CN113763358B - Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation - Google Patents
Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation Download PDFInfo
- Publication number
- CN113763358B CN113763358B CN202111049692.5A CN202111049692A CN113763358B CN 113763358 B CN113763358 B CN 113763358B CN 202111049692 A CN202111049692 A CN 202111049692A CN 113763358 B CN113763358 B CN 113763358B
- Authority
- CN
- China
- Prior art keywords
- sampling
- semantic segmentation
- module
- training
- transformer substation
- 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
- 230000011218 segmentation Effects 0.000 title claims abstract description 54
- 239000002184 metal Substances 0.000 title claims abstract description 40
- 230000007797 corrosion Effects 0.000 title claims abstract description 23
- 238000005260 corrosion Methods 0.000 title claims abstract description 23
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 47
- 238000005070 sampling Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims description 16
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 2
- 230000003628 erosive effect Effects 0.000 claims 1
- 230000008961 swelling Effects 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 13
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000001360 synchronised effect Effects 0.000 abstract 1
- 238000005286 illumination Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000003466 welding Methods 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
- G06T7/0008—Industrial image inspection checking presence/absence
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
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)
- Software Systems (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)
- Quality & Reliability (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
Abstract
The invention discloses a method and a system for detecting oil leakage and metal corrosion of a transformer substation based on semantic segmentation, which belong to the technical field of defect detection of transformer substation equipment and comprise the following steps: s1: training a network; s2: segmentation and identification; s3: fusing areas; s4: sampling at intervals; s5: repeating sampling; s6: a target area is determined. The invention adopts the technology of semantic segmentation to detect the defects, improves the defect detection rate, meets the requirement of high detection rate of the transformer substation, can ensure higher accuracy rate while having high detection rate, has more advantages than synchronous improvement of the detection rate and the false detection rate after the threshold value is reduced in the target detection, and is worth being popularized and used.
Description
Technical Field
The invention relates to the technical field of transformer substation equipment defect detection, in particular to a method and a system for detecting transformer substation oil leakage and metal corrosion based on semantic segmentation.
Background
Most of equipment of the transformer substation, such as main transformer, sleeve and the like, operates under high voltage and strong magnetic field environment, and the equipment is often filled with oil liquid for insulation, isolation, heat dissipation and other functions. After long-term use, the equipment can appear oil leakage because of problems such as sealing or welding. Once the oil in the equipment is reduced to a certain extent, the insulation performance of the equipment is greatly reduced, and a large fault hidden trouble can be caused to the normal operation of the equipment, so that the method has great value for accurately detecting the oil leakage condition of the equipment.
Metal tarnish failure occurs more frequently on metal-type equipment exposed outdoors, and oxidation occurs on the equipment surface. Such faults are of somewhat less importance than oil leakage, but also affect the safe operation of the substation.
There are two solutions currently available, one is a traditional image processing-based solution, and the other is a solution by using a target detection technology after deep learning.
Scheme based on image processing: color space transformation such as HSV space is firstly carried out, then a color histogram is counted, threshold separation and color clustering is carried out. This solution is particularly robust and is susceptible to interference from various environmental factors, such as illumination, and reflectors. The adaptability under complex background is particularly poor, the false alarm rate is higher, and the detection rate is lower.
After deep learning is raised, trying to solve the problem of oil leakage and metal corrosion as target detection tasks, firstly marking a target area with a bounding box, and then training targets in the bounding box. Compared with the technical precision based on image processing, the technical precision of the scheme is improved more. However, due to the characteristics of non-rigidity and changeable shape of two targets, namely oil leakage and metal corrosion, the precision is a certain distance away from ideal practicability, and therefore, the method and the system for detecting the oil leakage and the metal corrosion of the transformer substation based on semantic segmentation are provided.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: according to the characteristics of non-rigidity and changeable shape of two targets, namely oil leakage and metal corrosion, the oil leakage and metal corrosion phenomenon is detected by adopting a semantic segmentation technology, so that the detection accuracy is greatly improved, and the method for detecting the oil leakage and the metal corrosion of the transformer substation based on the semantic segmentation is provided.
The invention solves the technical problems through the following technical proposal, and the invention comprises the following steps:
s1: training network
Training a semantic segmentation network by using the marked training data set;
s2: segmentation identification
Using the semantic segmentation network after training to carry out segmentation recognition on the picture to be detected;
s3: region fusion
Aiming at each type of information, adopting basic image processing operation to respectively fuse into oil seepage areas and metal rust areas;
s4: sampling at intervals
The sampling points are subjected to image acquisition again at fixed intervals of time t, and the steps S2 and S3 are repeated;
s5: repeated sampling
Repeating the step S4 until n times of sampling are accumulated for the same sampling point;
s6: determining a target area
Recording according to the sampling time sequence aiming at the same area length-width ratio, removing the detection result with unreasonable area change, and determining the final target area.
Further, the step S1 specifically includes the following steps:
s101: collecting a data picture, manually segmenting and labeling the picture by using a labelme data labeling tool, and segmenting out an oil leakage area and a metal rust area which exist in the picture to form a training data set;
s102: training the semantic segmentation network using the labeled data set.
Further, in the step S102, the semantic segmentation network is a U-Net network.
Further, in the step S3, each type of information is the pixel information of leaked oil and metal rust after being identified by segmentation.
Still further, in the step S3, the basic image processing operation includes a corrosion, expansion, connected domain extraction operation.
Further, in the step S4, the interval time t is not less than 0.5h.
Further, in the step S5, the sampling number n is not less than 3.
Further, in the step S6, the types of detection results with unreasonable area change include two types, the first is that the detected area becomes smaller with time in the multiple detection results, and the second is that the detected area has two kinds of change processes of becoming larger and smaller with time in the multiple detection results.
The invention also provides a transformer substation oil leakage and metal corrosion detection system based on semantic segmentation, which adopts the detection method to detect the oil leakage and metal corrosion phenomena of transformer substation equipment and comprises the following steps:
the training module is used for training the semantic segmentation network by using the marked training data set;
the segmentation module is used for carrying out segmentation recognition on the picture to be detected by using the semantic segmentation network after the training is completed;
the area fusion module is used for respectively fusing the oil seepage areas and the metal rust areas by adopting basic image processing operation aiming at each type of information;
the first sampling module is used for carrying out image acquisition on the sampling points again at fixed time t intervals, and repeating the step S2 and the step S3;
the second sampling module is used for repeating the step S4 until n times of sampling are accumulated for the same sampling point;
the region determining module is used for recording the same region length-width ratio according to the sampling time sequence, removing the detection result with unreasonable region change and determining a final target region;
the control processing module is used for sending control instructions to the modules to complete related actions;
the training module, the segmentation module, the region fusion module, the first sampling module, the second sampling module and the region determination module are electrically connected with the control processing module.
Compared with the prior art, the invention has the following advantages: the defect detection is carried out by adopting the semantic segmentation technology, so that the defect detection rate is improved, the requirement of high detection rate of the transformer substation is met, and the higher accuracy can be ensured while the detection rate is high, and the method is worthy of being popularized and used.
Drawings
FIG. 1 is a schematic flow chart of a method for detecting oil leakage and metal rust failure in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram of a U-Net network according to a first embodiment of the present invention;
fig. 3 is a schematic diagram of a specific flow of defect identification in the second embodiment of the present invention.
Detailed Description
The following describes in detail the examples of the present invention, which are implemented on the premise of the technical solution of the present invention, and detailed embodiments and specific operation procedures are given, but the scope of protection of the present invention is not limited to the following examples.
Example 1
As shown in fig. 1, the substation oil leakage and metal rust fault detection method based on image semantic segmentation comprises the following steps:
step 1: the semantic segmentation network is trained.
The semantic segmentation network adopted by the method is generally a U-Net network;
in this embodiment, the U-Net network is similar to a U-shape in network structure, and as shown in fig. 2, the U-Net network is divided into 2 parts, the first part: extracting features; a second part: and an upsampling section.
The left side of the figure is the overall network structure, wherein the lower right basic structure is utilized, the lower right label part conv3x3 represents the convolution kernel size of 3x3, and Relu represents the activation function; copy and crop represents copy and cut operations of the feature map; max pool 2x2 represents a maxpooling operation of 2x 2; up-conv 2x2 represents a deconvolution operation with a convolution kernel size of 2x 2; conv 1x1 represents the convolution kernel size of 1x 1.
In this embodiment, the training steps of the semantic segmentation network are as follows:
step 101: collecting a data picture, manually segmenting and labeling the picture by using a labelme data labeling tool, and segmenting out an oil leakage area and a metal rust area which exist in the picture to form a training data set;
step 102: training the semantic segmentation network using the labeled data set.
Step 2: and carrying out segmentation recognition on the picture to be detected by using the semantic segmentation network after training.
In this embodiment, semantic segmentation classifies the content of the picture from the pixels, that is, the pixels of the same class are classified into one class, so that some tiny areas cannot be avoided, or some areas with unreasonable aspect ratio, especially for the segmentation of metal rusting areas, the metal rusting is often distributed in a punctiform or blocky manner in the early stage, and the semantic segmentation network is used for identifying the metal rusting, so that large-scale and small-area segmentation areas can appear, and the subsequent processing fusion is required through the step 3.
Step 3: aiming at the situation that the interval between the oil leakage small area areas is lower than 5 pixel points, basic image processing operation is adopted, the basic image processing operation generally comprises operations of corrosion, expansion, connected domain extraction and the like, a plurality of small areas are fused into an oil leakage whole area, and metal corrosion is the same.
In this embodiment, the substation equipment is generally large and is distributed outdoors, and many external environmental factors such as rainwater, illumination and the like have influence on the detection result, so that in order to remove the error result generated by the external environmental factors, the final result is checked and determined through steps 4 to 6.
Step 4: carrying out image acquisition again on the sampling points at fixed time t intervals, and repeating the step 2 and the step 3;
in this embodiment, the interval time is not too short, and is generally not less than 0.5h, because the time setting is too short for the influence factors in the external environment, such as rainwater, illumination, etc., so that the change of the defect area cannot be detected, the false detection caused by the influence of the rainwater or the illumination cannot be eliminated, and the false detection rate is increased, so that the false detection rate can be effectively reduced by selecting a proper sampling interval time, and the detection effect is ensured.
Step 5: repeating the step 4 until n times of sampling are accumulated on the same sampling point, and ending;
in this embodiment, the number of sampling n is generally not less than 3, and the number of sampling is too small to exclude erroneous results according to the area change.
Step 6: recording the aspect ratio of the same area, eliminating the detection result with unreasonable area change, and determining the final target area. The same area refers to a designated area defined for an area photographed by a fixed camera.
Both the leakage and the metal corrosion are slow-expanding defects, which indicate that the false detection is likely to be caused by rainwater if the area becomes small, and that the false detection is likely to be caused by illumination if the area is large for a while and small for a while. After such false detection is excluded, the remaining areas are oil leakage and metal rust areas.
Example two
In the embodiment, the detection method is applied to the intelligent inspection robot, so that the defect position can be accurately and effectively identified.
The specific flow is shown in fig. 3:
1. design detection algorithm for detecting oil leakage and metal corrosion of transformer
2. Integrating algorithm application on intelligent inspection robot platform
3. Image input platform is shot and acquired through camera of intelligent inspection robot, and intelligent detection and identification are carried out on the input image by using detection algorithm
4. And finally, respectively outputting specific positions of the oil leakage and the metal corrosion.
The intelligent identification can completely avoid the defect condition of checking by people, and reduces great potential safety hazards; meanwhile, the algorithm also has high detection rate and accuracy.
In summary, the method for detecting the oil leakage and the metal corrosion of the transformer substation based on the semantic segmentation in the embodiment adopts the semantic segmentation technology to detect the defects, improves the defect detection rate, meets the requirement of high detection rate of the transformer substation, ensures higher accuracy rate while being high in detection rate, and is worth being popularized and used.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (6)
1. The method for detecting the leakage oil and the metal rust of the transformer substation based on semantic segmentation is characterized by comprising the following steps of:
s1: training network
Training a semantic segmentation network by using the marked training data set;
s2: segmentation identification
Using the semantic segmentation network after training to carry out segmentation recognition on the picture to be detected;
s3: region fusion
Aiming at each type of information, adopting basic image processing operation to respectively fuse into oil seepage areas and metal rust areas;
s4: sampling at intervals
The sampling points are subjected to image acquisition again at fixed intervals of time t, and the steps S2 and S3 are repeated;
in the step S4, the interval time t is not less than 0.5h;
s5: repeated sampling
Repeating the step S4 until n times of sampling are accumulated for the same sampling point;
in the step S5, the sampling number n is not less than 3;
s6: determining a target area
Recording according to the sampling time sequence aiming at the same area length-width ratio, removing detection results with unreasonable area change, and determining a final target area;
in the step S6, the types of detection results with unreasonable area change include two types, the first is that the detected area becomes smaller with time in the multiple detection results, and the second is that the detected area has two changing processes of becoming larger and smaller with time in the multiple detection results.
2. The method for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation according to claim 1, wherein the method comprises the following steps: the step S1 specifically comprises the following steps:
s101: collecting a data picture, manually segmenting and labeling the picture by using a labelme data labeling tool, and segmenting out an oil leakage area and a metal rust area which exist in the picture to form a training data set;
s102: training the semantic segmentation network using the labeled data set.
3. The method for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation according to claim 2, wherein the method comprises the following steps: in the step S102, the semantic segmentation network is a U-Net network.
4. The method for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation according to claim 1, wherein the method comprises the following steps: in the step S3, each type of information is the pixel information of leaked oil and metal rust after being identified by segmentation.
5. The method for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation according to claim 1, wherein the method comprises the following steps: in the step S3, the basic image processing operation includes erosion, swelling, connected domain extraction operations.
6. A system for detecting leakage oil and metal rust of a transformer substation based on semantic segmentation, characterized in that the detection method according to any one of claims 1 to 5 is used for detecting leakage oil and metal rust phenomena of transformer substation equipment, comprising:
the training module is used for training the semantic segmentation network by using the marked training data set;
the segmentation module is used for carrying out segmentation recognition on the picture to be detected by using the semantic segmentation network after the training is completed;
the area fusion module is used for respectively fusing the oil seepage areas and the metal rust areas by adopting basic image processing operation aiming at each type of information;
the first sampling module is used for carrying out image acquisition on the sampling points again at fixed time t intervals, and repeating the step S2 and the step S3;
the second sampling module is used for repeating the step S4 until n times of sampling are accumulated for the same sampling point;
the region determining module is used for recording the same region length-width ratio according to the sampling time sequence, removing the detection result with unreasonable region change and determining a final target region;
the control processing module is used for sending control instructions to the modules to complete related actions;
the training module, the segmentation module, the region fusion module, the first sampling module, the second sampling module and the region determination module are electrically connected with the control processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111049692.5A CN113763358B (en) | 2021-09-08 | 2021-09-08 | Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111049692.5A CN113763358B (en) | 2021-09-08 | 2021-09-08 | Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113763358A CN113763358A (en) | 2021-12-07 |
| CN113763358B true CN113763358B (en) | 2024-01-09 |
Family
ID=78793887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111049692.5A Active CN113763358B (en) | 2021-09-08 | 2021-09-08 | Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113763358B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456292B (en) * | 2023-12-26 | 2024-04-19 | 浙江晶盛机电股份有限公司 | Sapphire defect detection method, device, electronic device and storage medium |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036239A (en) * | 2014-05-29 | 2014-09-10 | 西安电子科技大学 | Fast high-resolution SAR (synthetic aperture radar) image ship detection method based on feature fusion and clustering |
| CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O＆M intelligent monitoring system and method |
| CN109285142A (en) * | 2018-08-07 | 2019-01-29 | 广州智能装备研究院有限公司 | A kind of head and neck neoplasm detection method, device and computer readable storage medium |
| CN109886238A (en) * | 2019-03-01 | 2019-06-14 | 湖北无垠智探科技发展有限公司 | Unmanned plane Image Change Detection algorithm based on semantic segmentation |
| CN110992317A (en) * | 2019-11-19 | 2020-04-10 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect detection method based on semantic segmentation |
| CN111104829A (en) * | 2018-10-29 | 2020-05-05 | 山东理工大学 | General method for cell or nanoparticle segmentation and identification based on image processing |
| CN111368687A (en) * | 2020-02-28 | 2020-07-03 | 成都市微泊科技有限公司 | Sidewalk vehicle illegal parking detection method based on target detection and semantic segmentation |
| CN111507398A (en) * | 2020-04-16 | 2020-08-07 | 浙江华云信息科技有限公司 | Transformer substation metal instrument corrosion identification method based on target detection |
| CN111815576A (en) * | 2020-06-23 | 2020-10-23 | 深圳供电局有限公司 | Method, device, equipment and storage medium for detecting corrosion condition of metal part |
| CN112258496A (en) * | 2020-11-02 | 2021-01-22 | 郑州大学 | Underground drainage pipeline disease segmentation method based on full convolution neural network |
| CN112348770A (en) * | 2020-09-09 | 2021-02-09 | 陕西师范大学 | Bridge crack detection method based on multi-resolution convolution network |
| CN112465840A (en) * | 2020-12-10 | 2021-03-09 | 重庆紫光华山智安科技有限公司 | Semantic segmentation model training method, semantic segmentation method and related device |
| CN112967272A (en) * | 2021-03-25 | 2021-06-15 | 郑州大学 | Welding defect detection method and device based on improved U-net and terminal equipment |
| CN113344739A (en) * | 2021-06-09 | 2021-09-03 | 北京九贺科技有限公司 | Data safety supervision method and system for automatically checking power failure |
-
2021
- 2021-09-08 CN CN202111049692.5A patent/CN113763358B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036239A (en) * | 2014-05-29 | 2014-09-10 | 西安电子科技大学 | Fast high-resolution SAR (synthetic aperture radar) image ship detection method based on feature fusion and clustering |
| CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O＆M intelligent monitoring system and method |
| CN109285142A (en) * | 2018-08-07 | 2019-01-29 | 广州智能装备研究院有限公司 | A kind of head and neck neoplasm detection method, device and computer readable storage medium |
| CN111104829A (en) * | 2018-10-29 | 2020-05-05 | 山东理工大学 | General method for cell or nanoparticle segmentation and identification based on image processing |
| CN109886238A (en) * | 2019-03-01 | 2019-06-14 | 湖北无垠智探科技发展有限公司 | Unmanned plane Image Change Detection algorithm based on semantic segmentation |
| CN110992317A (en) * | 2019-11-19 | 2020-04-10 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect detection method based on semantic segmentation |
| CN111368687A (en) * | 2020-02-28 | 2020-07-03 | 成都市微泊科技有限公司 | Sidewalk vehicle illegal parking detection method based on target detection and semantic segmentation |
| CN111507398A (en) * | 2020-04-16 | 2020-08-07 | 浙江华云信息科技有限公司 | Transformer substation metal instrument corrosion identification method based on target detection |
| CN111815576A (en) * | 2020-06-23 | 2020-10-23 | 深圳供电局有限公司 | Method, device, equipment and storage medium for detecting corrosion condition of metal part |
| CN112348770A (en) * | 2020-09-09 | 2021-02-09 | 陕西师范大学 | Bridge crack detection method based on multi-resolution convolution network |
| CN112258496A (en) * | 2020-11-02 | 2021-01-22 | 郑州大学 | Underground drainage pipeline disease segmentation method based on full convolution neural network |
| CN112465840A (en) * | 2020-12-10 | 2021-03-09 | 重庆紫光华山智安科技有限公司 | Semantic segmentation model training method, semantic segmentation method and related device |
| CN112967272A (en) * | 2021-03-25 | 2021-06-15 | 郑州大学 | Welding defect detection method and device based on improved U-net and terminal equipment |
| CN113344739A (en) * | 2021-06-09 | 2021-09-03 | 北京九贺科技有限公司 | Data safety supervision method and system for automatically checking power failure |
Non-Patent Citations (2)
| Title |
|---|
| 基于聚合通道特征的防震锤锈蚀缺陷识别算法;孙长翔;邱翔;罗希;黄前华;曹成功;;计算技术与自动化(第02期);全文 * |
| 融合深度图像的卷积神经网络语义分割方法;王孙平;陈世峰;;集成技术(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113763358A (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210409B (en) | Method and system for detecting form frame lines in form document | |
| CN110427860B (en) | Lane line identification method and device and storage medium | |
| CN110232380B (en) | Fire night scene restoration method based on Mask R-CNN neural network | |
| CN110321933B (en) | Fault identification method and device based on deep learning | |
| CN110346699B (en) | Insulator discharge information extraction method and device based on ultraviolet image processing technology | |
| CN111626145B (en) | Simple and effective incomplete form identification and page-crossing splicing method | |
| CN110619623B (en) | Automatic identification method for heating of joint of power transformation equipment | |
| CN113763358B (en) | Method and system for detecting oil leakage and metal corrosion of transformer substation based on semantic segmentation | |
| CN115063725B (en) | Aircraft skin defect identification system based on multi-scale self-adaptive SSD algorithm | |
| CN114519694B (en) | Seven-segment nixie tube liquid crystal display screen identification method and system based on deep learning | |
| CN110991414A (en) | High-precision traffic element segmentation method, electronic equipment and storage medium | |
| CN114241310B (en) | Improved YOLO model-based intelligent identification method for piping dangerous case of dike | |
| CN113486856A (en) | Driver irregular behavior detection method based on semantic segmentation and convolutional neural network | |
| CN112597996B (en) | Method for detecting traffic sign significance in natural scene based on task driving | |
| Song et al. | Bolt looseness detection based on Canny edge detection algorithm | |
| CN112330659B (en) | Geometric tolerance symbol segmentation method combining LSD (least squares) linear detection and connected domain marking method | |
| CN115984378A (en) | Track foreign matter detection method, device, equipment and medium | |
| CN114419026B (en) | Visual attention-based aeroengine fuse winding direction identification system and method | |
| CN112734745B (en) | Unmanned aerial vehicle thermal infrared image heating pipeline leakage detection method fusing GIS data | |
| WO2022198507A1 (en) | Obstacle detection method, apparatus, and device, and computer storage medium | |
| CN112419316A (en) | Cross-device visible light texture defect detection method and device | |
| CN114037840A (en) | Power transmission line visual object extraction method and system based on multi-feature fusion | |
| Wazalwar et al. | Design flow for robust license plate localization | |
| CN112307873A (en) | Automatic illegal building identification method based on full convolution neural network | |
| Lee et al. | High-Speed Maritime Object Detection Scheme for the Protection of the Aid to Navigation |
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 |