CN110852320A - Transmission channel foreign matter intrusion detection method based on deep learning - Google Patents
Transmission channel foreign matter intrusion detection method based on deep learning Download PDFInfo
- Publication number
- CN110852320A CN110852320A CN201911092068.6A CN201911092068A CN110852320A CN 110852320 A CN110852320 A CN 110852320A CN 201911092068 A CN201911092068 A CN 201911092068A CN 110852320 A CN110852320 A CN 110852320A
- Authority
- CN
- China
- Prior art keywords
- transmission channel
- power transmission
- distance
- foreign matter
- wire
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Abstract
The invention relates to a foreign matter intrusion detection method for a power transmission channel based on deep learning, aiming at overhead line conductors in the power transmission channel, the semantic segmentation of the conductors is realized by utilizing an improved Mask-RCNN conductor detection method; when the distance between the wire and the foreign matter is subjected to intrusion detection, an optimized deduplication algorithm is provided, and more accurate pixel level shapes of the wire are obtained; when the foreign matter invades, a monocular distance measurement method is used, and the distance between the lead and the foreign matter is obtained by conversion of a reference object; and for the detection of the foreign matter invasion of the power transmission channel, a real-time monitoring and early warning method is adopted to monitor and early warn the foreign matter invasion. The method uses the guide image aiming at the power transmission channel to perform guide filtering, thereby enhancing the target characteristics, weakening the influence caused by weather and improving the robustness; an improved Mask R-CNN algorithm is provided for carrying out pixel level division on a target, so that distance measurement is facilitated; and an algorithm is provided to fit the shape of the wire again, and the weight is removed, so that the accuracy is improved.
Description
Technical Field
The invention belongs to the technical field of power grid operation and maintenance, particularly relates to a deep learning-based power transmission channel foreign matter intrusion detection method, relates to the fields of filtering algorithm, image processing, deep learning, semantic segmentation and the like, and can be used for detecting abnormal conditions such as intrusion of hoisting machinery into a power transmission channel and the like which endanger safe operation of a power grid.
Background
During the power transmission, the following two main problems are faced: firstly, due to strong wind and other reasons, the distance between the leads is reduced, the limit discharge distance of the high-voltage line is shortened, the safety of a transmission line is endangered, and even large-area power failure is caused; secondly, large-scale operation equipment such as cranes near the power transmission line invade the power transmission channel, and the power transmission line is often damaged.
The current common solution is manual investigation, however, the method is time-consuming, labor-consuming and costly. Based on this, an efficient and automatic foreign matter identification method is needed to help identify and early warn various potential safety hazards of the power channel. The prior art methods comprise:
(1) the wire identification method based on the SSD comprises the following steps: the method has the main idea that intensive sampling is uniformly carried out at different positions of a picture, different scales and aspect ratios can be adopted during sampling, then classification and regression are directly carried out after the characteristics are extracted by using CNN, and the whole process only needs one step, so that the method has the advantages of high speed, but the accuracy cannot achieve the aim of carrying out pixel-level distance detection on a target.
(2) The improved fast R-CNN high-voltage cable target detection method comprises the following steps: the improved fast R-CNN high-voltage cable target detection method introduces skip connection and adjusts the sequence of an active layer and a convolution layer; then, a candidate frame generation mechanism is improved, and the performance of the network for detecting the small target is improved; and finally, extracting the characteristics of each region by using the ROI pooling layer, and finishing classification and frame regression tasks. But the effect is not ideal and the anti-interference capability is poor under the complex weather environment.
(3) The target detection method under the power environment by using semantic segmentation comprises the following steps: semantic segmentation gives us a more detailed understanding of an image than image classification or object detection, because it achieves a pixel-to-pixel mapping, specifically to the pixel level, when processing an image.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the invention uses a semantic segmentation algorithm to detect the invasion of foreign matters such as wires and cranes, so as to achieve the judgment of pixel level, and simultaneously uses a deep learning algorithm to optimize the shape and remove the duplicate of the target of the detection result, thereby avoiding false detection. The technical scheme adopted by the invention is as follows:
the foreign matter intrusion detection method for the power transmission channel based on deep learning comprises the following steps:
step 1, aiming at overhead line conductors in a power transmission channel, semantic segmentation of the conductors is realized by using an improved Mask-RCNN conductor detection method;
step 2, when intrusion detection is carried out on the distance between the wire and the foreign matter, an optimized deduplication algorithm is provided, and more accurate pixel level shapes of the wire are obtained;
step 3, when the foreign matter invades, a monocular distance measurement method is used, and the distance between the conducting wire and the foreign matter is obtained through conversion of a reference object;
and 4, monitoring and early warning the intrusion of the foreign matters in the power transmission channel by adopting a real-time monitoring and early warning method.
Preferably, the specific steps of implementing semantic segmentation of the wire in step 1 are as follows: aiming at overhead line conductors in a power transmission channel, an improved Mask-RCNN conductor detection method is designed, a feature extraction is carried out through a convolution neural Network to generate a feature map of a conductor image, anchors (anchor points, fixed reference frames generated during detection) aiming at conductor features and with different aspect ratios are added into an RPN (Region pro-polar Network), the RPN is utilized to generate a Region ROI (Region of interest) with possible conductors in an original image feature map, then an ROI operation (a Region feature aggregation mode is carried out through a bilinear interpolation method, the problem of Region mismatching caused by two quantization in ROI Pooling operation is well solved), pixel alignment is carried out, further convolution and Pooling operation are carried out on each candidate Region, whether the ROI comprises the conductors or not is judged, and the coordinates are subjected to regression correction, meanwhile, Mask branches are added, Mask information of the guide lines is generated in the guide line target area (namely, covering identification is carried out in the area where the target possibly exists, and the Mask is called as a Mask), and semantic segmentation of the guide lines is achieved.
Preferably, the specific steps of obtaining a more accurate pixel-level shape of the conductive line in step 2 are as follows: when the distance between a wire and a foreign body is subjected to intrusion detection, an optimized deduplication algorithm is provided for the problem that the foreign body characteristics and the shape are discontinuous during wire detection, small areas adjacent to the foreign body shape are merged, repeated problems are removed, all wire mask pixel points are averaged at the same time, the wire shape is modeled, wire strip shapes with different lengths are generated, and by properly prolonging the wire strips, the wire segments with shorter lengths are merged on long wire segments with the same direction and the same positions, and more accurate pixel level shapes of the wires are obtained.
Preferably, the specific steps of obtaining the distance between the wire and the foreign object in step 3 are as follows: when the foreign matter invades, a monocular distance measurement method is used, a reference object with a known length is preset under a fixed camera, the preset reference object, a lead and the foreign matter under the fixed camera are respectively identified, and then the distance between the lead and the foreign matter is converted according to the proportion of the distance between the lead and the foreign matter relative to the reference object.
Preferably, the specific steps of monitoring and early warning by using the real-time monitoring and early warning method in step 4 are as follows: for the detection of the invasion of the foreign matters in the power transmission channel, a real-time monitoring and early warning method is adopted, after the conducting wires and the foreign matters are identified, pixel points of the conducting wires and the foreign matters are traversed, the shortest Euclidean distance between the conducting wires and the foreign matters on the pixels is calculated, the detected shortest conducting wires and foreign matters distance is drawn, a threshold value is set according to the set safety distance between the conducting wires and the foreign matters, and when the distance is smaller than the threshold value, the monitoring and early warning are carried out on the conducting wires and the.
Preferably, in order to improve robustness of the distance detection method in the case of foreign matter intrusion, the recognition difficulty is increased in consideration of the fact that a power transmission channel is in a foggy and rainy day, and therefore preprocessing of an image is added. The image is processed using a guide filter for the power transmission channel, the image is noise-reduced using a guide image, and a target image P (input image) is filter-processed through a guide map G (guide map) so that the final output image is substantially similar to the target image P, but the texture portion is similar to the guide map G. This preserves image edge information to make the target features more visible than isotropic filtering (e.g., simple smoothing or gaussian smoothing).
The invention has the beneficial effects that:
(1) and the guide image aiming at the power transmission channel is used for guiding and filtering, so that the target characteristic is enhanced, the influence caused by weather is weakened, and the robustness is improved.
(2) An improved Mask R-CNN algorithm is provided for carrying out pixel level division on the target, so that distance measurement is facilitated.
(3) And an algorithm is provided to fit the shape of the wire again, and the weight is removed, so that the accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flow chart of intrusion detection of a hoisting machine of a power transmission channel according to an embodiment of the present invention;
FIG. 2 is a front and back comparison of the guided filtering effect of a power transmission channel according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the lead recognition principle of the improved Mask-RCNN algorithm;
fig. 4 is a diagram of an effect of the detection and early warning in the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention takes the intrusion detection of hoisting machinery of a power transmission channel as an example for explanation. As shown in fig. 1, a flowchart of intrusion detection of a hoisting machine in a power transmission channel according to an embodiment of the present invention includes the following steps:
and S1, power transmission channel image input.
S2, preprocessing the image by using a guide filtering method aiming at the power transmission channel;
as shown in fig. 2, which is a front-back comparison diagram of the power transmission channel oriented filtering effect according to the embodiment of the present invention, it can be seen that the image edge information can be retained after the preprocessing, so that the target feature is more obvious.
S3, conducting wire recognition through conducting wire semantic segmentation by using an improved Mask R-CNN algorithm, and carrying out optimal shape fitting duplication elimination on the recognized result by using a shape fitting duplication elimination algorithm;
and identifying the hoisting machine.
Aiming at the lead under the power transmission channel, the embodiment of the invention designs an improved Mask-RCNN lead detection method, which is shown in fig. 3 and is a lead recognition principle schematic diagram of an improved Mask-RCNN algorithm.
Aiming at the problem of discontinuous wire detection, an optimized wire shape fitting and de-duplication algorithm is provided, and more accurate pixel level shapes of the wires are obtained.
And S4, obtaining the Euclidean pixel distance between the targets generated in the step 3 by using a distance detection algorithm, and then obtaining the actual distance by comparing the Euclidean pixel distance with the length of the reference object.
And S5, early warning is carried out on the invading hoisting machinery through distance judgment.
For the detection of the invasion of foreign matters such as a power transmission channel hoisting machine and the like, a real-time monitoring and early warning method is adopted, a threshold value is set according to a set safe distance between a lead and large-scale equipment such as a crane and the like, and when the distance is smaller than the threshold value, the monitoring and early warning are carried out on the lead. Fig. 4 is a diagram illustrating an effect of the detection and early warning in the embodiment of the present invention. In the figure, the distance between the hoisting machinery and the lead is 13 meters, and early warning is needed.
The power transmission channel hoisting machinery intrusion detection early warning flow chart based on deep learning can early warn potential safety hazards under a power transmission channel, reduce interference of weather environment on identification and have strong robustness and high accuracy.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. The foreign matter intrusion detection method for the power transmission channel based on deep learning is characterized by comprising the following steps:
step 1, aiming at overhead line conductors in a power transmission channel, semantic segmentation of the conductors is realized by using an improved Mask-RCNN conductor detection method;
step 2, when the distance between the wire and the foreign matter is subjected to intrusion detection, an optimized deduplication algorithm is utilized to obtain a more accurate pixel-level shape of the wire;
step 3, when the foreign matter invades, a monocular distance measurement method is used, and the distance between the conducting wire and the foreign matter is obtained through conversion of a reference object;
and 4, monitoring and early warning the intrusion of the foreign matters in the power transmission channel by adopting a real-time monitoring and early warning method.
2. The method for detecting the intrusion of the foreign matters into the power transmission channel according to claim 1, wherein the specific steps of realizing the semantic segmentation of the leads in the step 1 are as follows:
an improved Mask-RCNN lead detection method is designed for overhead line leads in a power transmission channel, feature extraction is carried out through a convolutional neural network to generate a feature map of a lead image, anchors with different aspect ratios are added to RPN according to lead features, the RPN is utilized to generate a region ROI possibly having leads in an original image feature map, then ROIAlign operation is carried out through a bilinear interpolation method to carry out pixel alignment, further convolution and pooling operation are carried out on each candidate region to judge whether the ROI includes the leads or not, regression correction is carried out on a coordinate, Mask branches are added at the same time, Mask information of the leads is generated in a lead target region, and semantic segmentation of the leads is achieved.
3. The method for detecting foreign matter intrusion in power transmission channels according to claim 1, wherein the specific steps of obtaining a more accurate pixel-level shape of the conductor in step 2 are as follows:
when the distance between a wire and a foreign body is subjected to intrusion detection, an optimized deduplication algorithm is provided for the problem that the foreign body characteristics and the shape are discontinuous during wire detection, small areas adjacent to the foreign body shape are merged, repeated problems are removed, all wire mask pixel points are averaged at the same time, the wire shape is modeled, wire strip shapes with different lengths are generated, and by properly prolonging the wire strips, the wire segments with shorter lengths are merged on long wire segments with the same direction and the same positions, and more accurate pixel level shapes of the wires are obtained.
4. The method for detecting the intrusion of the foreign matters into the power transmission channel according to claim 1, wherein the specific steps of obtaining the distance between the lead and the foreign matters in the step 3 are as follows:
when the foreign matter invades, a monocular distance measurement method is used, a reference object with a known length is preset under a fixed camera, the preset reference object, a lead and the foreign matter under the fixed camera are respectively identified, and then the distance between the lead and the foreign matter is converted according to the proportion of the distance between the lead and the foreign matter relative to the reference object.
5. The method for detecting foreign matter intrusion in a power transmission channel according to claim 1, wherein the specific steps of monitoring and early warning by adopting a real-time monitoring and early warning method in the step 4 are as follows:
for the detection of the invasion of the foreign matters in the power transmission channel, a real-time monitoring and early warning method is adopted, after the conducting wires and the foreign matters are identified, pixel points of the conducting wires and the foreign matters are traversed, the shortest Euclidean distance between the conducting wires and the foreign matters on the pixels is calculated, the detected shortest conducting wires and foreign matters distance is drawn, a threshold value is set according to the set safety distance between the conducting wires and the foreign matters, and when the distance is smaller than the threshold value, the monitoring and early warning are carried out on the conducting wires and the.
6. The method according to any one of claims 1 to 5, wherein preprocessing of the image is added, the image is processed using guided filtering for the power transmission channel, the image is de-noised using the guide image, the target image P is filtered through a guide map G, so that the final output image is substantially similar to the target image P, but the texture part is similar to the guide map G.
7. The method for detecting foreign matter intrusion in a power transmission channel according to claim 6, wherein the method for detecting mechanical intrusion in a hoisting machine in a power transmission channel comprises the following steps:
s1, inputting a power transmission channel image;
s2, preprocessing the image by using a guide filtering method aiming at the power transmission channel;
s3, conducting wire recognition through conducting wire semantic segmentation by using an improved Mask R-CNN algorithm, and carrying out optimal shape fitting duplication elimination on the recognized result by using a shape fitting duplication elimination algorithm; and identifying the hoisting machinery;
s4, obtaining Euclidean pixel distance between the targets generated in the step 3 by using a distance detection algorithm, and then obtaining an actual distance by comparing the Euclidean pixel distance with the length of a reference object;
and S5, early warning is carried out on the invading hoisting machinery through distance judgment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911092068.6A CN110852320B (en) | 2019-11-08 | 2019-11-08 | Transmission channel foreign matter intrusion detection method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911092068.6A CN110852320B (en) | 2019-11-08 | 2019-11-08 | Transmission channel foreign matter intrusion detection method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110852320A true CN110852320A (en) | 2020-02-28 |
| CN110852320B CN110852320B (en) | 2022-06-24 |
Family
ID=69601022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911092068.6A Active CN110852320B (en) | 2019-11-08 | 2019-11-08 | Transmission channel foreign matter intrusion detection method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110852320B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111428688A (en) * | 2020-04-16 | 2020-07-17 | 成都旸谷信息技术有限公司 | Intelligent vehicle driving lane identification method and system based on mask matrix |
| CN112001066A (en) * | 2020-07-30 | 2020-11-27 | 四川大学 | Deep learning-based method for calculating limit transmission capacity |
| CN112766137A (en) * | 2021-01-14 | 2021-05-07 | 华南理工大学 | Dynamic scene foreign matter intrusion detection method based on deep learning |
| CN112909824A (en) * | 2021-03-24 | 2021-06-04 | 南方电网电力科技股份有限公司 | Method and device for identifying suspended foreign matters of power transmission line |
| CN113343856A (en) * | 2021-06-09 | 2021-09-03 | 北京容联易通信息技术有限公司 | Image recognition method and system |
| CN114677568A (en) * | 2022-05-30 | 2022-06-28 | 山东极视角科技有限公司 | Linear target detection method, module and system based on neural network |
| CN115205796A (en) * | 2022-07-07 | 2022-10-18 | 北京交通大学 | Method and system for monitoring foreign matter invasion limit and early warning risk of track line |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104215217A (en) * | 2014-09-19 | 2014-12-17 | 成都翼比特科技有限责任公司 | Monocular distance measurement method based on pseudo-image distance |
| CN106127778A (en) * | 2016-06-27 | 2016-11-16 | 安徽慧视金瞳科技有限公司 | A kind of line detection method for projecting interactive system |
| CN107680095A (en) * | 2017-10-25 | 2018-02-09 | 哈尔滨理工大学 | The electric line foreign matter detection of unmanned plane image based on template matches and optical flow method |
| US20180098137A1 (en) * | 2016-06-12 | 2018-04-05 | Green Grid Inc. | Method and system for utility power lines vegetation proximity monitoring and controlling |
| CN108734143A (en) * | 2018-05-28 | 2018-11-02 | 江苏迪伦智能科技有限公司 | A kind of transmission line of electricity online test method based on binocular vision of crusing robot |
| CN109284739A (en) * | 2018-10-26 | 2019-01-29 | 福州大学 | A kind of preventing damage to power transmission line caused by external force method for early warning and system based on deep learning |
| CN109712116A (en) * | 2018-12-04 | 2019-05-03 | 华南理工大学 | The fault recognition method of transmission line of electricity and its attachment |
| CN110400267A (en) * | 2019-07-03 | 2019-11-01 | 安徽继远软件有限公司 | A kind of preprocess method based on inspection image |
-
2019
- 2019-11-08 CN CN201911092068.6A patent/CN110852320B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104215217A (en) * | 2014-09-19 | 2014-12-17 | 成都翼比特科技有限责任公司 | Monocular distance measurement method based on pseudo-image distance |
| US20180098137A1 (en) * | 2016-06-12 | 2018-04-05 | Green Grid Inc. | Method and system for utility power lines vegetation proximity monitoring and controlling |
| CN106127778A (en) * | 2016-06-27 | 2016-11-16 | 安徽慧视金瞳科技有限公司 | A kind of line detection method for projecting interactive system |
| CN107680095A (en) * | 2017-10-25 | 2018-02-09 | 哈尔滨理工大学 | The electric line foreign matter detection of unmanned plane image based on template matches and optical flow method |
| CN108734143A (en) * | 2018-05-28 | 2018-11-02 | 江苏迪伦智能科技有限公司 | A kind of transmission line of electricity online test method based on binocular vision of crusing robot |
| CN109284739A (en) * | 2018-10-26 | 2019-01-29 | 福州大学 | A kind of preventing damage to power transmission line caused by external force method for early warning and system based on deep learning |
| CN109712116A (en) * | 2018-12-04 | 2019-05-03 | 华南理工大学 | The fault recognition method of transmission line of electricity and its attachment |
| CN110400267A (en) * | 2019-07-03 | 2019-11-01 | 安徽继远软件有限公司 | A kind of preprocess method based on inspection image |
Non-Patent Citations (3)
| Title |
|---|
| WEISHAN ZHANG ETC.: "RCNN-based foreign object detection for securing power transmission lines (RCNN4SPTL)", 《PROCEDIA COMPUTER SCIENCE》 * |
| 刘召 等: "基于改进的 Faster R-CNN 高压线缆目标检测方法", 《智能系统学报》 * |
| 李保坤: "基于深度学习的输电导线检测算法研究", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111428688A (en) * | 2020-04-16 | 2020-07-17 | 成都旸谷信息技术有限公司 | Intelligent vehicle driving lane identification method and system based on mask matrix |
| CN111428688B (en) * | 2020-04-16 | 2022-07-26 | 成都旸谷信息技术有限公司 | Intelligent vehicle driving lane identification method and system based on mask matrix |
| CN112001066A (en) * | 2020-07-30 | 2020-11-27 | 四川大学 | Deep learning-based method for calculating limit transmission capacity |
| CN112766137A (en) * | 2021-01-14 | 2021-05-07 | 华南理工大学 | Dynamic scene foreign matter intrusion detection method based on deep learning |
| CN112766137B (en) * | 2021-01-14 | 2023-02-10 | 华南理工大学 | Dynamic scene foreign matter intrusion detection method based on deep learning |
| CN112909824A (en) * | 2021-03-24 | 2021-06-04 | 南方电网电力科技股份有限公司 | Method and device for identifying suspended foreign matters of power transmission line |
| CN113343856A (en) * | 2021-06-09 | 2021-09-03 | 北京容联易通信息技术有限公司 | Image recognition method and system |
| CN113343856B (en) * | 2021-06-09 | 2022-03-29 | 北京容联易通信息技术有限公司 | Image recognition method and system |
| CN114677568A (en) * | 2022-05-30 | 2022-06-28 | 山东极视角科技有限公司 | Linear target detection method, module and system based on neural network |
| CN115205796A (en) * | 2022-07-07 | 2022-10-18 | 北京交通大学 | Method and system for monitoring foreign matter invasion limit and early warning risk of track line |
| CN115205796B (en) * | 2022-07-07 | 2023-05-16 | 北京交通大学 | Rail line foreign matter intrusion monitoring and risk early warning method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110852320B (en) | 2022-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110852320B (en) | Transmission channel foreign matter intrusion detection method based on deep learning | |
| CN109785285B (en) | Insulator damage detection method based on ellipse characteristic fitting | |
| CN107742301B (en) | Image classification-based power transmission line image processing method under complex background | |
| CN108307146B (en) | System and method for detecting potential safety hazard of high-voltage transmission line | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| CN110321933B (en) | Fault identification method and device based on deep learning | |
| CN106290388A (en) | A kind of insulator breakdown automatic testing method | |
| CN111814686A (en) | Vision-based power transmission line identification and foreign matter invasion online detection method | |
| CN107818563B (en) | Space measurement and positioning method for split conductor spacing of power transmission line | |
| CN109300118B (en) | High-voltage power line unmanned aerial vehicle inspection method based on RGB image | |
| CN107179479B (en) | Visible light image-based transmission conductor strand scattering defect detection method | |
| CN113313677B (en) | Quality detection method for X-ray image of wound lithium battery | |
| CN111402247A (en) | Machine vision-based method for detecting defects of suspension clamp on power transmission line | |
| CN115631116B (en) | Aircraft power inspection system based on binocular vision | |
| CN108734704B (en) | Transmission conductor strand breakage detection method based on gray variance normalization | |
| CN115512252B (en) | Unmanned aerial vehicle-based power grid inspection automation method and system | |
| CN110910443A (en) | Contact net geometric parameter real-time measuring method and device based on single monitoring camera | |
| CN112819784A (en) | Method and system for detecting broken strands and scattered strands of wires of distribution line | |
| CN110276787B (en) | Conductor galloping monitoring method based on marker image detection | |
| CN111008967B (en) | Insulator RTV coating defect identification method | |
| CN111127381B (en) | Non-parallel detection method for pantograph slide plate | |
| CN114220009A (en) | Infrared image-based wire windage yaw identification method and system | |
| CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
| CN110782411B (en) | Anti-vibration hammer slip identification method based on image processing | |
| CN107609556B (en) | Detection method for high-altitude operation machine in power transmission line environment |
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 |