CN112700403A - Charged interval false-entry prevention detection method, medium and system - Google Patents
Charged interval false-entry prevention detection method, medium and system Download PDFInfo
- Publication number
- CN112700403A CN112700403A CN202011429994.0A CN202011429994A CN112700403A CN 112700403 A CN112700403 A CN 112700403A CN 202011429994 A CN202011429994 A CN 202011429994A CN 112700403 A CN112700403 A CN 112700403A
- Authority
- CN
- China
- Prior art keywords
- eye image
- left eye
- moving target
- points
- detection method
- 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 46
- 230000002265 prevention Effects 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 11
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 2
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005316 response function Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T7/248—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a charged interval false-entry prevention detection method, medium and system. The method comprises the following steps: arranging a binocular camera around a charged interval, and acquiring a left eye image and a right eye image around the charged interval in real time by adopting the binocular camera; tracking angular points in the left eye image in real time, and determining the number of the angular points with changed positions; if the number of the corner points with changed positions reaches a set threshold value, determining that a moving target appears; determining the area where the moving target is located; matching the feature points of the detected left eye image and the right eye image; calculating the depth of the characteristic point according to the horizontal coordinate of the characteristic point which is matched with the left eye image and the right eye image and is positioned in the area where the moving target is positioned; calculating the average value of the depths of all the characteristic points of the area where the moving target is located to obtain the distance between the moving target and the charged interval; and if the distance is smaller than the preset safe distance, marking the moving target. The invention has stable and objective detection and is not easy to miss detection or false detection.
Description
Technical Field
The invention relates to the technical field of charged interval safety, in particular to a charged interval false-entry prevention detection method, medium and system.
Background
In the present society, the rapid development of various technologies cannot be supported by electric power, and meanwhile, the pressure on electric power equipment is increased, so that the maintenance of the electric power equipment is particularly important. In recent years, casualty accidents caused by mistaken entry into a live interval in a safety accident of a power system occur frequently, and due to the fact that safety awareness of operators is low, no alarm prompting device is arranged before a person mistakenly enters the live interval, and the vigilance cannot be improved timely. The false entry prevention detection system can timely detect moving targets around the electrified interval and timely send out alarm sound to remind the targets of being far away.
Electrified interval prevents mistake and goes into detecting system belongs to this technical field of electric power safety, and traditional prevent mistake and go into system and expose all sorts of drawbacks and defects in actual operation: for example, the sensor device is excessively dependent, the sensitivity of the sensor device directly affects the detection result, the situation of false detection or detection omission is easy to occur, and the situation of uneven hardware quality and the like all affect the power safety.
Disclosure of Invention
The embodiment of the invention provides a charged interval false-entry prevention detection method, medium and system, and aims to solve the problem that false detection or detection missing is easy to occur due to the fact that the prior art excessively depends on sensor equipment.
In a first aspect, a method for detecting a false access prevention in a live interval is provided, which includes: arranging a binocular camera around a charged interval, and acquiring a left eye image and a right eye image around the charged interval in real time by adopting the binocular camera, wherein the left eye image is acquired by a left eye camera in the binocular camera, and the right eye image is acquired by a right eye camera in the binocular camera; tracking angular points in the left eye image in real time, and determining the number of angular points with changed positions; if the number of the corner points with the changed positions reaches a set threshold value, determining that a moving target appears; triangulation is carried out on the detected angular points, and the area where the moving target is located is determined; respectively carrying out feature point detection on the left eye image and the right eye image, and matching feature points of the left eye image and the right eye image obtained by detection; calculating the depth of the characteristic point according to the horizontal coordinate of the characteristic point which is matched with the left eye image and the right eye image and is positioned in the area where the moving target is positioned; calculating the average value of the depths of all characteristic points of the area where the moving target is located to obtain the distance between the moving target and the charged interval; and if the distance is smaller than a preset safe distance, marking the moving target.
In a second aspect, a computer-readable storage medium having computer program instructions stored thereon is provided; the computer program instructions, when executed by a processor, implement the electrified interval false-entry-prevention detection method as described in the first aspect of the embodiments above.
In a third aspect, a live interval false-entry-prevention detection system is provided, which includes: a computer readable storage medium as described in the second aspect of the embodiments above.
Therefore, the embodiment of the invention can detect the moving target around the electrified interval so as to find the moving target close to the electrified interval in time and record and warn the moving target, has higher automation degree and higher accuracy of detection, is not easy to have the condition of false detection or missing detection, can stably and objectively detect by depending on sensor equipment, and has great promotion effect on the development of the field of electric power safety.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a flow chart of a method for detecting a false entry into a live interval according to a preferred embodiment of the present invention;
fig. 2 is a flowchart of a method for detecting a false entrance of a charged interval according to another preferred 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 some, not all, embodiments of the present invention. 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 discloses a charged interval false-entry-prevention detection method. As shown in fig. 1, the detection method includes the following steps:
step S1: the binocular camera is arranged around the charged interval, and the left eye image and the right eye image around the charged interval are collected in real time by the binocular camera.
Wherein, the left eye image is collected by a left eye camera in the binocular cameras. The right eye image is captured by a right eye camera of the binocular cameras. In addition, the specific position at which the binocular camera is disposed around the charged interval may be determined according to actual circumstances.
Step S2: and tracking angular points in the left eye image in real time, and determining the number of angular points with changed positions.
Specifically, the Shi-Tomasi corner detection method may be used to detect the corners in the left eye image in advance. Wherein, the coordinates of the pixel points in the left eye image are (x, y). The corresponding gray value is I (x, y). The corner response function is R. The amount of change in pixel value resulting from the window moving on the image is E (u, v), and the amount of movement is (u, v), and this can be obtainedω (x, y) is a window function at position (x, y) representing the weight of each pixel within the window, IxAnd IyIs the partial differential of I. Lambda [ alpha ]1And λ2Is the eigenvalue of the matrix M. By the formula R ═ min (λ)1,λ2) Selecting a corner point, if the value of R is larger than a set threshold value, considering that the corner point is a cornerAnd (4) point.
When a moving object is present, the position of the partially tracked corner points in successive image frames changes. The method for tracking the corner points in the left eye image in real time is an LK optical flow method.
Step S3: and if the number of the corner points with the changed positions reaches a set threshold value, determining that the moving target appears.
Wherein the set threshold may be set empirically.
Step S4: and triangulating the detected angular points, and determining the area where the moving target is located.
Triangulation algorithms have the property of describing the outer contour of a moving object. Specifically, triangulation is performed on the angular points, every three angular points form a specific triangle in a group, and finally an irregular graphic area formed by combining all the triangles is the outline of the moving target, so that the position of the moving target on the left eye image is obtained.
Step S5: and respectively carrying out characteristic point detection on the left eye image and the right eye image, and matching the characteristic points of the left eye image and the right eye image obtained by detection.
Wherein, the detection algorithm of the feature points is an ORB algorithm. The matching algorithm of the feature points is a fast nearest neighbor matching algorithm.
Step S6: and calculating the depth of the characteristic point according to the horizontal coordinate of the characteristic point which is matched with the left eye image and the right eye image and is positioned in the area where the moving target is positioned.
Specifically, the depth of the feature point is calculated by the following formula:
wherein Z represents the depth of the feature point, T represents the base line distance of the binocular camera, f represents the focal length of the binocular camera, and xlHorizontal coordinate, x, representing a feature point in a left eye imagerHorizontal coordinates representing feature points in the right eye image. The base line distance and the focal length can be obtained through pre-calibration and factory parameters.
Step S7: and calculating the average value of the depths of all the characteristic points of the area where the moving target is located to obtain the distance between the moving target and the charged interval.
Because the area of the moving target formed after triangulation describes the contour information of the target, the feature points in the area are selected, so that the interference of the feature points on the surrounding environment can be reduced, and the precision in distance measurement is improved. Because the binocular camera is arranged around the charged interval and images around the interval are acquired, the calculated depth average is the distance from the moving object to the binocular camera, i.e., the distance between the moving object and the charged interval.
Step S8: and if the distance is smaller than the preset safe distance, marking the moving target.
The preset safe distance can adopt an empirical value, the positions of the binocular cameras are different, and the safe distance is correspondingly changed. If the moving object is a human, marking the moving object can make sure which worker is specific, and remind the worker of secondary attention. And if the moving target is non-human, marking the moving target so as to research protective measures according to actual conditions.
The charged interval false-entry-prevention detection method provided by the embodiment of the invention can also automatically detect the type of the moving target. Specifically, as shown in fig. 2, the method for detecting a false entry of a charged interval according to an embodiment of the present invention may further include the following steps:
step S1: the binocular camera is arranged around the charged interval, and the left eye image and the right eye image around the charged interval are collected in real time by the binocular camera.
Step S2: and tracking angular points in the left eye image in real time, and determining the number of angular points with changed positions.
Step S3: and if the number of the corner points with the changed positions reaches a set threshold value, determining that the moving target appears.
Steps S1-S3 are the same as steps S1-S3, and are not repeated herein.
Step S4: and inputting the left eye image into a neural network, and outputting the category of the moving target in the left eye image.
The neural network is a YOLOv3 neural network. The loss function loss used by the YOLOv3 neural network is as follows:
wherein the content of the first and second substances,andthe control function is expressed, and the pixel point value of the object is defined as 1, and the pixel point value of the object is 0 if the object is not contained.The loss function is represented. S2Indicating the number of meshes into which the picture is divided. i represents the current number of grids at the time of summation. B represents the number of prediction boxes per cell. j denotes a current cell prediction box. The center point of the prediction box is (x, y). The width and height of the prediction box are (ω, h). The confidence of the single prediction box is C. The class probability of an object is p. c represents the category of the object. Lambda [ alpha ]coordAnd λnoobjFor controlling the weight of the pixel containing the object.
Step S5: and alarming according to the category of the moving object.
Specifically, if the moving object is a human, a reminding alarm is given to remind the human to notice that the human enters a dangerous area; if the moving object is a non-human, an expulsion alarm is given, the sound of the expulsion alarm is sharp so as to expel animals and the like, and in addition, field personnel can be reminded to help the expulsion.
The embodiment of the invention also discloses a computer readable storage medium, which stores computer program instructions; the computer program instructions, when executed by a processor, implement the electrified interval false-entry-prevention detection method as described in the above embodiments.
The embodiment of the invention also discloses a charged interval false-entry-prevention detection system, which comprises: a computer readable storage medium as in the above embodiments.
In summary, the embodiment of the invention can detect the moving target around the charged interval so as to find the moving target close to the charged interval in time and record and warn the moving target, has high detection automation degree and high accuracy, is not easy to have false detection or missing detection, can stably and objectively detect by depending on sensor equipment, and has great promotion effect on the development of the field of electric power safety.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method for detecting false entry in a live interval is characterized by comprising the following steps:
arranging a binocular camera around a charged interval, and acquiring a left eye image and a right eye image around the charged interval in real time by adopting the binocular camera, wherein the left eye image is acquired by a left eye camera in the binocular camera, and the right eye image is acquired by a right eye camera in the binocular camera;
tracking angular points in the left eye image in real time, and determining the number of angular points with changed positions;
if the number of the corner points with the changed positions reaches a set threshold value, determining that a moving target appears;
triangulation is carried out on the detected angular points, and the area where the moving target is located is determined;
respectively carrying out feature point detection on the left eye image and the right eye image, and matching feature points of the left eye image and the right eye image obtained by detection;
calculating the depth of the characteristic point according to the horizontal coordinate of the characteristic point which is matched with the left eye image and the right eye image and is positioned in the area where the moving target is positioned;
calculating the average value of the depths of all characteristic points of the area where the moving target is located to obtain the distance between the moving target and the charged interval;
and if the distance is smaller than a preset safe distance, marking the moving target.
2. The charged interval false entry prevention detection method according to claim 1, wherein before the step of tracking and detecting the corner points in the left eye image in real time, the detection method further comprises:
and detecting the corner points in the left eye image by adopting a Shi-Tomasi corner point detection method.
3. The charged interval false entry prevention detection method according to claim 1, wherein the depth of the feature point is calculated by the equation:
wherein Z represents the depth of the feature point, T represents the base line distance of the binocular camera, f represents the focal length of the binocular camera, and xlHorizontal coordinate, x, representing a feature point in the left eye imagerHorizontal coordinates representing feature points in the right eye image.
4. The charged interval false-entry-prevention detection method according to claim 1, characterized in that: the method for tracking the corner points in the left eye image in real time is an LK optical flow method.
5. The charged interval false-entry-prevention detection method according to claim 1, characterized in that: the detection algorithm of the feature points is an ORB algorithm, and the matching algorithm of the feature points is a rapid nearest neighbor matching algorithm.
6. The charged interval false entry prevention detection method according to claim 1, wherein after the step of determining the area where the moving object is located, the method further comprises:
inputting the left eye image into a neural network, and outputting the category of the moving target in the left eye image;
and alarming according to the category of the moving target.
7. The charged interval false-entry-prevention detection method according to claim 6, wherein the step of alarming according to the category of the moving object includes:
if the moving target is a human, sending out a reminding alarm;
issuing an eviction alert if the moving object is non-human.
8. The charged interval false-entry-prevention detection method according to claim 6, wherein the neural network is a YOLOv3 neural network.
9. A computer-readable storage medium characterized by: the computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement a live interval false entry detection method as claimed in any one of claims 1 to 8.
10. The utility model provides a live interval prevents mistake income detecting system which characterized in that includes: the computer-readable storage medium of claim 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011429994.0A CN112700403A (en) | 2020-12-09 | 2020-12-09 | Charged interval false-entry prevention detection method, medium and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011429994.0A CN112700403A (en) | 2020-12-09 | 2020-12-09 | Charged interval false-entry prevention detection method, medium and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112700403A true CN112700403A (en) | 2021-04-23 |
Family
ID=75505640
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011429994.0A Pending CN112700403A (en) | 2020-12-09 | 2020-12-09 | Charged interval false-entry prevention detection method, medium and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112700403A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103426267A (en) * | 2013-03-06 | 2013-12-04 | 珠海电力设计院有限公司 | Video safety fence system based on video analysis technology |
CN104994334A (en) * | 2015-06-09 | 2015-10-21 | 海南电网有限责任公司 | Automatic substation monitoring method based on real-time video |
CN105915846A (en) * | 2016-04-26 | 2016-08-31 | 成都通甲优博科技有限责任公司 | Monocular and binocular multiplexed invading object monitoring method and system |
CN108151750A (en) * | 2017-12-13 | 2018-06-12 | 西华大学 | Positioning method and device |
CN111753712A (en) * | 2020-06-22 | 2020-10-09 | 中国电力科学研究院有限公司 | Method, system and equipment for monitoring safety of power production personnel |
-
2020
- 2020-12-09 CN CN202011429994.0A patent/CN112700403A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103426267A (en) * | 2013-03-06 | 2013-12-04 | 珠海电力设计院有限公司 | Video safety fence system based on video analysis technology |
CN104994334A (en) * | 2015-06-09 | 2015-10-21 | 海南电网有限责任公司 | Automatic substation monitoring method based on real-time video |
CN105915846A (en) * | 2016-04-26 | 2016-08-31 | 成都通甲优博科技有限责任公司 | Monocular and binocular multiplexed invading object monitoring method and system |
CN108151750A (en) * | 2017-12-13 | 2018-06-12 | 西华大学 | Positioning method and device |
CN111753712A (en) * | 2020-06-22 | 2020-10-09 | 中国电力科学研究院有限公司 | Method, system and equipment for monitoring safety of power production personnel |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113255481B (en) | Crowd state detection method based on unmanned patrol car | |
US20140119608A1 (en) | System and Methods for Improving Accuracy and Robustness of Abnormal Behavior Detection | |
US11917333B2 (en) | Systems and methods for personnel location at a drilling site | |
CN104821056A (en) | Intelligent guarding method based on radar and video integration | |
CN111461078B (en) | Fishing preventing and monitoring method based on computer vision technology | |
CN116743970B (en) | Intelligent management platform with video AI early warning analysis | |
CN114898523A (en) | Power transmission line image monitoring method and device with distance measuring function | |
CN115272425B (en) | Railway site area intrusion detection method and system based on three-dimensional point cloud | |
CN114235050A (en) | Marine environment monitoring and early warning method, device and system | |
CN115993597A (en) | Visual radar perception fusion method and terminal equipment | |
CN114419537A (en) | Power transmission line warning method and device based on deep learning | |
CN112643719A (en) | Tunnel security detection method and system based on inspection robot | |
CN111523386A (en) | Machine vision-based high-speed railway platform door monitoring and protecting method and system | |
CN113965733A (en) | Binocular video monitoring method, system, computer equipment and storage medium | |
CN115865988B (en) | Passenger ship passenger tread event monitoring system and method by utilizing mobile phone sensor network | |
CN112700403A (en) | Charged interval false-entry prevention detection method, medium and system | |
JP2021163401A (en) | Person detection system, person detection program, leaned model generation program and learned model | |
CN111414967A (en) | Method for improving robustness of temperature measurement system and monitoring system | |
CN116797967A (en) | Visual monitoring hidden trouble identification method and device for overhead transmission line | |
KR102407202B1 (en) | Apparatus and method for intelligently analyzing video | |
CN113963502B (en) | All-weather illegal behavior automatic inspection method and system | |
CN112149579A (en) | Improved yolo _ v 3-based bird nest hidden danger detection method | |
CN117649734B (en) | Intelligent security monitoring method and system based on multidimensional sensor | |
CN106803322B (en) | A kind of anti-limited region dividing method | |
CN116665440B (en) | Expressway inter-vehicle distance measurement method, system and equipment |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210423 |
|
RJ01 | Rejection of invention patent application after publication |