CN110458176B - Foreign body intrusion detection method for laser foreign body cleaner - Google Patents
Foreign body intrusion detection method for laser foreign body cleaner Download PDFInfo
- Publication number
- CN110458176B CN110458176B CN201910625698.9A CN201910625698A CN110458176B CN 110458176 B CN110458176 B CN 110458176B CN 201910625698 A CN201910625698 A CN 201910625698A CN 110458176 B CN110458176 B CN 110458176B
- Authority
- CN
- China
- Prior art keywords
- image
- points
- laser
- matching
- foreign matter
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 24
- 230000005540 biological transmission Effects 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000003044 adaptive effect Effects 0.000 claims abstract description 9
- 230000000877 morphologic effect Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims abstract description 7
- 230000009545 invasion Effects 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000012217 deletion Methods 0.000 claims description 2
- 230000037430 deletion Effects 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims 2
- 241001465754 Metazoa Species 0.000 abstract description 7
- 238000012423 maintenance Methods 0.000 abstract description 4
- 238000011161 development Methods 0.000 abstract description 3
- 230000018109 developmental process Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 description 4
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010285 flame spraying Methods 0.000 description 1
- 239000002985 plastic film Substances 0.000 description 1
- 229920006255 plastic film Polymers 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
- Burglar Alarm Systems (AREA)
Abstract
A foreign body intrusion detection method for a laser foreign body remover, comprising: collecting an n-k frame image, an nth frame image and an n + k frame image according to a time sequence, and respectively compressing the images; image enhancement processing; extracting feature points by adopting an adaptive threshold FAST corner detection algorithm; matching the characteristics; deleting the mismatching points; calculating a transmission matrix to obtain a registration image; differentiating the images and carrying out threshold segmentation on the differential images; morphological treatment; performing bitwise AND operation to obtain a binary image of the suspicious invading foreign object target; and correcting the moving target of the binary image, finally obtaining a detection result of foreign matter invasion, sending a laser closing instruction and giving an alarm. According to the invention, the invasion foreign matter is automatically detected, and the laser closing instruction is automatically sent, so that the potential safety hazard in the working process of a laser foreign matter removing instrument is eliminated, the safety of people, animals, vehicles and the like is ensured, and the development requirement of automation and intelligence in the field of power grid maintenance is met.
Description
Technical Field
The invention relates to a graph processing method for maintaining a power transmission line network, in particular to a foreign matter intrusion detection method for a laser foreign matter cleaner.
Background
The transmission line network, referred to as the grid for short, is the main carrier of power transmission. In recent years, foreign matters such as balloons, advertisement banners, kites, agricultural plastic films, sunshade nets, kongming lights and the like are often wound around electric wires to cause alternate short circuits, which can cause power failure accidents and even cause personnel injuries and equipment faults in severe cases. Most of the lines are arranged in places with complex terrain and severe environment, so that the maintenance of the normal operation of the transmission line is particularly important and difficult.
The method for removing the foreign matters in the power grid comprises manual tower-climbing live-wire operation, unmanned aerial vehicle flaming operation, laser foreign matter removing operation and the like. The manual operation has low working efficiency, potential safety hazards and serious environmental influence; the flame spraying operation of the unmanned aerial vehicle is limited by the load and cannot work continuously for a long time.
The laser foreign matter removing method is a method for rapidly removing foreign matters by adopting remote laser, and a laser beam with proper energy is irradiated on the foreign matters of a power transmission line at a longer distance, so that the parts wound on a lead are burnt under the condition of ensuring safety, and the foreign matters fall. However, the energy of the laser beam is large, and certain safety hazards exist, for example, when the laser is turned on, walking people, animals or running vehicles suddenly appear in the visual field of the camera.
Therefore, foreign matters intruding suddenly need to be detected, when the invasion of the foreign matters is detected (the foreign matters suddenly appear in the radiation range of the laser), the laser needs to be automatically turned off, the potential safety hazard in the working process of removing the foreign matters by the laser is eliminated, and the safety of people, animals and vehicles is guaranteed.
Disclosure of Invention
The invention aims to provide a foreign matter intrusion detection method for a laser foreign matter cleaner, which can automatically identify foreign matters entering the visual field of a camera, send out a laser closing instruction and give an alarm, realize the function of automatically detecting the invading foreign matters and ensure the safety of removing the laser foreign matters.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a foreign matter intrusion detection method for a laser foreign matter cleaner is characterized in that after preparation work of the laser foreign matter cleaner is completed, a foreign matter intrusion detection function is started before a laser is turned on to start foreign matter cleaning, and the method comprises the following steps:
(1) collecting target area images through a laser foreign matter remover, collecting an n-k frame image, an n frame image and an n + k frame image according to time sequence, and respectively compressing the images to obtain a compressed image P t+1 、P t+2 、P t+3 The purpose of image compression is to reduce the calculation amount, improve the calculation efficiency and meet the real-time requirement;
(2) for compressed image P t+1 、P t+2 、P t+3 Respectively processed by an image enhancement algorithm to obtain an enhanced image S t+1 、S t+2 、S t+3 The purpose of image enhancement is to increase the contrast of a target and a background and facilitate the detection of the following features;
(3) for the enhanced image S t+1 、S t+2 、S t+3 Respectively extracting characteristic points by adopting a FAST corner detection algorithm of a self-adaptive threshold value to obtain K1, K2 and K3;
(4) constructing a feature point description operator, describing feature points K1, K2 and K3 as D1, D2 and D3, and performing feature matching on the described feature points D1, D2 and D3 to obtain M12 and M23;
(5) deleting the mismatching points to obtain deleted matching points Md12 and Md23;
(6) respectively generating transmission transformation matrixes according to the deleted matching points Md12 and Md23 to obtain registration images H12 and H23;
(7) registering image H12 and image P t+2 Performing image difference, and registering image H23 and image P t+3 Carrying out image difference, and respectively carrying out threshold segmentation on the difference images to obtain segmented images S12 and S23;
(8) performing morphological processing on the segmented image to obtain processed images O12 and O23;
(9) performing bitwise AND operation on the images O12 and O23 after the morphological processing to obtain a binary image T of the suspicious moving target;
and (3) performing moving object correction on the binary image at the R (R), namely performing pixel constraint on the binary image at first, then performing linear detection, and if no linear exists, directly skipping to the step
If the straight line exists, removing the pixels participating in the straight line fitting part to obtain a binary image after the straight line is removed;
Further, in the step (3), the implementation method of Threshold adaptive selection of the FAST corner detection algorithm is as shown in formula (1):
wherein: threshold is a fast detection Threshold, mGrey is the average gray level of the image, std is the standard deviation of the image, and the Threshold selection mode of the formula (1) is used for respectively selecting the enhanced image S t+1 、S t+2 、S t+3 Calculating corresponding adaptive threshold, and applying the corresponding adaptive threshold to the image S t+1 、S t+2 、S t+3 And carrying out corner detection to further obtain characteristic points K1, K2 and K3.
Further, the step (5) deletes the mismatching points, specifically as shown in step 5.1 and step 5.2:
step 5.1, deleting the mismatching points roughly: calculating the maximum value max _ dist of the distance in the matching result, reserving the matching points with the distance smaller than 0.6 max _dist in the matching result, and deleting the matching points with the distance larger than 0.6 max _dist in the matching result to obtain the roughly screened matching points;
step 5.2, deleting the mismatching points in detail: and (3) calculating a basic matrix for the matching points obtained in the step (5.1) by adopting a RANSAC algorithm, finely deleting the matching points obtained in the step (5.1) according to the state (0 or 1,0 represents the rejected matching points, and 1 represents the reserved matching points) of each marked matching point when the basic matrix is generated, and screening the matching points with the marked state of 1, namely the finely screened matching points.
Further, in step r, the moving object correction is performed on the binary image, where the pixel constraint is to eliminate false detection caused by branch shaking, and the main implementation method is as follows:
(1) Detecting a connected region of the binary image;
(2) Primary screening: presetting a threshold value T 1 The area is smaller than the threshold value T 1 Removing the connected region to obtain a sequence set P with the small-area connected region removed;
(3) Sorting the set obtained by the processing in the step (2) according to the sizes of the connected regions, and selecting the median of the area sequence as a threshold T 2 ；
(4) Making the area smaller than a threshold value T 2 Removing the connected region to obtain the image after the pixel restriction.
Further, the step r corrects the moving object of the binary image, wherein the line detection is to eliminate the false detection caused by the deviation of the power transmission line in the image registration process, and the implementation method is to perform line detection on the binary image after the pixels are restricted, and then delete the pixels participating in the line fitting part to obtain the binary image after the lines are removed.
Compared with the prior art, the invention has the beneficial effects that:
through the intrusion foreign matter real-time detection, the laser closing instruction is automatically sent, the potential safety hazard in the working process of the laser foreign matter removing instrument is eliminated, the safety of people, animals and vehicles is ensured, and the development requirement of automation and intelligence in the field of power grid maintenance is met.
Drawings
Fig. 1 is a flowchart of a foreign object intrusion detection method for a laser foreign object remover according to the present invention.
Detailed Description
The invention is further described below with reference to the following figures and examples.
Referring to fig. 1, fig. 1 is a flowchart illustrating a foreign object intrusion detection method for a laser foreign object remover according to the present invention. It can be seen from the figure that the foreign body intrusion detection method for the laser foreign body cleaner of the present invention starts the foreign body intrusion detection after the preparation work of the laser foreign body cleaner is completed and before the laser is turned on to start the foreign body cleaning, and comprises the following steps:
1. collecting target area images through a laser foreign matter cleaner, collecting an n-k frame image, an n frame image and an n + k frame image according to time sequence, and respectively compressing the length and the width of the images by 0.5 times to obtain a compressed image P t+1 、P t+2 、P t+3 Assuming the original image size is 1280 x 960, the compressed image P t+1 The size of (2) is 640 x 480, wherein the purpose of image compression is to reduce the amount of calculation, improve the calculation efficiency and meet the real-time requirement;
2. for compressed image P t+1 、P t+2 、P t+3 Respectively processed by an image enhancement algorithm to obtain an enhanced image S t+1 、S t+2 、S t+3 (ii) a The purpose of image enhancement is to increase the contrast between the target and the background and facilitate the detection of the following features, wherein the image enhancement algorithm is realized by adopting a homomorphic filtering algorithm.
3. For the enhanced image S t+1 、S t+2 、S t+3 Extracting feature points by using a FAST corner detection algorithm of a self-adaptive threshold value respectively to obtain K1, K2 and K3; the implementation method of Threshold self-adaptive selection is shown as formula (1):
wherein: threshold is the fast detection Threshold, mcley is the average gray level of the image, and Std is the standard deviation of the image.
Respectively selecting the enhanced images S by the threshold selection mode of the formula (1) t+1 、S t+2 、S t+3 Calculating corresponding adaptive threshold, and applying the corresponding adaptive threshold to the image S t+1 、S t+2 、S t+3 And detecting corner points to obtain characteristic points K1, K2 and K3.
4. Constructing surf feature point description operators, describing the feature points K1, K2 and K3 as D1, D2 and D3, and performing feature matching on the described feature points D1, D2 and D3 by adopting a BruteForce algorithm to obtain M12 and M23.
5. Deleting the mismatching points to obtain deleted matching points Md12 and Md23; the specific steps for deleting the mismatching points are as follows:
1) Rough deletion of mismatching points: calculating the maximum value max _ dist of the distance in the matching result, reserving the matching points with the distance smaller than 0.6 max _dist in the matching result, and deleting the matching points with the distance larger than 0.6 max _dist in the matching result to obtain the roughly screened matching points;
2) And deleting mismatching points in detail: and (3) calculating a basic matrix for the matching points obtained in the step (1) by using a RANSAC algorithm, and screening the matching points marked with the state of 1 according to the state (0 or 1,0 represents the rejected matching points and 1 represents the reserved matching points) of each marked matching point when the basic matrix is generated, namely the carefully screened matching points.
6. Respectively generating transmission transformation matrixes according to the deleted matching points Md12 and Md23 to obtain registration images H12 and H23;
7. registering images H12 and P t+2 Performing image difference, and registering image H23 and image P t+3 Carrying out image difference, and respectively carrying out OTSU self-adaptive threshold segmentation on the difference image to obtain segmented images S12 and S23;
8. performing morphological processing on the segmented image to obtain processed images O12 and O23;
9. performing bitwise AND operation on the images O12 and O23 after the morphological processing to obtain a binary image T of the suspicious moving target;
10. the method comprises the following steps of correcting a moving target of a binary image, namely firstly carrying out pixel constraint on the binary image and then carrying out linear detection, wherein the pixel constraint is to eliminate false detection caused by treeing shaking, the method is to carry out connected region detection on the binary image and remove a part of the connected region with the area smaller than a threshold value, and the specific steps are as follows:
(1) Detecting a connected region of the binary image;
(2) Primary screening: presetting a threshold value T 1 =50, area less than threshold T 1 Removing the connected region to obtain a sequence set P with the small-area connected region removed;
(3) Sorting the sets obtained by the processing in the step (2) according to the sizes of the areas of the connected regions, and selecting the median of the area sequence as a threshold T 2 ；
(4) Make the area less than the threshold T 2 Removing the connected region to obtain the image after the pixel restriction.
The straight line detection is to eliminate the false detection caused by the deviation of the power transmission line in the image registration process, and the realization method is to perform the straight line detection on the binary image after the pixel constraint, and then delete the pixels participating in the straight line fitting part to obtain the binary image after the straight line is removed.
11. Obtaining a moving object detection result, namely a foreign object intrusion detection result, wherein the intruded pedestrians, animals and vehicles are collectively called as a moving object, the moving object is represented in a binary image in a pixel point form with a gray value of 1, when pixels with the gray value of 1 do not exist in the binary image, the moving object is represented without intruding foreign objects, and the laser foreign object remover keeps the current working state; and when a pixel point with the gray value of 1 exists in the binary image, the intrusion foreign matter is represented, and the controller of the laser foreign matter cleaner sends a laser closing instruction and sends a safety warning.
According to the invention, the invasion foreign matter is detected in real time, and the laser closing instruction is automatically sent, so that the potential safety hazard in the working process of the laser foreign matter removing instrument is eliminated, the safety of people, animals and vehicles is ensured, the working efficiency is improved, and the development requirement of automation and intelligence in the field of power grid maintenance is met.
Claims (4)
1. A foreign body intrusion detection method for a laser foreign body remover is characterized by comprising the following steps:
(1) collecting target area images by using a laser foreign matter remover, collecting an n-k frame image, an nth frame image and an n + k frame image according to time sequence, and respectively compressing the images to obtain a compressed image P t+1 、P t+2 、P t+3 ；
(2) For compressed image P t+1 、P t+2 、P t+3 Respectively processed by an image enhancement algorithm to obtain an enhanced image S t+1 、S t+2 、S t+3 ；
(3) For the enhanced image S t+1 、S t+2 、S t+3 Extracting characteristic points K1, K2 and K3 by adopting a FAST corner detection algorithm of a self-adaptive threshold value respectively;
(4) constructing a feature point description operator, describing feature points K1, K2 and K3 as D1, D2 and D3, and performing feature matching on the described feature points D1, D2 and D3 to obtain M12 and M23;
(5) deleting mismatching points in M12 and M23 to obtain deleted matching feature points Md12 and Md23;
(6) respectively generating transmission transformation matrixes according to the deleted matched feature points Md12 and Md23 to obtain registration images H12 and H23;
(7) registering image H12 and image P t+2 Performing image difference, and registering image H23 and image P t+3 Carrying out image difference, and respectively carrying out threshold segmentation on the difference images to obtain segmented images S12 and S23;
(8) performing morphological processing on the segmented images S12 and S23 to obtain images O12 and O23 after the morphological processing;
(9) performing bitwise AND operation on the images O12 and O23 after morphological processing to obtain a binary image T of the suspicious moving object;
the method comprises that the binary image T of the suspicious moving object is corrected by the object (R), namely, the binary image is firstly imagedElement constraint, then straight line detection is carried out, if no straight line exists, the step is directly skipped to
If the straight line exists, removing the pixels participating in the straight line fitting part to obtain a binary image after the straight line is removed;
2. The method of claim 1, wherein the Threshold adaptive FAST corner detection algorithm is implemented by adaptively selecting the Threshold according to the following steps:
wherein: threshold is a fast detection Threshold, mGrey is the average gray level of the image, std is the standard deviation of the image, and the Threshold selection mode of the formula (1) is used for respectively selecting the enhanced image S t+1 、S t+2 、S t+3 Calculating corresponding adaptive threshold, and applying the corresponding adaptive threshold to the image S t+1 、S t+2 、S t+3 And performing Fast corner detection to further obtain characteristic points K1, K2 and K3.
3. The method for detecting intrusion of a foreign substance into a laser foreign substance remover according to claim 1, wherein said algorithm for deleting the mismatching point of step (5) comprises the steps of:
step 3.1, rough deletion of mismatching points:
calculating the maximum value max _ dist of the distance in the matching result, reserving the matching points with the distance smaller than 0.6 max _dist in the matching result, and deleting the matching points with the distance larger than 0.6 max _dist in the matching result to obtain the roughly screened matching points;
step 3.2, deleting the mismatching points in detail:
and (3) calculating a basic matrix for the matching points obtained in the step (3.1) by adopting a RANSAC algorithm, and screening the matching points marked with the state of 1 according to the state 0 or 1,0 of each marked matching point in the generation of the basic matrix to represent the rejected matching points and 1 to represent the reserved matching points, so that the carefully screened matching points are obtained.
4. The method for detecting the invasion of the foreign matters for the laser foreign matter remover according to claim 1, wherein the pixel constraint is to eliminate the false detection caused by the treeing shake, and the realization method is to detect the connected region of the binary image and remove the part of the connected region with the area smaller than the threshold value; the straight line detection is to eliminate false detection caused by deviation of the power transmission line in the image registration process, and the implementation method is to perform straight line detection on the binary image after the pixels are constrained, and then delete the pixels participating in the straight line fitting part to obtain the binary image after the straight line is removed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625698.9A CN110458176B (en) | 2019-07-11 | 2019-07-11 | Foreign body intrusion detection method for laser foreign body cleaner |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625698.9A CN110458176B (en) | 2019-07-11 | 2019-07-11 | Foreign body intrusion detection method for laser foreign body cleaner |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110458176A CN110458176A (en) | 2019-11-15 |
| CN110458176B true CN110458176B (en) | 2022-11-04 |
Family
ID=68482702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910625698.9A Active CN110458176B (en) | 2019-07-11 | 2019-07-11 | Foreign body intrusion detection method for laser foreign body cleaner |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110458176B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114371179B (en) * | 2021-12-29 | 2023-12-26 | 武汉科技大学 | Online detection and removal system, method and device for foreign matters on surface of fine blanking die |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013104783A (en) * | 2011-11-14 | 2013-05-30 | Mitsubishi Electric Corp | Foreign substance detection device |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
| CN109191438A (en) * | 2018-08-17 | 2019-01-11 | 中科光绘（上海）科技有限公司 | A kind of method for recognizing impurities for laser foreign matter remover |
-
2019
- 2019-07-11 CN CN201910625698.9A patent/CN110458176B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013104783A (en) * | 2011-11-14 | 2013-05-30 | Mitsubishi Electric Corp | Foreign substance detection device |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
| CN109191438A (en) * | 2018-08-17 | 2019-01-11 | 中科光绘（上海）科技有限公司 | A kind of method for recognizing impurities for laser foreign matter remover |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110458176A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dhara et al. | Color cast dependent image dehazing via adaptive airlight refinement and non-linear color balancing | |
| CN107679495B (en) | Detection method for movable engineering vehicles around power transmission line | |
| CN104796582B (en) | Video image denoising and Enhancement Method and device based on random injection retinex | |
| CN110458176B (en) | Foreign body intrusion detection method for laser foreign body cleaner | |
| Kim et al. | Adaptive patch based convolutional neural network for robust dehazing | |
| CN111582074A (en) | Monitoring video leaf occlusion detection method based on scene depth information perception | |
| CN107481214A (en) | A kind of twilight image and infrared image fusion method | |
| CN114881869A (en) | Inspection video image preprocessing method | |
| CN115660998A (en) | Image defogging method based on deep learning and traditional priori knowledge fusion | |
| KR101534646B1 (en) | Video processing apparatus and method for removing fog from video | |
| CN108921147B (en) | Black smoke vehicle identification method based on dynamic texture and transform domain space-time characteristics | |
| CN111402250A (en) | Machine vision defect detection method and platform based on edge calculation | |
| CN108122209B (en) | License plate deblurring method based on countermeasure generation network | |
| CN105608683A (en) | Defogging method of single image | |
| Jia et al. | A wavelet-based approach to improve foggy image clarity | |
| CN111161196A (en) | Adaptive enhancement method for aerial image of power transmission line | |
| CN109753932A (en) | Forest fire smoke detection method based on Wavelet transformation image enhancement and multiple features | |
| CN115641431A (en) | Power transmission line engineering vehicle detection method based on improved deep learning | |
| Noh et al. | Haze removal on superpixel domain | |
| CN113450289A (en) | Method for automatically enhancing low illumination of face image in passenger traffic scene | |
| Yu et al. | Haze removal algorithm using color attenuation prior and guided filter | |
| Kansal et al. | Fusion based fast de-fogging for foggy images | |
| CN110796610A (en) | Image defogging method, device and equipment and storage medium | |
| Joy et al. | Deep Multi-scale Residual Learning for Single Image De-Hazing | |
| Zhang et al. | Temperature-based collision detection in extreme low light condition with bio-inspired LGMD neural network |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231129 Address after: Room 146, Building 6, Chuangzhi Green Valley Development Center, No. 788 Hongpu Road, Shangcheng District, Hangzhou City, Zhejiang Province, 310000 Patentee after: Xianheng Electrical Technology (Hangzhou) Co.,Ltd. Address before: 201800 room 1102, 811 Pingcheng Road, Juyuan New District, Jiading District, Shanghai Patentee before: ZHONGKE GUANGHUI (SHANGHAI) TECHNOLOGY CO.,LTD. |