CN114022482B - Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium - Google Patents
Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN114022482B CN114022482B CN202210012032.8A CN202210012032A CN114022482B CN 114022482 B CN114022482 B CN 114022482B CN 202210012032 A CN202210012032 A CN 202210012032A CN 114022482 B CN114022482 B CN 114022482B
- Authority
- CN
- China
- Prior art keywords
- hot spots
- photovoltaic panel
- determining
- punctiform
- mirror image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a method, a device and equipment for detecting punctiform hot spots of a photovoltaic panel and a readable storage medium, wherein the method comprises the following steps: acquiring an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected; constructing mirror image gradient non-similarity measurement in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image; based on the size characteristics of the punctiform hot spots and the omnidirectional mirror image gradient difference characteristics on the pixel points, a local radiation structure with a set size scale from inside to outside is constructed by taking the pixel points of the punctiform hot spots as centers; constructing an omnidirectional mirror image gradient difference measurement according to the mirror image gradient non-similarity measurement and the local radiation structure; and determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference measurement. The method can improve the accuracy of detecting the punctiform hot spots.
Description
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to a method and an apparatus for detecting a punctiform hot spot of a photovoltaic panel, an electronic device, and a computer-readable storage medium.
Background
In recent years, the energy crisis has become more and more severe worldwide, and this has become one of the important factors that restrict the economic development of the world. On one hand, the environmental pollution problem is aggravated by a large amount of toxic and harmful gas generated by the combustion of traditional energy coal, petroleum and natural gas; on the other hand, rapid depletion of traditional energy reserves has made it difficult to meet the sustainable development of global economy in the near future. Therefore, the trend of vigorously developing clean and renewable energy sources is raised in various countries around the world, and the energy source safety new strategy that new energy sources such as photovoltaic power generation and wind power generation replace traditional energy sources is developed. Photovoltaic power generation, as a clean renewable energy form converting solar energy into electric energy, is an important ring in the development of new energy nowadays.
Since 2021, China pays attention to the development of photovoltaic power generation and puts forward a new energy strategy of 'carbon peak reaching and carbon neutralization'. With the rapid increase of the photovoltaic loading capacity in recent years in China, the scale of a photovoltaic power station is getting larger and larger, the daily operation and maintenance of a large photovoltaic power station face brand new challenges, how to rapidly overhaul faults of photovoltaic components and reduce economic loss of the power station become a pain point for the development of the photovoltaic industry.
In an actual operation and maintenance scene, the hot spot effect is caused when the photovoltaic panel breaks down, and the hot spot effect of the photovoltaic panel can be effectively induced by the infrared image of the photovoltaic panel shot by the infrared camera. Of all types of photovoltaic panel hotspots, punctiform hotspots are the most common type of hotspot.
The prior patent CN 113076816A-infrared and visible light image-based solar photovoltaic module hot spot identification method provides a method for constructing a photovoltaic module infrared image and visible light image data set to construct a hot spot detection model, but it is difficult to detect a point hot spot with sparse features, and meanwhile, it is easy to cause false detection to high brightness noise in an image.
How to rapidly and accurately detect the dotted hot spots of the photovoltaic panel directly affects the operation and maintenance level of the photovoltaic power station, that is, how to overcome the technical defects, is a problem to be solved urgently by those skilled in the art.
Disclosure of Invention
The application aims to provide a method and a device for detecting punctiform hot spots of a photovoltaic panel, electronic equipment and a computer readable storage medium.
To achieve the above object, the present application provides, in a first aspect, a method for detecting a punctiform hot spot of a photovoltaic panel, the method including: acquiring an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected; constructing mirror image gradient non-similarity measurement in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image; based on the size characteristics of the punctiform hot spots and the omnidirectional mirror image gradient difference characteristics on the pixel points, a local radiation structure with a set size scale from inside to outside is constructed by taking the pixel points of the punctiform hot spots as centers; constructing an omnidirectional mirror image gradient difference measurement according to the mirror image gradient non-similarity measurement and the local radiation structure; and determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference measurement.
Optionally, the method further includes:
carrying out median filtering on the infrared image to be detected by using a median filter to obtain a filtered image;
carrying out normalization processing on the filtered image to obtain a preprocessed image;
correspondingly, determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference measurement, and the method comprises the following steps:
and determining whether the preprocessed infrared image contains the punctiform hot spots according to the omnidirectional mirror image gradient difference measurement.
Optionally, the method further includes:
determining a target filtering scale of the median filter according to the size of salt and pepper noise and the size of the photovoltaic panel appearing in the infrared image to be detected;
correspondingly, performing median filtering on the infrared image to be detected by using a median filter to obtain a filtered image, and the method comprises the following steps:
and performing median filtering on the infrared image to be detected by using a median filter with the scale as a target filtering scale to obtain a filtered image.
Optionally, the normalizing the filtered image to obtain a preprocessed image includes:
determining the gray value of each pixel point forming the filtered image, and determining the minimum gray value and the maximum gray value according to the gray value of each pixel point;
respectively calculating a first gray difference between the actual gray value and the minimum gray value of each pixel point;
calculating a second gray difference between the maximum gray value and the minimum gray value;
respectively taking the quotient of the first gray value and the second gray value difference corresponding to each pixel point as a new gray value of each pixel point;
and generating a preprocessed image according to the new gray value of each pixel point.
Optionally, determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference metric includes:
determining the confidence coefficient of each pixel point of the infrared image to be detected, which belongs to the pixel points of the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement to obtain the global confidence coefficient;
determining a segmentation threshold for screening the punctate hot spots based on the global confidence;
determining pixel points with actual confidence degrees exceeding a segmentation threshold as target pixel points belonging to the punctiform hot spots;
and determining the contained point hot spots according to the target pixel points.
Optionally, determining a segmentation threshold for screening the punctate hot spots based on the global confidence includes:
determining the confidence coefficient mean value of each pixel point according to the global confidence coefficient;
determining a confidence standard deviation according to the actual confidence and the confidence mean value of each pixel point;
and determining a segmentation threshold value for screening the punctate hot spots according to the confidence coefficient mean value and the confidence coefficient standard deviation.
Optionally, the different directions of the mirror gradient dissimilarity measure include: horizontal direction, vertical direction, one-three quadrant direction and two-four quadrant direction.
To achieve the above object, the present application provides, in a second aspect, a dotted hot spot detection device for a photovoltaic panel, the device including: the infrared image acquisition unit to be detected is configured to acquire an infrared image to be detected, which is obtained by shooting the photovoltaic panel to be detected; the mirror image gradient non-similarity measurement construction unit is configured to construct mirror image gradient non-similarity measurements in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image; the local radiation structure construction unit is configured to construct a local radiation structure with a set size scale from inside to outside by taking the pixel point of the point hot spot as a center based on the size characteristic of the point hot spot and the omnidirectional mirror image gradient difference characteristic on the pixel point; an omnidirectional mirror image gradient difference metric construction unit configured to construct an omnidirectional mirror image gradient difference metric based on the mirror image gradient dissimilarity metric and the local radiation structure; and the dot-shaped hot spot detection unit is configured to determine whether the infrared image to be detected contains dot-shaped hot spots according to the omnidirectional mirror image gradient difference metric.
Optionally, the apparatus further comprises:
the median filtering unit is configured to perform median filtering on the infrared image to be detected by using a median filter to obtain a filtered image;
the normalization unit is configured to perform normalization processing on the filtered image to obtain a preprocessed image;
correspondingly, the punctiform hot spot detection unit is further configured to:
and determining whether the preprocessed infrared image contains the punctiform hot spots according to the omnidirectional mirror image gradient difference measurement.
Optionally, the apparatus further comprises:
a filtering scale determining unit configured to determine a target filtering scale of the median filter according to the size of the salt and pepper noise and the photovoltaic panel appearing in the infrared image to be detected;
correspondingly, the median filtering unit is further configured to:
and performing median filtering on the infrared image to be detected by using a median filter with the scale as a target filtering scale to obtain a filtered image.
Optionally, the normalization unit is further configured to:
determining the gray value of each pixel point forming the filtered image, and determining the minimum gray value and the maximum gray value according to the gray value of each pixel point;
respectively calculating a first gray difference between the actual gray value and the minimum gray value of each pixel point;
calculating a second gray difference between the maximum gray value and the minimum gray value;
respectively taking the quotient of the first gray value and the second gray value difference corresponding to each pixel point as a new gray value of each pixel point;
and generating a preprocessed image according to the new gray value of each pixel point.
Optionally, the punctiform hot spot detection unit includes:
the global confidence determining subunit is configured to determine the confidence of pixel points of the infrared image to be detected, which belong to the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement, so as to obtain a global confidence;
a segmentation threshold determination subunit configured to determine a segmentation threshold for screening the punctate hot spots based on the global confidence;
a segmentation subunit configured to determine pixel points having an actual confidence exceeding a segmentation threshold as target pixel points belonging to the punctate hot spots;
and the dot-shaped hot spot determining subunit is configured to determine the contained dot-shaped hot spots according to the target pixel points.
Optionally, the segmentation threshold determination subunit is further configured to:
determining the confidence coefficient mean value of each pixel point according to the global confidence coefficient;
determining a confidence standard deviation according to the actual confidence and the confidence mean value of each pixel point;
and determining a segmentation threshold value for screening the punctate hot spots according to the confidence coefficient mean value and the confidence coefficient standard deviation.
Optionally, the different directions of the mirror gradient dissimilarity measure include: horizontal direction, vertical direction, one-three quadrant direction and two-four quadrant direction.
To achieve the above object, the present application provides, in a third aspect, an electronic apparatus comprising:
a memory for storing a computer program;
a processor configured to implement the steps of the method for detecting punctiform hot spots of a photovoltaic panel as described in any of the embodiments of the first aspect described above when executing the computer program stored in the memory.
In order to achieve the above object, the present application provides, in a fourth aspect, a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of detecting a punctiform hot spot of a photovoltaic panel as described in any of the embodiments of the first aspect.
Compared with the prior art, the photovoltaic panel dotted hot spot detection method provided by the application can be used for providing mirror image gradient non-similarity measurement by fully representing the local gradient characteristics of dotted hot spot pixel points, and can be used for effectively enhancing the signal intensity of dotted hot spots in an image; by fully considering the gradient change characteristics of the punctiform hot spots in different directions and different spatial scales, a local radiation structure is provided, so that the signal enhancement capability of an algorithm on the punctiform hot spot pixel points is ensured while the change of the punctiform hot spot scales is adapted; by fully considering the difference of the dot hot spot pixel point and the background edge pixel point on the image gradient, the omnidirectional image gradient difference measurement is provided, the interference of the background edge pixel point on the detection result can be effectively inhibited, and the detection precision is improved.
The application also provides a photovoltaic panel punctiform hot spot detection device, electronic equipment and computer readable storage medium simultaneously, has above-mentioned beneficial effect, and no longer gives details here.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a method for detecting a punctiform hot spot of a photovoltaic panel according to an embodiment of the present disclosure;
fig. 2 is an original infrared image to be detected without preprocessing provided by an embodiment of the present application;
fig. 3 is a schematic view of a local radiation structure provided in an embodiment of the present application;
FIG. 4 is a schematic view of another local radiating structure provided by an embodiment of the present application;
fig. 5 is a processed infrared image obtained by visualizing the omnidirectional mirror image gradient difference metric proposed in the embodiment of the present application in combination with fig. 2;
fig. 6 is a flowchart of an image preprocessing method in the method for detecting a punctiform hot spot of a photovoltaic panel according to the embodiment of the present application;
fig. 7 is a flowchart of a dotted hot spot detection method in the method for detecting a dotted hot spot of a photovoltaic panel according to the embodiment of the present application;
fig. 8 is a block diagram of a structure of a dotted hot spot detection device of a photovoltaic panel according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Example one
Referring to fig. 1, fig. 1 is a flowchart of a dotted hot spot detection of a photovoltaic panel according to an embodiment of the present application, which includes the following steps:
step 101: acquiring an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected;
this step is intended to obtain an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected, by an executing main body (for example, a local server or a cloud server for data processing and analysis) adapted to execute the photovoltaic panel punctiform hot spot detection method provided by the present application.
The infrared image to be detected can be an image obtained by shooting a solar photovoltaic panel below by an infrared camera arranged on the unmanned aerial vehicle, or an image obtained by shooting through other ways, and the image is not specifically limited here.
In order to improve the detection effect of whether the infrared image to be detected contains the punctiform hot spots as much as possible, the shot original infrared image can be subjected to various preprocessing operations before the actual infrared image is analyzed, so that noise or other types of interference information in the image can be removed as much as possible, and the punctiform hot spots contained in the image can be highlighted as much as possible. For example, performing gray scale processing, normalizing by gray scale, binarization processing, deblurring, and the like, one specific preprocessing method may be: the method comprises the steps of firstly carrying out median filtering on an infrared image to be detected by utilizing a median filter to obtain a filtered image, then carrying out normalization processing on the filtered image to obtain a preprocessed image, and removing salt and pepper noise and uneven gray distribution which are commonly contained in the infrared image obtained by aerial photography of the photovoltaic panel through the mode, so that the follow-up analysis effect on the preprocessed image is improved.
Step 102: constructing mirror image gradient non-similarity measurement in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image;
on the basis of step 101, in this step, mirror image gradient non-similarity measurements in different directions and different orders are constructed by the execution main body, so that the mirror image gradient characteristics of the edge of the photovoltaic panel on the pixel points of the infrared image are fully embodied by means of the constructed mirror image gradient non-similarity measurements.
As shown in fig. 2, the main characteristic of the punctate hot spot in the infrared image is represented by the difference in gray distribution between the hot spot pixel point and the neighborhood pixel point. That is, in fig. 2, the black uniform portion near the right side is a background portion not including the photovoltaic panel, the gray color blocks uniformly distributed in an obvious block shape on the left side are normal photovoltaic panels, and an obvious white bright spot on the photovoltaic panel on the lower side is a spot hot spot concerned by the present application. Compared with the normal photovoltaic panel pixel point with higher gray level and the background pixel point with more uniform gray level distribution, the local neighborhood of the gray level of the point hot spot pixel point shows the discontinuity of the gray level distribution. Edge pixel points at the junction of the photovoltaic panel and the background also have discontinuous gray distribution in a local area, and the false detection of the edge pixels is easily caused only by considering the discontinuity of the gray distribution. Different from the background edge pixel points, the gray gradient change of the point hot spot pixel points in each mirror image direction has consistency.
Therefore, according to the characteristics of the dotted hot spots of the photovoltaic panel, in order to eliminate the interference of the background edge pixel points of various types, the non-similarity measurement of the image gradients of the pixel points in different directions and different orders is constructed on the basis of the image gradient characteristics of the background edge pixel points of different types possibly existing in the infrared image and in consideration of the variation differences of the image gradients of the edge pixel points and the dotted hot spot pixel points in different directions and different spatial scales.
The different directions of the mirror image gradient non-similarity measurement comprise a horizontal direction, a vertical direction, a first-three-quadrant direction and a second-four-quadrant direction, and the specific implementation mode of the first-order mirror image gradient non-similarity measurement in the corresponding direction is as follows:
wherein (A), (B), (C), (D), (C), (B), (C)x, y) Indicating the coordinate position of the current pixel point in the image,I() And expressing the gray value of the corresponding pixel point.
The specific implementation manner of the mirror gradient dissimilarity measure in different directions and different orders can be further seen in the following formula:
wherein the content of the first and second substances,re {1,2, …, R } is the order of the mirror gradient dissimilarity measure, R is an optional maximum order, and in particular embodiments, the mirror gradient dissimilarity measure may be constructed with the maximum order of R taken to be 4.
Step 103: based on the size characteristics of the punctiform hot spots and the omnidirectional mirror image gradient difference characteristics on the pixel points, a local radiation structure with a set size scale from inside to outside is constructed by taking the pixel points of the punctiform hot spots as centers;
the step aims to construct a local radiation structure with a set size scale from inside to outside by taking the pixel point of the executing main body for constructing the point hot spot as a center, so that the size characteristic of the point hot spot and the omnidirectional mirror image gradient difference characteristic on the pixel point are combined more fully by means of the constructed local radiation structure.
Considering that the characteristic of the dot-shaped hot spot of the photovoltaic panel is mainly expressed in the local area taking the hot spot pixel point as the center, the dot-shaped hot spot center pixel point is used for constructing the local structure of the infrared image shown in fig. 3.
Furthermore, considering that the local gray distribution difference between the point-shaped hot spot pixel point and the background uniform pixel point and the background edge pixel point is mainly reflected on the difference of the image gradient distribution, the image gradient distribution of the background uniform pixel point is close to zero, and the background edge pixel point shows image gradient non-similarity only in a specific direction. In contrast, the dot-shaped hot spot pixels show mirror image gradient dissimilarity in different directions and different scales.
Thus, the inside-out dimension of the structure is from the center of the pixel of the punctiform hot spot
The local radiation structure of (a) is shown by the following formula:
wherein the content of the first and second substances,
，
，
，
respectively representing the radiation structures in the horizontal direction, the vertical direction, the one-three quadrant direction and the two-four quadrant direction. The specific implementation modes of the radiation structures in the horizontal direction, the vertical direction, the first three-quadrant direction and the second four-quadrant direction are respectively as follows:
in one embodiment, the localized radiating structures are constructed as shown in FIG. 4, in which
Representing the coordinates of the pixel points in the corresponding radiation structure.
Step 104: constructing an omnidirectional mirror image gradient difference measurement according to the mirror image gradient non-similarity measurement and the local radiation structure;
on the basis of step 102 and step 103, this step is intended to construct by the executing entity an omnidirectional mirror image gradient difference metric further based on the mirror image gradient non-similarity metric and the local radiation structure, so as to better identify a punctate hot spot in the infrared image by means of the constructed omnidirectional mirror image gradient difference metric.
The characteristic that the mirror image gradients of the point-shaped hot spot pixel points in a certain spatial scale range have direction consistency is considered, so that the gray gradients on all scales can be integrated along different directions. Further considering that the discretization of the gradient integral over the image is represented as a gradient accumulation along a certain direction, a mirror cumulative gradient dissimilarity measure can also be calculated from the cumulative gradients in the mirror direction. And finally, calculating the difference measurement of the omnidirectional mirror image gradient of the central pixel point according to the obtained mirror image cumulative gradient dissimilarity measurement based on that the punctiform hot spot pixels show mirror image gradient dissimilarity in all directions.
In particular, a specific implementation of the mirror cumulative gradient dissimilarity metric can be seen in the following formula:
wherein (A), (B), (C), (D), (C), (B), (C)x, y) Coordinates representing a central pixel point, R representing a local radiation structureIn the embodiment, R is 4.
On the basis, the specific implementation of the omni-directional mirror gradient difference metric can be expressed as the following formula:
where the ReLU function represents a linear rectification function, determined by ReLU (x) = max (x, 0). By applying the above scheme, the result of calculating and visualizing the omnidirectional mirror image gradient difference metric for the infrared image shown in fig. 2 can be seen in fig. 5.
Step 105: and determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference measurement.
On the basis of step 104, this step is intended to determine, by the executing entity, whether a punctiform hotspot is contained in the infrared image to be detected, based on the omnidirectional mirror image gradient difference metric. Specifically, the constructed omnidirectional mirror image gradient difference metric may determine a critical value or a threshold value when detecting the punctate hot spots, and determine that the punctate hot spots belong to characteristics that the punctate hot spots should have when the punctate hot spots are located on a certain side of the critical value or are greater than or less than the threshold value.
Further, if the original infrared image is pre-processed in advance according to the above description, this step adapts to: and determining whether the preprocessed infrared image contains the punctiform hot spots according to the omnidirectional mirror image gradient difference measurement.
Compared with the prior art, the photovoltaic panel dotted hot spot detection method provided by the application can be used for providing the mirror image gradient non-similarity measurement by fully representing the local gradient characteristics of the dotted hot spot pixel points, and can be used for effectively enhancing the signal intensity of the dotted hot spots in the image; by fully considering the gradient change characteristics of the punctiform hot spots in different directions and different spatial scales, a local radiation structure is provided, so that the signal enhancement capability of an algorithm on the punctiform hot spot pixel points is ensured while the change of the punctiform hot spot scales is adapted; by fully considering the difference of the dot hot spot pixel point and the background edge pixel point on the image gradient, the omnidirectional image gradient difference measurement is provided, the interference of the background edge pixel point on the detection result can be effectively inhibited, and the detection precision is improved.
Example two
In order to deepen understanding of how to preprocess the original infrared image, the embodiment further provides an image preprocessing method by sequentially performing median filtering and gray level normalization processing through fig. 6, and the remaining steps in the overall scheme still adopt the first embodiment, which includes the following steps:
step 601: determining a target filtering scale of the median filter according to the size of salt and pepper noise and the size of the photovoltaic panel appearing in the infrared image to be detected;
step 602: performing median filtering on the infrared image to be detected by using a median filter with the scale as a target filtering scale to obtain a filtered image;
steps 601-602 are intended to determine a suitable median filtering metric from the executing entity to avoid negative impact on the final result due to median filtering with an unsuitable metric.
Step 603: determining the gray value of each pixel point forming the filtered image, and determining the minimum gray value and the maximum gray value according to the gray value of each pixel point;
step 604: respectively calculating a first gray difference between the actual gray value and the minimum gray value of each pixel point;
step 605: calculating a second gray difference between the maximum gray value and the minimum gray value;
step 606: and taking the quotient of the first gray value and the second gray value difference corresponding to each pixel point as a new gray value of each pixel point.
On the basis of step 605, this step is intended to be performed by the above-described execution subject.
Step 607: and generating a preprocessed image according to the new gray value of each pixel point.
Considering that various interferences suffered in the infrared imaging process can be mainly divided into sensor noise interference and external environment noise interference, the sensor noise interference refers to thermal noise interference spontaneously generated in the imaging process of an infrared camera, and the external environment noise interference is noise interference caused by external factors such as weather, illumination, imaging angle, ground environment and the like in the thermal imaging process.
The two forms of noise interference are mainly represented by salt and pepper noise and uneven image gray distribution in the infrared image. Therefore, the present embodiment sequentially performs median filtering and normalization processing on the input infrared image through steps 601 to 607 to respectively suppress noise interference formed by the two imaging processes.
In a specific implementation, a 3 × 3 median filter may be used to perform filtering preprocessing on the input image, and the preprocessed image may be normalized according to the following formula:
wherein the content of the first and second substances,
representing the minimum and maximum gray scale values of the image pixel, respectively.
The formula exists only as an exemplary normalization mode, and does not mean that there is no other normalization mode capable of achieving a similar gray-scale distribution uniformity effect, such as adjusting the numerator of the above formula to be
And the like.
It should be noted that, this embodiment is a specific implementation given under the idea of performing the median filtering first and then performing the normalization processing, but it is not meant that the specific implementation of the median filtering provided in steps 601 to 602 is necessarily connected to the specific implementation of the normalization provided in steps 603 to 607, and other specific implementations that can achieve the same or similar purpose may be replaced.
EXAMPLE III
In order to better understand how step 105 determines the punctate hot spots based on the constructed omnidirectional mirror gradient difference metric, this embodiment further provides a specific implementation scheme through fig. 7, and the remaining steps in the overall scheme can still be referred to as embodiment one or embodiment two, including the following steps:
step 701: determining the confidence coefficient of each pixel point of the infrared image to be detected, which belongs to the pixel points of the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement to obtain the global confidence coefficient;
step 702: determining a segmentation threshold for screening the punctate hot spots based on the global confidence;
step 703: determining pixel points with actual confidence degrees exceeding a segmentation threshold as target pixel points belonging to the punctiform hot spots;
step 704: and determining the contained point hot spots according to the target pixel points.
That is, in this embodiment, the omni-directional mirror image gradient difference measurement is calculated for all the pixel points of the input image, and the measurement value is used as the confidence that the pixel point is the point hot spot pixel point, and the related statistics is calculated according to the confidence of all the pixel points and used as the adaptive segmentation threshold, and finally, the pixel point with the confidence greater than the segmentation threshold is used as the point hot spot pixel point through screening.
A method of determining a segmentation threshold, including and not limited to, includes:
firstly, determining a confidence coefficient mean value of each pixel point according to the global confidence coefficient; then, determining a confidence standard deviation according to the actual confidence and the confidence mean value of each pixel point; and finally, determining a segmentation threshold value for screening the punctate hot spots according to the confidence coefficient mean value and the confidence coefficient standard deviation.
Specifically, the adaptive segmentation threshold can be obtained by the following formula:
wherein the content of the first and second substances,
which represents the calculated segmentation threshold value(s),
a global mean value representing the confidence level S,
the standard deviation of the confidence S is represented,
representing empirical weight parameters.
Finally, traversing confidence coefficients of all pixel points in the infrared image, and screening out confidence coefficients larger than the confidence coefficients
The pixel point of (2) is used as the detection result of the point hot spot.
Because the situation is complicated and cannot be illustrated by a list, a person skilled in the art can realize that many examples exist according to the basic method principle provided by the application and the practical situation, and the protection scope of the application should be protected without enough inventive work.
Referring to fig. 8, fig. 8 is a block diagram of a structure of a dotted hot spot detection device 800 of a photovoltaic panel according to an embodiment of the present disclosure, where the present embodiment exists as an embodiment of a device corresponding to the foregoing method embodiment, and the dotted hot spot detection device 800 of the photovoltaic panel may include:
the infrared image acquisition unit 801 is configured to acquire an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected; a mirror image gradient non-similarity measurement constructing unit 802 configured to construct mirror image gradient non-similarity measurements in different directions and different orders according to mirror image gradient characteristics exhibited by edges of the photovoltaic panel on pixel points of the infrared image; a local radiation structure constructing unit 803, configured to construct a local radiation structure with a set size scale from inside to outside with a pixel point of the spot hot spot as a center based on a size characteristic of the spot hot spot and an omnidirectional mirror image gradient difference characteristic on the pixel point; an omnidirectional mirror gradient difference metric construction unit 804 configured to construct an omnidirectional mirror gradient difference metric based on the mirror gradient dissimilarity metric and the local radiation structure; a punctiform hot spot detection unit 805 configured to determine whether a punctiform hot spot is included in the infrared image to be detected according to the omnidirectional mirror image gradient difference metric.
Further, the photovoltaic panel dotted hot spot detection apparatus 800 may further include:
the median filtering unit is configured to perform median filtering on the infrared image to be detected by using a median filter to obtain a filtered image;
the normalization unit is configured to perform normalization processing on the filtered image to obtain a preprocessed image;
correspondingly, the punctiform hot spot detection unit 805 may be further configured to:
and determining whether the preprocessed infrared image contains the punctiform hot spots according to the omnidirectional mirror image gradient difference measurement.
Further, the photovoltaic panel dotted hot spot detection apparatus 800 may further include:
a filtering scale determining unit configured to determine a target filtering scale of the median filter according to the size of the salt and pepper noise and the photovoltaic panel appearing in the infrared image to be detected;
correspondingly, the median filtering unit may be further configured to:
and performing median filtering on the infrared image to be detected by using a median filter with the scale as a target filtering scale to obtain a filtered image.
Wherein the normalization unit may be further configured to:
determining the gray value of each pixel point forming the filtered image, and determining the minimum gray value and the maximum gray value according to the gray value of each pixel point;
respectively calculating a first gray difference between the actual gray value and the minimum gray value of each pixel point;
calculating a second gray difference between the maximum gray value and the minimum gray value;
respectively taking the quotient of the first gray value and the second gray value difference corresponding to each pixel point as a new gray value of each pixel point;
and generating a preprocessed image according to the new gray value of each pixel point.
The dotted hot spot detection unit 805 may include:
the global confidence determining subunit is configured to determine the confidence of pixel points of the infrared image to be detected, which belong to the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement, so as to obtain a global confidence;
a segmentation threshold determination subunit configured to determine a segmentation threshold for screening the punctate hot spots based on the global confidence;
a segmentation subunit configured to determine pixel points having an actual confidence exceeding a segmentation threshold as target pixel points belonging to the punctate hot spots;
and the dot-shaped hot spot determining subunit is configured to determine the contained dot-shaped hot spots according to the target pixel points.
Wherein the segmentation threshold determination subunit may be further configured to:
determining the confidence coefficient mean value of each pixel point according to the global confidence coefficient;
determining a confidence standard deviation according to the actual confidence and the confidence mean value of each pixel point;
and determining a segmentation threshold value for screening the punctate hot spots according to the confidence coefficient mean value and the confidence coefficient standard deviation.
Optionally, the different directions of the mirror gradient dissimilarity measure include: horizontal direction, vertical direction, one-three quadrant direction and two-four quadrant direction.
This embodiment exists as an apparatus embodiment corresponding to the method embodiment described above.
Compared with the prior art, the photovoltaic panel dotted hot spot detection device provided by the embodiment can provide mirror image gradient non-similarity measurement by fully representing the local gradient characteristics of dotted hot spot pixel points, and can effectively enhance the signal intensity of dotted hot spots in an image; by fully considering the gradient change characteristics of the punctiform hot spots in different directions and different spatial scales, a local radiation structure is provided, so that the signal enhancement capability of an algorithm on the punctiform hot spot pixel points is ensured while the change of the punctiform hot spot scales is adapted; by fully considering the difference of the dot hot spot pixel point and the background edge pixel point on the image gradient, the omnidirectional image gradient difference measurement is provided, the interference of the background edge pixel point on the detection result can be effectively inhibited, and the detection precision is improved.
Based on the foregoing embodiments, the present application further provides an electronic device, which may include a memory and a processor, where the memory stores a computer program, and the processor, when calling the computer program in the memory, may implement the steps provided by the foregoing embodiments. Of course, the electronic device may also include various necessary network interfaces, power supplies, other components, and the like.
The present application also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by an execution terminal or processor, can implement the steps provided by the above-mentioned embodiments. The storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It will be apparent to those skilled in the art that various changes and modifications can be made in the present invention without departing from the principles of the invention, and these changes and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (9)
1. A method for detecting punctiform hot spots of a photovoltaic panel is characterized by comprising the following steps:
acquiring an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected;
constructing mirror image gradient non-similarity measurement in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image;
based on the size characteristics of the punctiform hot spots and the omnidirectional mirror image gradient difference characteristics on the pixel points, constructing a local radiation structure with a set size scale from inside to outside by taking the pixel points of the punctiform hot spots as centers;
constructing an omnidirectional mirror image gradient difference metric according to the mirror image gradient non-similarity metric and the local radiation structure;
determining whether the infrared image to be detected contains punctate hot spots according to the omnidirectional mirror image gradient difference measurement, wherein the determining comprises the following steps: determining the confidence coefficient of each pixel point of the infrared image to be detected, which belongs to the pixel points of the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement, so as to obtain the global confidence coefficient; determining a segmentation threshold for screening punctate hotspots based on the global confidence; determining the pixel points with actual confidence degrees exceeding the segmentation threshold as target pixel points belonging to the punctiform hot spots; and determining the contained point hot spots according to the target pixel points.
2. The method for detecting the punctiform hot spots of the photovoltaic panel as recited in claim 1, further comprising:
carrying out median filtering on the infrared image to be detected by using a median filter to obtain a filtered image;
carrying out normalization processing on the filtered image to obtain a preprocessed image;
correspondingly, determining whether the infrared image to be detected contains the punctate hot spots according to the omnidirectional mirror image gradient difference measurement, and the method comprises the following steps:
and determining whether the preprocessed infrared image contains punctiform hot spots according to the omnidirectional mirror image gradient difference measurement.
3. The method for detecting the punctiform hot spots of the photovoltaic panel as recited in claim 2, further comprising:
determining a target filtering scale of the median filter according to the size of salt and pepper noise and the size of the photovoltaic panel appearing in the infrared image to be detected;
correspondingly, the performing median filtering on the infrared image to be detected by using a median filter to obtain a filtered image includes:
and performing median filtering on the infrared image to be detected by using a median filter with the scale as the target filtering scale to obtain a filtered image.
4. The method for detecting the punctiform hot spots of the photovoltaic panel as recited in claim 2, wherein the step of normalizing the filtered image to obtain a preprocessed image comprises:
determining the gray value of each pixel point forming the filtered image, and determining the minimum gray value and the maximum gray value according to the gray value of each pixel point;
respectively calculating a first gray difference between the actual gray value of each pixel point and the minimum gray value;
calculating a second gray difference between the maximum gray value and the minimum gray value;
taking the quotient of the first gray value corresponding to each pixel point and the second gray value difference as a new gray value of each pixel point;
and generating the preprocessed image according to the new gray value of each pixel point.
5. The method for detecting the punctiform hot spots of the photovoltaic panel according to claim 1, wherein the determining the segmentation threshold for screening the punctiform hot spots based on the global confidence comprises:
determining the confidence coefficient mean value of each pixel point according to the global confidence coefficient;
determining a confidence standard deviation according to the actual confidence of each pixel point and the confidence mean value;
and determining a segmentation threshold value for screening the punctate hot spots according to the confidence coefficient mean value and the confidence coefficient standard deviation.
6. The method for detecting the punctiform hot spots of the photovoltaic panel according to any one of claims 1 to 5, wherein the different directions of the mirror gradient dissimilarity measure comprise: horizontal direction, vertical direction, one-three quadrant direction and two-four quadrant direction.
7. A photovoltaic panel punctiform hot spot detection device is characterized by comprising:
an infrared image acquisition unit to be detected configured to: acquiring an infrared image to be detected, which is obtained by shooting a photovoltaic panel to be detected;
a mirror gradient dissimilarity metric construction unit configured to: constructing mirror image gradient non-similarity measurement in different directions and different orders according to mirror image gradient characteristics of the edge of the photovoltaic panel on pixel points of the infrared image;
a local radiating structure constructing unit configured to: based on the size characteristics of the punctiform hot spots and the omnidirectional mirror image gradient difference characteristics on the pixel points, constructing a local radiation structure with a set size scale from inside to outside by taking the pixel points of the punctiform hot spots as centers;
an omnidirectional image gradient difference metric construction unit configured to: constructing an omnidirectional mirror image gradient difference metric according to the mirror image gradient non-similarity metric and the local radiation structure;
a punctiform hot spot detection unit configured to: determining whether the infrared image to be detected contains punctiform hot spots or not according to the omnidirectional mirror image gradient difference measurement; the punctiform hot spot detection unit is specifically configured to: determining the confidence coefficient of each pixel point of the infrared image to be detected, which belongs to the pixel points of the punctiform hot spots, according to the omnidirectional mirror image gradient difference measurement, so as to obtain the global confidence coefficient; determining a segmentation threshold for screening punctate hotspots based on the global confidence; determining the pixel points with actual confidence degrees exceeding the segmentation threshold as target pixel points belonging to the punctiform hot spots; and determining the contained point hot spots according to the target pixel points.
8. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the method for detecting punctiform hot spots of a photovoltaic panel according to any one of claims 1 to 6 when executing a computer program stored on the memory.
9. A readable storage medium, characterized in that the readable storage medium has stored thereon a computer program, which, when being executed by a processor, can implement the steps of the method for detecting punctiform hot spots of a photovoltaic panel according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210012032.8A CN114022482B (en) | 2022-01-07 | 2022-01-07 | Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210012032.8A CN114022482B (en) | 2022-01-07 | 2022-01-07 | Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114022482A CN114022482A (en) | 2022-02-08 |
| CN114022482B true CN114022482B (en) | 2022-04-12 |
Family
ID=80069649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210012032.8A Active CN114022482B (en) | 2022-01-07 | 2022-01-07 | Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114022482B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456371B (en) * | 2023-12-26 | 2024-04-12 | 浙江正泰智维能源服务有限公司 | Group string hot spot detection method, device, equipment and medium |
| CN117934474B (en) * | 2024-03-22 | 2024-06-11 | 自贡市第一人民医院 | Gastrointestinal microscopy image enhancement processing method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108923749A (en) * | 2018-06-11 | 2018-11-30 | 东北电力大学 | Photovoltaic module hot spot based on infrared video detects localization method |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107314819B (en) * | 2017-07-03 | 2019-04-23 | 南京绿谷信息科技有限公司 | A kind of detection of photovoltaic plant hot spot and localization method based on infrared image |
| CN109215042B (en) * | 2018-09-28 | 2021-09-03 | 吉林电力股份有限公司科技开发分公司 | Photovoltaic cell panel hot spot effect detection system based on computer vision and calculation method thereof |
| KR20210036044A (en) * | 2019-09-25 | 2021-04-02 | 현대자동차주식회사 | Fault diagnosis method for bypass-diode of solar cell |
| CN111860404A (en) * | 2020-07-28 | 2020-10-30 | 华润智慧能源有限公司 | Photovoltaic panel hot spot positioning method and system |
| CN112200764B (en) * | 2020-09-02 | 2022-05-03 | 重庆邮电大学 | Photovoltaic power station hot spot detection and positioning method based on thermal infrared image |
| CN112164038A (en) * | 2020-09-16 | 2021-01-01 | 上海电力大学 | Photovoltaic hot spot detection method based on deep convolutional neural network |
| CN112184711B (en) * | 2020-11-05 | 2024-04-02 | 杭州青枭科技有限公司 | Photovoltaic module defect detection and positioning method and system |
| CN113076816B (en) * | 2021-03-17 | 2023-06-02 | 上海电力大学 | Solar photovoltaic module hot spot identification method based on infrared and visible light images |
| CN113538391B (en) * | 2021-07-25 | 2022-08-02 | 吉林大学 | Photovoltaic defect detection method based on Yolov4 and thermal infrared image |
-
2022
- 2022-01-07 CN CN202210012032.8A patent/CN114022482B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108923749A (en) * | 2018-06-11 | 2018-11-30 | 东北电力大学 | Photovoltaic module hot spot based on infrared video detects localization method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114022482A (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114022482B (en) | Photovoltaic panel dotted hot spot detection method, device, equipment and readable storage medium | |
| CN107045629B (en) | Multi-lane line detection method | |
| CN107369159B (en) | Threshold segmentation method based on multi-factor two-dimensional gray level histogram | |
| US7672507B2 (en) | Image processing methods and systems | |
| CN102855622B (en) | A kind of infrared remote sensing image sea ship detection method based on significance analysis | |
| JP6305171B2 (en) | How to detect objects in a scene | |
| CN112669318B (en) | Surface defect detection method, device, equipment and storage medium | |
| CN107784669A (en) | A kind of method that hot spot extraction and its barycenter determine | |
| US20060029257A1 (en) | Apparatus for determining a surface condition of an object | |
| Rau et al. | Bridge crack detection using multi-rotary UAV and object-base image analysis | |
| CN110910350B (en) | Nut loosening detection method for wind power tower cylinder | |
| JP2016166853A (en) | Location estimation device and location estimation method | |
| CN114018160A (en) | Lug imaging device and lug size detection method | |
| CN110660065A (en) | Infrared fault detection and identification algorithm | |
| CA3136674A1 (en) | Methods and systems for crack detection using a fully convolutional network | |
| CN115457001A (en) | Photovoltaic panel foreign matter detection method, system, device and medium based on VGG network | |
| CN115345821A (en) | Steel coil binding belt loosening abnormity detection and quantification method based on active visual imaging | |
| CN115018785A (en) | Hoisting steel wire rope tension detection method based on visual vibration frequency identification | |
| CN107767366B (en) | A kind of transmission line of electricity approximating method and device | |
| CN108898080B (en) | Ridge line neighborhood evaluation model-based crack connection method | |
| CN106960437B (en) | Liquid level detection method for oil-immersed power transformer of power inspection robot | |
| Zhu et al. | A filtering strategy for interest point detecting to improve repeatability and information content | |
| Ge et al. | An optimized design of the pointer meter image enhancement and automatic reading system in low illumination environment | |
| Dargazany et al. | Stereo-based terrain traversability estimation using surface normals | |
| CN105427295A (en) | Welding seam based image identification method and image identification system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |