; The captured image set is segmented to obtain the outline of the photovoltaic panel.

5. The method according to claim 4, characterized in that The pre-cropping of the infrared image set to be detected comprises: Obtain the ratios Y1 and Y2 of the edge length of the photovoltaic panel and the corresponding edge of the infrared image to be detected, where: Y1=y1 / y, where Y1 is the ratio of the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected to the side length of the infrared image to be detected, y1 is the distance from the upper edge of the photovoltaic panel to the upper edge of the infrared image to be detected, and y is the side length of the infrared image to be detected; Y2=y2 / y, y2 is the distance from the lower edge of the photovoltaic panel to the upper edge of the infrared image to be detected; The portion of the infrared image to be detected that is within the range from Y1 to Y2 is retained.

6. The method according to claim 4, characterized in that Pre-cropping the visible light image set to be detected according to the pre-cropped infrared image set to be detected includes: According to each pre-cropped position of the infrared image to be detected, the same position of the corresponding visible light image to be detected is cropped.

7. The method according to claim 4, characterized in that The image segmentation of the intercepted image set to obtain the photovoltaic panel contour includes: The image set was segmented using DeeplabV3 and processed using morphology.

8. The method according to claim 4, characterized in that The target detection on the pre-cropped infrared image set to be detected and the visible light image set to be detected comprises: The yolov7 algorithm is used to perform target detection on the pre-cropped infrared image set and visible light image set to be detected.

9. An abnormal photovoltaic panel detection system, characterized in that: include: An image acquisition module is used to acquire an infrared image set and a visible light image set of the photovoltaic panel to be detected, wherein the photos in the infrared image set and the visible light image set correspond to each other, and every two photos in the one-to-one correspondence are acquired at the same position; An image preprocessing module is used to preprocess the infrared image set to be detected and the visible light image set to be detected; The abnormal target confirmation module is used to input the infrared image set to be detected into a pre-trained infrared image photovoltaic panel detection model, and input the visible light image set to be detected into a pre-trained visible light photovoltaic panel detection model, detect abnormal photovoltaic panels, and obtain the location information of the abnormal photovoltaic panels in the image; The position information is shared in the infrared image set to be detected and the visible light image set to be detected.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method described in any one of claims 1 to 8 is implemented.

Images