CN115018784B - Method, device, equipment and medium for detecting wire strand scattering defect - Google Patents
Method, device, equipment and medium for detecting wire strand scattering defect Download PDFInfo
- Publication number
- CN115018784B CN115018784B CN202210615506.8A CN202210615506A CN115018784B CN 115018784 B CN115018784 B CN 115018784B CN 202210615506 A CN202210615506 A CN 202210615506A CN 115018784 B CN115018784 B CN 115018784B
- Authority
- CN
- China
- Prior art keywords
- infrared
- image
- detection image
- wire
- fusion
- 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
- 238000000034 method Methods 0.000 title claims abstract description 71
- 230000007547 defect Effects 0.000 title claims abstract description 54
- 238000001514 detection method Methods 0.000 claims abstract description 155
- 230000004927 fusion Effects 0.000 claims abstract description 63
- 238000003709 image segmentation Methods 0.000 claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims description 43
- 238000012549 training Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 5
- 239000004020 conductor Substances 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 abstract description 20
- 238000004891 communication Methods 0.000 description 8
- 238000005286 illumination Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000004438 eyesight Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004313 glare Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method, a device, equipment and a medium for detecting a wire strand scattering defect. The method comprises the following steps: acquiring an infrared detection image and a visible light detection image in a power line scene; performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, wherein pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics; inputting the fusion detection image into a pre-trained image segmentation model, and obtaining a wire mask image corresponding to the fusion detection image; and dividing an infrared lead region from the infrared detection image according to the lead mask diagram, and detecting whether a lead stranding defect exists in the electric power line detection region according to the infrared lead region. By adopting the technical scheme, the accuracy of detecting the scattered strand defects of the lead can be improved.
Description
Technical Field
The invention relates to the technical field of electric power detection, in particular to a method, a device, equipment and a medium for detecting a wire strand scattering defect.
Background
In an electric power system, safe delivery of electric energy is critical to stable operation of the electric power system. The wires are responsible for conveying electric energy in a power grid, and the defect of wire stranding easily causes the sharp increase of power of a power line, so that the phenomena of line short circuit, wire strand breakage and the like are generated. The defect of the scattered strand of the wire is accurately detected and solved at the early stage, and the wire can be effectively prevented from being further damaged, so that the safety and stability of the transmission of the power line are ensured.
At present, the wire stranding defect detection method can be divided into two types, wherein one type adopts a computer vision algorithm to process a visible light image, and the stranding characteristic is utilized to directly detect the wire stranding region; and the other type of the method utilizes the characteristic of temperature rise of the resistance of the wire stranding region to judge the temperature threshold value of the thermodynamic diagram mapped by the infrared image and predict the wire stranding condition.
However, the method for detecting the scattered strand area of the wire in the visible light image by using the computer vision algorithm is difficult to detect the situation of slight scattered strand of the wire, and the recall rate of the algorithm is poor; the method for predicting the wire stranding condition by using the thermodynamic diagram mapped by the infrared image is easy to generate a large number of misjudgments when judging the wire stranding defect due to the complex background, low contrast and large noise of the infrared image.
Disclosure of Invention
The invention provides a method, a device, equipment and a medium for detecting a wire strand scattering defect, which are used for improving the accuracy of wire strand scattering defect detection.
According to an aspect of the present invention, there is provided a method for detecting a wire strand break, the method comprising:
acquiring an infrared detection image and a visible light detection image in a power line scene;
performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, wherein pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics;
inputting the fusion detection image into a pre-trained image segmentation model, and obtaining a wire mask image corresponding to the fusion detection image;
and dividing an infrared lead region from the infrared detection image according to the lead mask diagram, and detecting whether a lead stranding defect exists in the electric power line detection region according to the infrared lead region.
According to another aspect of the present invention, there is provided a device for detecting a wire strand break, comprising:
the image acquisition module is used for acquiring an infrared detection image and a visible light detection image in the power line scene;
the image fusion module is used for carrying out image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, and pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics;
the mask image acquisition module is used for inputting the fusion detection image into a pre-trained image segmentation model to acquire a wire mask image corresponding to the fusion detection image;
and the scattered strand defect detection module is used for dividing an infrared wire area from the infrared detection image according to the wire mask image and detecting whether the wire scattered strand defect exists in the electric power line detection area according to the infrared wire area.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method for detecting a conductor strand break according to any of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute the method for detecting a conductor strand break according to any of the embodiments of the present invention.
According to the technical scheme, the infrared detection image and the visible light detection image in the power line scene are obtained, the infrared detection image and the visible light detection image of the same frame are fused to obtain the four-channel fusion detection image, the image segmentation model is used for obtaining the wire mask image from the fusion detection image, the infrared wire region is segmented from the infrared detection image according to the wire mask image, the abnormal temperature threshold value is calculated by using the inter-class variance method, and all pixel points in the segmented infrared wire region which are larger than the abnormal temperature threshold value are combined to obtain the abnormal temperature region.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for detecting a wire strand break according to a first embodiment of the present invention;
fig. 2a is a flowchart of a method for detecting a wire stranding defect according to a second embodiment of the present invention;
fig. 2b is a block diagram of a UNet model according to the second embodiment of the present invention;
FIG. 2c is a diagram of a residual block diagram according to a second embodiment of the present invention;
FIG. 2d is a block diagram of a decoding module according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a device for detecting a wire strand break defect according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device for implementing a method for detecting a wire stranding defect according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It is noted that the terms "comprises" and "comprising," and any variations thereof, in the description and claims of the present invention and in the foregoing figures, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a method for detecting a wire stranding defect according to an embodiment of the present invention, where the method may be performed by a device for detecting a wire stranding defect, and the device for detecting a wire stranding defect may be implemented in hardware and/or software, and the device for detecting a wire stranding defect may be configured in a device with an image processing function, where the device is adapted to use an infrared image and a visible light image in a power line scene captured by an unmanned aerial vehicle to determine whether a wire area in the image has a wire stranding defect. As shown in fig. 1, the method includes:
s110, acquiring an infrared detection image and a visible light detection image in a power line scene.
In a specific embodiment, the infrared detection image and the visible light detection image may be obtained by shooting through an unmanned aerial vehicle, and the unmanned aerial vehicle is matched with an infrared sensor and a visible light sensor. Alternatively, in a power line scenario, the drone flies about 10 meters from the power line while capturing infrared and visible light images at a low frame rate.
The infrared image detection image is obtained through an infrared sensor carried on the unmanned aerial vehicle and is a single-channel image represented by gray values. The infrared sensor maps the received heat of the outward radiation of the scene (including the dynamic object, the static object and the background) into gray values, and converts the gray values into an infrared detection image, wherein the larger the heat of the outward radiation of a certain part of the scene, namely the higher the temperature of the certain part of the scene, the higher the gray value of the part reflected in the infrared detection image, and the brighter the image. The gray scale value ranges typically from 0 to 255, being an integer value, 0 representing black, 255 representing white, the intermediate value being some different level of gray.
The visible light detection image is obtained through a visible light sensor mounted on the unmanned aerial vehicle, and visible light is a part which can be perceived by human eyes in electromagnetic spectrum. The visible light detection image is an RGB mode image, the RGB mode image is composed of three color channels, namely a Red color channel (Red), a Green color channel (Green) and a Blue color channel (Blue), and the RGB value can represent the brightness of each color, and the range of the RGB value is generally 0 to 255, which is an integer value.
The frame rate refers to the number of frames of images generated per second by the infrared sensor and the visible light sensor.
S120, performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, wherein pixel points in the fusion detection image comprise visible light color features and infrared gray features.
In this embodiment, the feature that the wires in the visible light detection image are clearly distinguished from the background is considered under the condition that the light source is sufficient, however, the detection performance of the visible light sensor is limited under the conditions of dim light, glare, dense fog and the like, so that a clear visible light detection image cannot be obtained, the accuracy is low when the wire area and the non-wire area of the image are segmented later, and the infrared sensor images according to the temperature of the object, so that even under the condition of poor illumination, the wire target with higher temperature can be captured. Therefore, in this embodiment, the manner of performing image fusion on the visible light detection image and the infrared detection image is adopted, so that the fused detection image contains information in the infrared detection image and the visible light detection image at the same time, and is not affected by illumination conditions.
Wherein the pixels are the smallest units that make up the image, each smallest unit can be considered as a pixel, each pixel having a well-defined position in the image and assigned color values, which in this embodiment are the gray value and the RGB value of the image.
The image fusion specifically means that the features of each channel of three channels of visible light RGB images and single channel infrared gray scale images of the same frame are fused to obtain a four-channel image containing visible light color features and infrared gray scale features.
The visible light color features are visible light RGB value features, and the infrared gray scale features are infrared gray scale value features.
The advantages of this arrangement are that: the defect that the visible light image obtained under the condition of poor illumination is not clear is overcome, the information contained in the fused image is more comprehensive, and the lead area and the non-lead area can be easily segmented in the fused image.
S130, inputting the fusion detection image into a pre-trained image segmentation model, and acquiring a wire mask image corresponding to the fusion detection image.
The image segmentation model may be a UNet segmentation model. The UNet segmentation model is a model based on a deep convolutional neural network for training, has higher recognition accuracy and generalization capability, has stronger representativeness of extracted features, and overcomes the defects that the traditional image processing method needs specific background conditions or illumination conditions, has poor real-time monitoring, has weak generalization capability and the like.
Specifically, the wire mask pattern is mainly used for dividing a wire area and a non-wire area, and is a binary image composed of 0 and 1, wherein the wire area is composed of 1, and is displayed as a white area in the mask pattern; the non-conductive regions are each composed of 0, and are shown as black regions in the mask pattern.
Based on the embodiment, the invention creatively provides that the fusion detection image obtained by fusing the visible light detection image and the infrared detection image is input into the image segmentation model, and the wire mask image corresponding to the fusion detection image is obtained, so that the UNet model has higher accuracy when the image with poor illumination condition is segmented.
And S140, dividing an infrared wire region from the infrared detection image according to the wire mask diagram, and detecting whether the wire stranding defect exists in the electric power circuit detection region according to the infrared wire region.
The step of dividing the infrared lead area from the infrared detection image according to the lead mask image specifically refers to finding out corresponding pixel points in the corresponding infrared detection image before image fusion according to the pixel point positions contained in the divided lead area in the lead mask image, wherein the area formed by the pixel points is the infrared lead area.
According to the infrared wire area, detecting whether the wire stranding defect exists in the electric power circuit detection area or not specifically includes: and identifying an abnormal temperature region in the infrared wire region, and determining that the wires in the abnormal temperature region have wire stranding defects.
Further, identifying an abnormal temperature region in the infrared wire region may specifically include:
obtaining an abnormal temperature threshold matched with the infrared wire region by adopting an inter-class variance method;
and combining all pixel points with gray values larger than the abnormal temperature threshold in the infrared lead area to obtain the abnormal temperature area.
Further, the method for obtaining the abnormal temperature threshold matched with the infrared wire region by adopting the inter-class variance method specifically comprises the following steps:
acquiring a gray level range [0, L-1] matched with the infrared wire region, wherein L is an integer greater than 1;
acquiring the number n of pixel points belonging to each gray level i in the infrared wire area i The total pixel number of the wire area is N;
according to the formula: p is p i ＝n i N, i=0, 1,2 …, L-1, calculating the probability p corresponding to each gray level i i ；
Sequentially acquiring a current processing gray level T in the gray level range, and acquiring gray values in the infrared wire area to be in [0, T ]]First pixel point set C of (2) 0 And the gray value is located at [ T+1, L-1]]Second pixel point set C of (2) 1 ；
According to the formula:
calculated and C 0 Corresponding mean value u 0 And (b)And C 1 Corresponding mean value u 1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein,ω 1 ＝1-ω 0 ；
according to the formula: u (u) T ＝ω 0 u 0 +ω 1 u 1 Calculating to obtain an image mean value u of the infrared lead area T ；
According to the formula:calculating to obtain the inter-class variance which is matched with the current processing gray level T>
Returning to execute the operation of sequentially obtaining a current processing gray level T in the gray level range until the inter-class variance corresponding to each gray level is calculated;
and determining the target gray level corresponding to the maximum inter-class variance as the abnormal temperature threshold.
On the basis of the above embodiment, it can be simply understood that the inter-class variance matching with the current processing gray level T is calculated by using known parameters such as parameter points and gray values in the image and the current processing gray level T acquired in the gray level rangeThe variance between all classes calculated +.>In (c) finding the value of the greatest +.>Will be in charge of this>The corresponding gray level T is determined to be abnormalA temperature threshold.
According to the technical scheme, the infrared detection image and the visible light detection image in the power line scene are obtained, the infrared detection image and the visible light detection image of the same frame are fused to obtain the four-channel fusion detection image, the image segmentation model is used for obtaining the wire mask image from the fusion detection image, the infrared wire region is segmented from the infrared detection image according to the wire mask image, the abnormal temperature threshold value is calculated by using the inter-class variance method, and all pixel points in the segmented infrared wire region which are larger than the abnormal temperature threshold value are combined to obtain the abnormal temperature region.
Example two
Fig. 2a is a flowchart of a method for detecting a wire stranding defect according to a second embodiment of the present invention, where the method further embodies a wire stranding defect detection process based on the above embodiment. As shown in fig. 2a, the method comprises:
s210, acquiring an original UNet model to be trained.
A block diagram of the UNet model is shown in fig. 2 b. As shown in fig. 2b, the UNet model is composed of an encoding network and a decoding network, the left half of fig. 2b is the encoding network from the input to the lowest Res Block, the right half of fig. 2b is the decoding network from the lowest decoding Block to the output, and the whole network performs 5 downsampling and 5 upsampling. The size of the feature map is halved in each downsampling, the number of channels is doubled, and the feature map is smaller and denser from flat and more stereoscopic; each up-sampling doubles the size of the feature map, halving the number of channels, and the feature map eventually reverts to the size at the time of input.
Wherein the coding network of the original UNet model comprises a ResNet50 network with 6 cross-layer connections; the decoding network of the original UNet model comprises 5 decoding modules, and 4 jump connections are arranged between the encoding network and the decoding network.
The coding network of the original UNet model adopts a ResNet50 network with 6 cross-layer connections, and the following benefits are: because the semantics of the wire are simpler, the edge detail information is difficult to capture, the deep semantic information and the shallow feature detail of the image are important, 6 cross-layer connections are added on the basis of ResNet50, a dense network structure is formed through unified dimensionality of maximum pooling operation (Maxpool), the shallow features can be reused by the structure, the depth extension of the features is enhanced, and meanwhile, gradient disappearance during back propagation can be prevented by the cross-layer connections.
The coding network and the decoding network have 4 jump connections, and the advantage of the arrangement is that: and features in the same dimension are spliced by using a jump connection mode, so that multiplexing of shallow features is realized.
The coding network comprises 4 residual modules (Res Block), the residual module structure is shown in fig. 2c, each module comprises two Convolution layers (convolume, conv), an activation function layer (Rectified Linear Unit, reLU) and two batch normalization layers (Batch Normalization, BN), and the Convolution kernel of the Convolution layers has a size of 3*3.
The decoding network comprises five decoding blocks (decoding blocks), and the decoding blocks are structured as shown in fig. 2d, where each Block includes Up-sampling (Up Sample), two convolutional layers Conv2d, and two active function layers ReLU. The up-sampling method adopts a nearest neighbor interpolation method for doubling the size of the feature map, the convolution kernel is 3*3, the step length is 1, and the up-sampling method is used for fusing features.
S220, defining a loss function.
The loss function L dice The calculation formula is as follows:
the loss function adopts a Dice function, wherein W is the image width, H is the image height, P is the model output image, and P ij Is a pixel point in P, T is a training sample image, T ij Is one of TAnd a pixel point.
The advantages of this arrangement are that: in the power line image, the number of the background negative samples is far greater than that of the lead positive samples, and a Dice function is commonly used for calculating the similarity among the samples, and the method focuses on whether the foreground image is correctly classified and does not focus on background pixels.
S230, performing iterative training on the original UNet model by using a pre-labeled training sample set to obtain an UNet segmentation model.
The training sample set is a fusion image of a plurality of electric power scenes and pre-labeled with matched wire mask images, and the fusion image is an image obtained by fusing an infrared image and a visible light image shot by an unmanned aerial vehicle.
The specific iterative training method comprises the following steps: training a UNet model on a physical computer platform Windows, wherein the model training adopts a self-adaptive moment estimation optimizer as a training optimization strategy of a network, and training is performed for 100 rounds. The learning rate is dynamically adjusted from 0.001, and after each round of updating is completed, the learning rate is multiplied by 0.9. And (3) observing the change of the training loss function, stopping training when the loss function value is not reduced for 5 consecutive rounds, and obtaining a model with the best convergence effect, namely a model with the lowest final loss function value.
S240, acquiring an infrared detection image and a visible light detection image in the power line scene.
S250, performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image.
It should be noted that each set of infrared monitoring images and visible light images used for image fusion must be images of the same frame, that is, pixel points at the same position in each set of infrared monitoring images and visible light images correspond to the same position point in real space.
And S260, inputting the fusion detection image into a pre-trained UNet model, and acquiring a wire mask diagram corresponding to the fusion detection image.
S270, dividing an infrared lead area from the infrared detection image according to the lead mask diagram.
Specifically, according to the pixel positions included in the divided lead areas in the lead mask diagram, corresponding pixel points are found out from the corresponding infrared detection image before image fusion, and the areas formed by the pixel points are the infrared lead areas.
S280, acquiring an abnormal temperature threshold matched with the infrared wire region by adopting an inter-class variance method.
S290, comparing the gray value in the infrared wire area with the abnormal temperature threshold value, and combining all pixel points with gray values larger than the abnormal temperature threshold value in the infrared wire area to obtain the abnormal temperature area.
And S2100, judging the abnormal temperature region as a region with wire strand scattering defects.
According to the technical scheme provided by the embodiment of the invention, the fusion detection image is input into the UNet segmentation model to obtain the wire mask image through pre-training the UNet segmentation model, the infrared wire region is segmented from the wire mask image, and the abnormal temperature region in the infrared wire region is judged to be the region with the wire strand scattering defect, so that the problem that the wire and background contrast in the infrared image is low and difficult to distinguish is solved, the false alarm rate of wire strand scattering detection is reduced, and the influence of illumination conditions on the wire strand scattering detection is eliminated.
Example III
Fig. 3 is a schematic structural diagram of a device for detecting a wire strand break defect according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes: an image acquisition module 310, an image fusion module 320, a mask image acquisition module 330, and a stranding defect detection module 340, wherein:
the image acquisition module 310 is configured to acquire an infrared detection image and a visible light detection image in a power line scene.
The image fusion module 320 is configured to perform image fusion on the infrared detection image and the visible light detection image to obtain a fused detection image, where pixel points in the fused detection image include visible light color features and infrared gray features.
And a mask map obtaining module 330, configured to input the fusion detection image into a pre-trained image segmentation model, and obtain a wire mask map corresponding to the fusion detection image.
And the stranding defect detection module 340 is configured to segment an infrared wire region from the infrared detection image according to the wire mask diagram, and detect whether a wire stranding defect exists in the power line detection region according to the infrared wire region.
According to the technical scheme, the infrared detection image and the visible light detection image in the power line scene are obtained, the infrared detection image and the visible light detection image of the same frame are fused to obtain the four-channel fusion detection image, the image segmentation model is used for obtaining the wire mask image from the fusion detection image, the infrared wire region is segmented from the infrared detection image according to the wire mask image, the abnormal temperature threshold value is calculated by using the inter-class variance method, and all pixel points in the segmented infrared wire region which are larger than the abnormal temperature threshold value are combined to obtain the abnormal temperature region.
On the basis of the above embodiments, the image segmentation model is a UNet segmentation model.
Based on the above embodiments, the image acquisition module 310 may specifically be configured to: and acquiring an infrared detection image and a visible light detection image acquired at a low frame rate when the unmanned aerial vehicle performs flight operation at a preset height from the power line scene.
Based on the above embodiments, the stranding defect detection module 340 may be specifically configured to: and identifying an abnormal temperature region in the infrared wire region, and determining that the wires in the abnormal temperature region have wire stranding defects.
On the basis of the above embodiments, the method may further include a UNet model training module, configured to:
acquiring an original UNet model to be trained; the coding network of the original UNet model comprises a ResNet50 network with 6 cross-layer connections; the decoding network of the original UNet model comprises 5 decoding modules, and 4 jump connections are arranged between the encoding network and the decoding network;
performing iterative training on the original UNet model by using a pre-labeled training sample set to obtain the UNet segmentation model;
wherein in training the UNet model, a Dice function is used as a loss function.
On the basis of the above embodiments, the method may further include: an abnormal temperature region detection unit comprising:
an abnormal temperature threshold obtaining subunit, configured to obtain an abnormal temperature threshold that is matched with the infrared wire region by using an inter-class variance method;
and the abnormal temperature region acquisition subunit is used for combining all pixel points with gray values larger than the abnormal temperature threshold in the infrared lead region to obtain the abnormal temperature region.
On the basis of the above embodiments, the abnormal temperature threshold acquisition subunit may be specifically configured to:
acquiring a gray level range [0, L-1] matched with the infrared wire region, wherein L is an integer greater than 1;
acquiring the number n of pixel points belonging to each gray level i in the infrared wire area i The total pixel number of the wire area is N;
according to the formula: p is p i ＝n i N, i=0, 1,2 …, L-1, calculating the probability p corresponding to each gray level i i ；
Sequentially acquiring a current processing gray level T in the gray level range, and acquiring gray values in the infrared wire area to be in [0, T ]]First pixel point set C of (2) 0 And the gray value is located at [ T+1, L-1]]Second pixel point set C of (2) 1 ；
According to the formula:
calculated and C 0 Corresponding mean value u 0 And with C 1 Corresponding mean value u 1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein,ω 1 ＝1-ω0；
according to the formula: u (u) T ＝ω 0 u 0+ ω 1 u 1 Calculating to obtain an image mean value u of the infrared lead area T ；
According to the formula:calculating to obtain the inter-class variance which is matched with the current processing gray level T>
Returning to execute the operation of sequentially obtaining a current processing gray level T in the gray level range until the inter-class variance corresponding to each gray level is calculated;
and determining the target gray level corresponding to the maximum inter-class variance as the abnormal temperature threshold.
The device for detecting the wire strand scattering defects, which is provided by the embodiment of the invention, can execute the method for detecting the wire strand scattering defects, which is provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example 4
Fig. 4 shows a schematic diagram of an electronic device 40 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 4, the electronic device 40 includes at least one processor 41, and a memory communicatively connected to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., in which the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 may also be stored. The processor 41, the ROM 42 and the RAM 43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.
Various components in electronic device 40 are connected to I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, a modem wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as the method of detecting a wire strand break.
Namely:
acquiring an infrared detection image and a visible light detection image in a power line scene;
performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, wherein pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics;
inputting the fusion detection image into a pre-trained image segmentation model, and obtaining a wire mask image corresponding to the fusion detection image;
and dividing an infrared lead region from the infrared detection image according to the lead mask diagram, and detecting whether a lead stranding defect exists in the electric power line detection region according to the infrared lead region.
In some embodiments, the method of detecting a wire strand defect may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM 42 and/or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the above-described method of detecting a wire-strand defect may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the method of detecting wire stranding defects in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (7)
1. The method for detecting the wire strand scattering defect is characterized by comprising the following steps of:
acquiring an infrared detection image and a visible light detection image in a power line scene;
performing image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, wherein pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics;
inputting the fusion detection image into a pre-trained image segmentation model, and obtaining a wire mask image corresponding to the fusion detection image; wherein the image segmentation model is a UNet segmentation model;
dividing an infrared lead region from the infrared detection image according to the lead mask image, and detecting whether a lead stranding defect exists in the electric power line detection region according to the infrared lead region;
wherein before inputting the fusion detection image into a pre-trained image segmentation model, further comprising:
acquiring an original UNet model to be trained; the coding network of the original UNet model comprises a ResNet50 network with 6 cross-layer connections; the decoding network of the original UNet model comprises 5 decoding modules, and 4 jump connections are arranged between the encoding network and the decoding network;
performing iterative training on the original UNet model by using a pre-labeled training sample set to obtain the UNet segmentation model;
in the process of training the UNet model, a Dice function is used as a loss function;
the coding network comprises 4 residual modules, wherein each residual module comprises two convolution layers, an activation function layer and two batch normalization layers, and the convolution kernel of the convolution layers is 3*3; the decoding network comprises five decoding modules, wherein each decoding module comprises up-sampling, two convolution layers and two activation function layers; the up-sampling adopts a nearest neighbor interpolation method;
according to the infrared wire area, detecting whether the wire stranding defect exists in the electric power circuit detection area comprises the following steps:
and identifying an abnormal temperature region in the infrared wire region, and determining that the wires in the abnormal temperature region have wire stranding defects.
2. The method of claim 1, wherein acquiring infrared detection images and visible light detection images in a power line scene comprises:
and acquiring an infrared detection image and a visible light detection image acquired at a low frame rate when the unmanned aerial vehicle performs flight operation at a preset height from the power line scene.
3. The method of claim 1, wherein identifying an abnormal temperature region in the infrared wire region comprises:
obtaining an abnormal temperature threshold matched with the infrared wire region by adopting an inter-class variance method;
and combining all pixel points with gray values larger than the abnormal temperature threshold in the infrared lead area to obtain the abnormal temperature area.
4. The method of claim 3, wherein obtaining an abnormal temperature threshold matching the infrared wire region using an inter-class variance method comprises:
acquiring a gray level range [0, L-1] matched with the infrared wire region, wherein L is an integer greater than 1;
acquiring the number n of pixel points belonging to each gray level i in the infrared wire area i The total pixel number of the infrared wire area is N;
according to the formula: p is p i ＝n i N, i=0, 1,2 …, L-1, calculating the probability p corresponding to each gray level i i ；
Sequentially acquiring a current processing gray level T in the gray level range, and acquiring gray values in the infrared wire area to be in [0, T ]]First pixel point set C of (2) 0 And the gray value is located at [ T+1, L-1]]Second pixel point set C of (2) 1 ；
According to the formula:
calculated and C 0 Corresponding mean value u 0 And with C 1 Corresponding mean value u 1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein,ω 1 ＝1-ω 0 ；
according to the formula: u (u) T ＝ω 0 u 0 +ω 1 u 1 Calculating to obtain an image mean value u of the infrared lead area T ；
According to the formula:calculating to obtain the inter-class variance which is matched with the current processing gray level T>
Returning to execute the operation of sequentially obtaining a current processing gray level T in the gray level range until the inter-class variance corresponding to each gray level is calculated;
and determining the target gray level corresponding to the maximum inter-class variance as the abnormal temperature threshold.
5. The utility model provides a detection device of wire strand defect which characterized in that includes:
the image acquisition module is used for acquiring an infrared detection image and a visible light detection image in the power line scene;
the image fusion module is used for carrying out image fusion on the infrared detection image and the visible light detection image to obtain a fusion detection image, and pixel points in the fusion detection image comprise visible light color characteristics and infrared gray scale characteristics;
the mask image acquisition module is used for inputting the fusion detection image into a pre-trained image segmentation model to acquire a wire mask image corresponding to the fusion detection image; wherein the image segmentation model is a UNet segmentation model;
the scattered strand defect detection module is used for dividing an infrared wire area from the infrared detection image according to the wire mask image and detecting whether a wire scattered strand defect exists in the electric power line detection area according to the infrared wire area;
the system further comprises a UNet model training module, which is used for:
acquiring an original UNet model to be trained; the coding network of the original UNet model comprises a ResNet50 network with 6 cross-layer connections; the decoding network of the original UNet model comprises 5 decoding modules, and 4 jump connections are arranged between the encoding network and the decoding network;
performing iterative training on the original UNet model by using a pre-labeled training sample set to obtain the UNet segmentation model;
in the process of training the UNet model, a Dice function is used as a loss function;
the coding network comprises 4 residual modules, wherein each residual module comprises two convolution layers, an activation function layer and two batch normalization layers, and the convolution kernel of the convolution layers is 3*3; the decoding network comprises five decoding modules, wherein each decoding module comprises up-sampling, two convolution layers and two activation function layers; the up-sampling adopts a nearest neighbor interpolation method;
the loose strand defect detection module is specifically used for: and identifying an abnormal temperature region in the infrared wire region, and determining that the wires in the abnormal temperature region have wire stranding defects.
6. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of detecting a conductor strand break of any one of claims 1-4.
7. A computer readable storage medium storing computer instructions for causing a processor to perform the method of detecting a conductor strand break according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210615506.8A CN115018784B (en) | 2022-05-31 | 2022-05-31 | Method, device, equipment and medium for detecting wire strand scattering defect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210615506.8A CN115018784B (en) | 2022-05-31 | 2022-05-31 | Method, device, equipment and medium for detecting wire strand scattering defect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115018784A CN115018784A (en) | 2022-09-06 |
| CN115018784B true CN115018784B (en) | 2024-03-12 |
Family
ID=83070868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210615506.8A Active CN115018784B (en) | 2022-05-31 | 2022-05-31 | Method, device, equipment and medium for detecting wire strand scattering defect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115018784B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117710366B (en) * | 2024-02-02 | 2024-05-14 | 杭州百子尖科技股份有限公司 | Quality inspection method and device for thermos cup and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018120936A1 (en) * | 2016-12-27 | 2018-07-05 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for fusing infrared image and visible light image |
| CN108492292A (en) * | 2018-03-20 | 2018-09-04 | 西安工程大学 | Conducting wire broken lot detection method based on infrared image processing |
| CN109544497A (en) * | 2018-11-21 | 2019-03-29 | 国网江苏省电力有限公司扬州供电分公司 | Image interfusion method and electronic equipment for transmission line faultlocating |
| CN112465745A (en) * | 2020-10-26 | 2021-03-09 | 脉得智能科技(无锡)有限公司 | Cell counting method based on full convolution regression network |
| CN112819784A (en) * | 2021-02-01 | 2021-05-18 | 广东电网有限责任公司广州供电局 | Method and system for detecting broken strands and scattered strands of wires of distribution line |
| CN113822865A (en) * | 2021-09-18 | 2021-12-21 | 长春理工大学 | Abdomen CT image liver automatic segmentation method based on deep learning |
-
2022
- 2022-05-31 CN CN202210615506.8A patent/CN115018784B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018120936A1 (en) * | 2016-12-27 | 2018-07-05 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for fusing infrared image and visible light image |
| CN108492292A (en) * | 2018-03-20 | 2018-09-04 | 西安工程大学 | Conducting wire broken lot detection method based on infrared image processing |
| CN109544497A (en) * | 2018-11-21 | 2019-03-29 | 国网江苏省电力有限公司扬州供电分公司 | Image interfusion method and electronic equipment for transmission line faultlocating |
| CN112465745A (en) * | 2020-10-26 | 2021-03-09 | 脉得智能科技(无锡)有限公司 | Cell counting method based on full convolution regression network |
| CN112819784A (en) * | 2021-02-01 | 2021-05-18 | 广东电网有限责任公司广州供电局 | Method and system for detecting broken strands and scattered strands of wires of distribution line |
| CN113822865A (en) * | 2021-09-18 | 2021-12-21 | 长春理工大学 | Abdomen CT image liver automatic segmentation method based on deep learning |
Non-Patent Citations (6)
| Title |
|---|
| A Deep Learning Method to Detect Foreign Objects for Inspecting Power Transmission Lines;J. Zhu等;IEEE;第8卷;全文 * |
| Attention-Based Multimodal Image Feature Fusion Module for Transmission Line Detection;H. Choi等;IEEE Transactions on Industrial Informatics;第2-30页 * |
| Real-time power line detection network using visible light and infrared images;Choi H等;IEEE;第1-6页 * |
| Research on antenna for detecting the corona discharge of transmission line;Yunfeng Xia等;The Journal of Engineering;全文 * |
| 吴良斌.《SAR图像处理与目标识别》.北京：航空工业出版社,2013,第152-153页. * |
| 基于无人机的输电线路立体巡检***设计及图像处理研究;王定松;中国优秀硕博论文全文库 信息科技辑;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115018784A (en) | 2022-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112801164A (en) | Training method, device and equipment of target detection model and storage medium | |
| CN117392733B (en) | Acne grading detection method and device, electronic equipment and storage medium | |
| CN112883902A (en) | Video detection method and device, electronic equipment and storage medium | |
| CN115018784B (en) | Method, device, equipment and medium for detecting wire strand scattering defect | |
| CN113569708A (en) | Living body recognition method, living body recognition device, electronic apparatus, and storage medium | |
| CN115471476A (en) | Method, device, equipment and medium for detecting component defects | |
| CN114037087B (en) | Model training method and device, depth prediction method and device, equipment and medium | |
| CN116703925B (en) | Bearing defect detection method and device, electronic equipment and storage medium | |
| CN113705380A (en) | Target detection method and device in foggy days, electronic equipment and storage medium | |
| CN116778386A (en) | Sulfur hexafluoride leakage detection method, device, equipment and storage medium | |
| US10146042B2 (en) | Image processing apparatus, storage medium, and image processing method | |
| CN114092739B (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN114037865B (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN115205939B (en) | Training method and device for human face living body detection model, electronic equipment and storage medium | |
| CN118212668A (en) | Target identification method, device, electronic equipment and medium | |
| CN115908982B (en) | Image processing method, model training method, device, equipment and storage medium | |
| CN116091785A (en) | Identification model generation method and device, electronic equipment and storage medium | |
| CN116977332A (en) | Camera light filling lamp performance test method and device, electronic equipment and storage medium | |
| CN117934897A (en) | Equipment abnormality detection method, device, equipment and storage medium | |
| CN115527257A (en) | Face recognition method and device | |
| CN116912194A (en) | Insulator detection method and device, electronic equipment and storage medium | |
| CN114757845A (en) | Light ray adjusting method and device based on face recognition, electronic equipment and medium | |
| CN116052181A (en) | Image direction determining method and device, electronic equipment and medium | |
| CN117808775A (en) | Defect level determining method and device, electronic equipment and storage medium | |
| CN114842464A (en) | Image direction recognition method, device, equipment, storage medium and program product |
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 |