CN114723678A - High-voltage wire foreign matter detection method and detection system based on video image - Google Patents
High-voltage wire foreign matter detection method and detection system based on video image Download PDFInfo
- Publication number
- CN114723678A CN114723678A CN202210276445.7A CN202210276445A CN114723678A CN 114723678 A CN114723678 A CN 114723678A CN 202210276445 A CN202210276445 A CN 202210276445A CN 114723678 A CN114723678 A CN 114723678A
- Authority
- CN
- China
- Prior art keywords
- image
- frame
- foreign matter
- images
- voltage 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 82
- 238000002407 reforming Methods 0.000 claims abstract description 3
- 238000000034 method Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 12
- 239000004744 fabric Substances 0.000 claims description 5
- 229920003023 plastic Polymers 0.000 claims description 5
- 239000004033 plastic Substances 0.000 claims description 5
- 230000003321 amplification Effects 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 206010070245 Foreign body Diseases 0.000 claims 5
- 230000005540 biological transmission Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000002349 favourable effect Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
-
- 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
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The application discloses high tension line foreign matter detection method based on video image, it includes: acquiring a video image which needs to be subjected to high-voltage wire foreign matter detection, acquiring two adjacent frames of images of the video image at a time interval of m minutes, and performing high-voltage wire foreign matter detection based on the two adjacent frames of images at the time interval of m minutes; and acquiring a next frame image at an interval of m minutes, reforming the next frame image and a next frame image in the two adjacent frame images into two adjacent frame images at an interval of m minutes, and continuously detecting the foreign matters in the high-voltage wire on the basis of the two adjacent frame images at the interval of m minutes. The step of detecting the foreign matters in the high-voltage wire based on the two adjacent frames of images comprises the following steps: detecting all foreign matters in each frame of image in two adjacent frames of images with the time interval of m minutes; judging whether each detected foreign matter in each frame image is positioned on the high-voltage wire or not; and carrying out foreign matter rechecking between two adjacent frames of images. The application also provides a high-voltage wire foreign matter detection system.
Description
Technical Field
The present disclosure relates to the field of image processing, and more particularly, to a method and a system for detecting foreign matters in a high-voltage cable based on video images.
Background
Electric power is an important basic industry of national economy, and safe and stable power supply is a precondition for ensuring the rapid and stable development of national economy. In a national power grid, the environment of a high-voltage transmission line is very complex and is easily damaged, so that it is very important to ensure the operation safety of the high-voltage transmission line.
In recent years, events that power grid safety is endangered by hanging foreign matters on kites, balloons, fabrics, plastic bags and the like are frequent in various places, and the limit discharge distance of high voltage electricity is shortened by hanging the foreign matters on power transmission lines, and even serious consequences of large-area power failure are caused. Therefore, the method has very important significance in timely identifying the foreign matters on the power transmission line. The existing foreign matter inspection of the power transmission line is mainly manual inspection or unmanned inspection; but artifical line inspection has the potential safety hazard big, and the artifical increase personnel working strength of patrolling and examining just influences detection efficiency, and the shortcoming of unmanned aerial vehicle mode is that unmanned aerial vehicle controls the degree of difficulty, battery continuation of the journey weak point, can not be for a long time uninterrupted duty.
However, the intelligent video processing technology performs high-speed analysis on mass data in a video monitoring picture by means of a powerful data processing function of a computer, filters information which is not concerned by monitoring personnel, provides useful key information for the monitoring personnel, can greatly reduce the working intensity of the video monitoring personnel, can reduce false alarm and missed alarm, and improves the timeliness of alarm processing. Therefore, it is currently urgently needed to design a method for detecting a high-voltage wire foreign object intrusion target based on video image data.
Disclosure of Invention
Aiming at the prior art, the technical problem to be solved by the application is to provide a video image-based high-voltage wire foreign matter detection method and a video image-based high-voltage wire foreign matter detection system which can effectively reduce manual participation, improve detection efficiency and alarm in time,
in order to solve the above technical problem, the present application provides a method for detecting a foreign object in a high-voltage wire based on a video image, including:
the method for detecting the foreign matters on the high-voltage wire comprises the following steps of acquiring a video image needing high-voltage wire foreign matter detection, acquiring two adjacent frames of images of the video image at a time interval of m minutes, and detecting the foreign matters on the high-voltage wire based on the two adjacent frames of images at the time interval of m minutes, wherein the method comprises the following steps:
detecting all foreign matters in each frame of image in two adjacent frames of images with the time interval of m minutes;
judging whether each detected foreign matter in each frame image is positioned on the high-voltage wire or not;
carrying out foreign matter rechecking between two adjacent frames of images: judging whether each foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image, if so, giving an alarm, and if not, deleting the foreign matter on the high-voltage wire in the previous frame; and the number of the first and second groups,
and acquiring a next frame image at an interval of m minutes, reforming the next frame image and a next frame image in the two adjacent frame images into two adjacent frame images at an interval of m minutes, and continuously detecting the foreign matters in the high-voltage wire on the basis of the two adjacent frame images at the interval of m minutes.
In one possible implementation, the step of detecting all the foreign objects in each of two adjacent frames of images separated by a time interval of m minutes comprises:
slicing each frame of image in two adjacent frames of images: equally slicing each frame image to obtain n grid images and recording the position of each grid image in the frame image;
carrying out batch foreign matter detection on each frame image corresponding to all the grid images to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image;
restoring the detected foreign matters and the coordinate frames corresponding to the foreign matters of each grid image into the frame images corresponding to the grid images, and expanding each coordinate frame in each frame image to be k times of the original coordinate frame.
In a possible implementation manner, inputting each frame of image corresponding to all the grid images into a foreign matter detection model for batch foreign matter detection, and obtaining the foreign matters framed by the coordinate frame in all the grid images of the frame of image comprises:
amplifying each grid image of the frame image to 4/3 times of the original grid image;
sequentially superposing the n grid images after the amplification treatment along an X axis and a Y axis to form a group of image groups;
and inputting the image group corresponding to each frame image into a foreign matter detection model to perform batch foreign matter detection to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image.
In one possible implementation, the determining whether each detected foreign object in each frame of image is located on the high voltage line includes:
acquiring a positive sample that the trained foreign matters stay on the high-voltage wire and a negative sample that the foreign matters do not stay on the high-voltage wire; inputting the positive sample and the negative sample into an efficientnet-b1 deep learning network model for training to obtain a recognition model of a classification result, wherein the classification result is foreign matters on a high-voltage wire and foreign matters on a non-high-voltage wire;
and inputting the image of the expanded coordinate framed area in each frame image into the recognition model.
In one possible implementation, the size of each frame of image obtained is 1920 × 1080, each frame of image is equally sliced to obtain 16 grid images with the size of 480 × 270, and each grid image is enlarged to be 640 × 360.
In one possible implementation, the foreign object detection model is obtained based on deep learning model training, and includes: acquiring training image data of plastic bags, kites, branches, balloons, fabrics and birds, and inputting the training image data into a yolov5 learning model to train to obtain the foreign matter detection model.
In one possible implementation, the step of determining whether each foreign object on the high-voltage electric wire of the previous frame image repeatedly appears on the subsequent frame image includes:
calculating the IOU intersection ratio of a coordinate frame A of each foreign matter on the high-voltage wire in the previous frame of image and a coordinate frame B of any foreign matter in the next frame of image, wherein if the IOU intersection ratio is more than or equal to 0.5, the foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image;
wherein the IOU intersection ratio is as follows: (A. andgate.B)/(A. U.B).
In one possible implementation, the m minutes is 1 minute or 2 minutes.
In one possible implementation, the value of k is 1.2 or 1.5.
In the video image-based high-voltage wire foreign matter detection method, the video image-based processing method is used for detecting foreign matters on the high-voltage wire, so that the manual participation can be effectively reduced, the detection efficiency is improved, and the alarm can be given in time; meanwhile, foreign matter detection is continuously carried out by continuously acquiring frame images, so that the effect of long-term detection is achieved; and the recheck between adjacent frames is favorable for improving the detection accuracy, and the false detection and false alarm phenomena can be effectively avoided.
The application also provides a high-voltage wire foreign matter detection system which comprises a collection device, a storage, a processor and an alarm, wherein the processor is connected with the storage, the collection device and the alarm; the device comprises a memory, a processor, a collecting device and an alarm, wherein the memory is used for storing executable codes, the processor is used for executing the executable codes stored in the memory so as to execute the high-voltage wire foreign matter detection method based on video images, the collecting device is used for collecting the video images needing high-voltage wire foreign matter detection, and the alarm is used for giving an alarm when the processor detects foreign matters on a high-voltage wire.
In the high-voltage wire foreign matter detection system, the foreign matter detection on the high-voltage wire is carried out by using a processing method based on video images, so that the manual participation can be effectively reduced, the detection efficiency is improved, and the alarm can be given in time; meanwhile, foreign matter detection is continuously carried out by continuously acquiring frame images, so that the effect of long-term detection is achieved; and the recheck between adjacent frames is favorable for improving the detection accuracy, and the false detection and false alarm phenomena can be effectively avoided.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a flowchart of a high-voltage wire foreign object detection method based on video images according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating a procedure of detecting all the alien materials in each of two adjacent frames of images with a time interval of m minutes according to an embodiment of the present application;
fig. 3 is a schematic view illustrating a determination of whether a detected foreign object is located on a high-voltage electric wire according to an embodiment of the present application;
fig. 4 is a block diagram of a high-voltage wire foreign matter detection system according to an embodiment of the present application.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
It will be understood that when an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element.
It will be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like, as used herein, refer to an orientation or positional relationship indicated in the drawings that is solely for the purpose of facilitating the description and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be considered as limiting the present application.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
The method and system for detecting foreign matters in high-voltage wires based on video images will now be described in detail with reference to the accompanying drawings.
Referring to fig. 1, a method for detecting foreign matters in a high-voltage wire based on a video image according to an embodiment of the present application includes the following steps:
step S1: acquiring a video image which needs to be subjected to high-voltage wire foreign matter detection and acquiring two adjacent frames of images of the video image at a time interval of m minutes;
step S2: the high voltage wire foreign matter detection is performed based on the two adjacent frames of images at the time interval of m minutes, and the step S2 includes:
step S21: detecting all foreign matters in each frame of image in two adjacent frames of images with time interval of m minutes;
step S22: judging whether each detected foreign matter in each frame image is positioned on the high-voltage wire or not;
step S23: carrying out foreign matter matching rechecking between two adjacent frames of images: judging whether each foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image, if so, giving an alarm, and if not, deleting the foreign matter on the high-voltage wire in the previous frame
Step S3: acquiring a next frame image at an interval of m minutes, re-forming the two adjacent frame images at an interval of m minutes with a subsequent frame image among the two adjacent frame images, continuing to perform high-voltage wire foreign object detection based on the two adjacent frame images at the interval of m minutes, and continuing to perform steps S21, S22 and S23 of step S2.
It is worth mentioning that the foreign matter may be caught on objects such as kites, balloons, plastics, fabrics, birds, etc. which are high voltage electric wires to cause a safety hazard.
In the step S1, performing the frame extraction operation at intervals of m minutes can reduce the amount of calculation, thereby saving resources, and performing the frame extraction operation at the same time can help to filter out flying or moving objects in advance to some extent, because the flying or moving objects do not stay on the high-voltage wires and generally do not affect the high-voltage wires. In one embodiment, the m minutes is 1 minute or 2 minutes.
With further reference to fig. 2, in the step S21, the step of detecting all the foreign objects in each of the two adjacent frames of images with the time interval of m minutes includes:
step S211: slicing each frame of image in two adjacent frames of images: equally slicing each frame image to obtain n grid images and recording the position of each grid image in the frame image;
step S212: carrying out batch foreign matter detection on each frame image corresponding to all the grid images to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image;
step S213: and restoring the detected foreign matters of each grid image and the coordinate frames corresponding to the foreign matters into the frame images corresponding to the grid images, and expanding each coordinate frame in each frame image to be k times of the original coordinate frame.
Further, in step S212, each frame of image is input to the foreign object detection model for performing a batch foreign object detection corresponding to all the grid images, and the step of obtaining the foreign objects framed by the coordinate frame in all the grid images of the frame of image includes:
step S2121: amplifying each grid image of the frame image to 4/3 times of the original grid image;
step S2122: sequentially superposing the n grid images after the amplification treatment along an X axis and a Y axis to form a group of image groups;
step S2123: and inputting the image group corresponding to each frame image into a foreign matter detection model to perform batch foreign matter detection to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image.
In step S21, on one hand, foreign matters staying on the high-voltage electric wire, which need to be detected, are generally processed in the overhead area, so that in the captured video of the overhead area, the foreign matters in each frame of image are relatively small targets compared with the overhead area, and detecting small targets in a large area increases the difficulty of target detection; therefore, each frame of image is sliced, and the sliced grid image is amplified, so that the detection efficiency of the small target is improved. On the other hand, each coordinate frame in each frame image is expanded to k times of the original coordinate frame, so that a part of the background around the coordinate frame of the foreign matter is included, and then whether each detected foreign matter in each frame image is located on the high-voltage wire or not is judged on the basis, so that whether the detected foreign matter is located on the high-voltage wire or not can be effectively ensured, because the phenomenon that the high-voltage wire is completely covered by the foreign matter in the grid image can exist when each coordinate frame in each frame image is not expanded to the original k times, the actually existing high-voltage wire can not be detected, the misjudgment condition can occur, and the foreign matter which is supposed to be located on the high-voltage wire is judged not to be located on the high-voltage wire. In one embodiment the value of k is 1.2 or 1.5.
In one embodiment, each frame of acquired image has a size of 1920 × 1080 and n is 16, that is, each frame of image is equally sliced to obtain 16 grid images with a size of 480 × 270, and each grid image is enlarged to obtain an image with a size of 640 × 360. In other embodiments, n may take the value 25, but is not limited to such.
In an embodiment, the foreign object detection model is obtained based on deep learning model training, which includes: acquiring training image data of plastic bags, kites, branches, balloons, fabrics and birds, and inputting the training image data into a yolov5 learning model to train to obtain the foreign matter detection model; the yolov5 learning model. In other embodiments, YOLO (You Only Look one: Unifield, Real-Time Object Detection) is the Object Detection model proposed by Joseph Redmon and Ali Farhadi et al in 2015, yolov5 is the fifth generation learning model of YOLO.
Referring further to fig. 3, in step S22, the determining whether each foreign object detected in each frame image is located on the high voltage line includes: acquiring a positive sample of trained foreign matters staying on the high-voltage wire and a negative sample of the foreign matters not staying on the high-voltage wire; inputting the positive sample and the negative sample into an efficientnet-b1 deep learning network model for training to obtain a recognition model of a classification result, wherein the classification result is foreign matters on a high-voltage wire and foreign matters on a non-high-voltage wire; and inputting the image of the expanded coordinate framed area in each frame image into the recognition model.
Further, in step S23, the step of determining whether each foreign object on the high-voltage electric wire of the previous frame image repeatedly appears on the subsequent frame image includes:
calculating the Intersection ratio of an IOU (Intersection over Unit) of a coordinate frame A of each foreign matter on the high-voltage wire in the previous frame of image and a coordinate frame B of any foreign matter in the next frame of image, wherein if the Intersection ratio of the IOU is more than or equal to 0.5, the foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image;
wherein the IOU intersection ratio is as follows: (A. andgate.B)/(A. U.B).
In step S23, if a foreign object detected in the previous frame of image is actually the object of moving or flying, the foreign object may be a foreign object located on the high-voltage wire in the previous frame of image, but after the foreign object moves, the foreign object is already far away from the high-voltage wire in the next frame of image, so performing the foreign object matching re-detection between the two adjacent frames of images is beneficial to improving the accuracy of the detection, and the false detection phenomenon can be effectively avoided.
In conclusion, in the video image-based high-voltage wire foreign matter detection method, the video image-based processing method is used for detecting the foreign matter on the high-voltage wire, so that the manual participation can be effectively reduced, the detection efficiency is improved, and the alarm can be given in time; meanwhile, foreign matter detection is continuously performed by continuously acquiring frame images, so that the effect of long-term detection is achieved; and the rechecking between adjacent frames is favorable for improving the detection accuracy, and the false detection and the false alarm can be effectively avoided.
Referring to fig. 4, the high-voltage wire foreign matter detection system 100 provided by the embodiment of the present application includes a collection device 20, a memory 30, a processor 10 and an alarm 40, wherein the processor 10 is connected to the memory 30, the collection device 20 and the alarm 40; the memory 30 is used for storing executable codes, the processor 10 is used for executing the executable codes stored in the memory 30 so as to execute the video image-based high-voltage wire foreign matter detection method, the acquisition device 20 is used for acquiring video images needing high-voltage wire foreign matter detection, and the alarm 40 is used for giving an alarm when the processor 10 detects foreign matters on a high-voltage wire.
In the present embodiment, the capturing device 20 is a camera for capturing video images, and the memory 30 may be, but is not limited to, an electronic, magnetic, optical or semiconductor system, device or device, and is specifically, but not limited to, a magnetic disk, a hard disk, a read only memory 30, a random access memory 30 or an erasable programmable read only memory 30. The Processor 10 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, and the processor 10 is the control center of the computer device and connects the various parts of the overall computer device using various interfaces and lines.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (10)
1. A high-voltage wire foreign matter detection method based on video images is characterized by comprising the following steps:
the method for detecting the foreign matters on the high-voltage wire comprises the following steps of acquiring a video image needing high-voltage wire foreign matter detection, acquiring two adjacent frames of images of the video image at a time interval of m minutes, and detecting the foreign matters on the high-voltage wire based on the two adjacent frames of images at the time interval of m minutes, wherein the method comprises the following steps:
detecting all foreign matters in each frame of image in two adjacent frames of images with the time interval of m minutes;
judging whether each detected foreign matter in each frame image is positioned on the high-voltage wire or not;
carrying out foreign matter matching rechecking between two adjacent frames of images: judging whether each foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image, if so, giving an alarm, and if not, deleting the foreign matter on the high-voltage wire in the previous frame; and
and acquiring a next frame image at an interval of m minutes, reforming the next frame image and a next frame image in the two adjacent frame images into two adjacent frame images at an interval of m minutes, and continuously detecting the foreign matters in the high-voltage wire on the basis of the two adjacent frame images at the interval of m minutes.
2. The video-image-based high-voltage electric wire foreign matter detection method according to claim 1, wherein the step of detecting all foreign matters in each of two adjacent frames of images at an interval of m minutes comprises:
slicing each frame of image in two adjacent frames of images: equally slicing each frame image to obtain n grid images and recording the position of each grid image in the frame image;
carrying out batch foreign matter detection on each frame image corresponding to all the grid images to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image;
restoring the detected foreign matters and the coordinate frames corresponding to the foreign matters of each grid image into the frame images corresponding to the grid images, and expanding each coordinate frame in each frame image to be k times of the original coordinate frame.
3. The method for detecting foreign matters in a high-voltage wire based on video images as claimed in claim 2, wherein each frame image corresponding to all grid images is input into a foreign matter detection model for batch foreign matter detection, and the step of obtaining the foreign matters framed by the coordinate frame in all the grid images of the frame image comprises:
amplifying each grid image of the frame image to 4/3 times of the original grid image;
sequentially superposing the n grid images after the amplification treatment along an X axis and a Y axis to form a group of image groups; and inputting the image group corresponding to each frame image into a foreign matter detection model to perform batch foreign matter detection to obtain foreign matters framed by the coordinate frame in all the grid images of the frame image.
4. The video-image-based high-voltage-wire foreign-object detection method according to claim 2, wherein the determining whether each foreign object detected in each frame image is located on the high-voltage wire is determining whether a foreign object in each enlarged coordinate frame in each frame image is located on the high-voltage wire, and includes:
acquiring a positive sample of trained foreign matters staying on the high-voltage wire and a negative sample of the foreign matters not staying on the high-voltage wire; inputting the positive sample and the negative sample into an efficientnet-b1 deep learning network model for training to obtain a recognition model of a classification result, wherein the classification result is foreign matters on a high-voltage wire and foreign matters on a non-high-voltage wire;
and inputting the image of the expanded coordinate framed area in each frame image into the recognition model.
5. The method according to claim 3, wherein the size of each frame of image is 1920 x 1080, each frame of image is equally sliced to obtain 16 grid images 480 x 270, and each grid image is enlarged to 640 x 360.
6. The video image-based foreign object detection method for high-voltage electric wires according to claim 3, wherein the foreign object detection model is obtained based on deep learning model training and comprises: acquiring training image data of plastic bags, kites, branches, balloons, fabrics and birds, and inputting the training image data into a yolov5 learning model to train to obtain the foreign matter detection model.
7. The video-image-based high-voltage electric wire foreign matter detection method according to claim 1, wherein the step of determining whether each foreign matter located on the high-voltage electric wire of the previous frame image repeatedly appears on the subsequent frame image comprises:
calculating the IOU intersection ratio of a coordinate frame A of each foreign matter on the high-voltage wire in the previous frame of image and a coordinate frame B of any foreign matter in the next frame of image, wherein if the IOU intersection ratio is more than or equal to 0.5, the foreign matter on the high-voltage wire in the previous frame of image repeatedly appears on the next frame of image;
wherein the IOU intersection ratio is as follows: (A. andgate.B)/(A. U.B).
8. The video-image-based high-voltage wire foreign matter detection method according to claim 1, wherein the m minutes is 1 minute or 2 minutes.
9. The video-image-based high-voltage wire foreign matter detection method according to claim 2, wherein the value of k is 1.2 or 1.5.
10. The high-voltage wire foreign matter detection system is characterized by comprising a collection device, a storage, a processor and an alarm, wherein the processor is connected with the storage, the collection device and the alarm; the memory is used for storing executable codes, the processor is used for executing the executable codes stored in the memory so as to execute the video image-based high-voltage wire foreign matter detection method according to any one of claims 1 to 9, the acquisition device is used for acquiring video images needing high-voltage wire foreign matter detection, and the alarm is used for giving an alarm when the processor detects foreign matters on a high-voltage wire.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210276445.7A CN114723678A (en) | 2022-03-21 | 2022-03-21 | High-voltage wire foreign matter detection method and detection system based on video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210276445.7A CN114723678A (en) | 2022-03-21 | 2022-03-21 | High-voltage wire foreign matter detection method and detection system based on video image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114723678A true CN114723678A (en) | 2022-07-08 |
Family
ID=82238678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210276445.7A Pending CN114723678A (en) | 2022-03-21 | 2022-03-21 | High-voltage wire foreign matter detection method and detection system based on video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114723678A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016413A (en) * | 2020-08-13 | 2020-12-01 | 南京领行科技股份有限公司 | Method and device for detecting abnormal behaviors between objects |
| CN113449769A (en) * | 2021-05-18 | 2021-09-28 | 内蒙古工业大学 | Power transmission line icing identification model training method, identification method and storage medium |
| CN113670929A (en) * | 2021-07-05 | 2021-11-19 | 国网宁夏电力有限公司电力科学研究院 | Power transmission line foreign matter detection method and device, storage medium and terminal equipment |
-
2022
- 2022-03-21 CN CN202210276445.7A patent/CN114723678A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016413A (en) * | 2020-08-13 | 2020-12-01 | 南京领行科技股份有限公司 | Method and device for detecting abnormal behaviors between objects |
| CN113449769A (en) * | 2021-05-18 | 2021-09-28 | 内蒙古工业大学 | Power transmission line icing identification model training method, identification method and storage medium |
| CN113670929A (en) * | 2021-07-05 | 2021-11-19 | 国网宁夏电力有限公司电力科学研究院 | Power transmission line foreign matter detection method and device, storage medium and terminal equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | An effective data augmentation strategy for CNN-based pest localization and recognition in the field | |
| CN111797890A (en) | Method and system for detecting defects of power transmission line equipment | |
| CN109672863A (en) | A kind of construction personnel's safety equipment intelligent monitoring method based on image recognition | |
| CN102915638A (en) | Surveillance video-based intelligent parking lot management system | |
| US8675917B2 (en) | Abandoned object recognition using pedestrian detection | |
| CN111274886B (en) | Deep learning-based pedestrian red light running illegal behavior analysis method and system | |
| CN111695493B (en) | Method and system for detecting hidden danger of power transmission line | |
| CN104657706A (en) | Image-based high-speed railway line pole breakage abnormality and connecting structural body abnormality detection method | |
| CN112541393A (en) | Transformer substation personnel detection method and device based on deep learning | |
| CN113298077A (en) | Transformer substation foreign matter identification and positioning method and device based on deep learning | |
| CN114863311A (en) | Automatic tracking method and system for inspection target of transformer substation robot | |
| Gao et al. | Adaptive deep learning for high-speed railway catenary swivel clevis defects detection | |
| CN114723678A (en) | High-voltage wire foreign matter detection method and detection system based on video image | |
| CN113916899B (en) | Method, system and device for detecting large transfusion soft bag product based on visual identification | |
| Zhang et al. | Fault detection of overhead contact systems based on multi-view Faster R-CNN | |
| US11954955B2 (en) | Method and system for collecting and monitoring vehicle status information | |
| CN116030412A (en) | Escalator monitoring video anomaly detection method and system | |
| Mansur et al. | Highway drivers drowsiness detection system model with r-pi and cnn technique | |
| CN114724091A (en) | Method and device for identifying foreign matters on transmission line wire | |
| CN111368726B (en) | Construction site operation face personnel number statistics method, system, storage medium and device | |
| Ge et al. | An automatic detection approach for wearing safety helmets on construction site based on YOLOv5 | |
| CN114243925B (en) | Intelligent fusion terminal-based distribution substation allergy sensing method and system | |
| CN110738692A (en) | spark cluster-based intelligent video identification method | |
| Yi et al. | Research on fruit recognition method based on improved yolov4 algorithm | |
| Giri et al. | Automated visual inspection of silicon detectors in CMS experiment |
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 |