CN109410186A - A kind of contact net positioner wire clamp image detecting method - Google Patents

A kind of contact net positioner wire clamp image detecting method Download PDF

Info

Publication number
CN109410186A
CN109410186A CN201811180568.0A CN201811180568A CN109410186A CN 109410186 A CN109410186 A CN 109410186A CN 201811180568 A CN201811180568 A CN 201811180568A CN 109410186 A CN109410186 A CN 109410186A
Authority
CN
China
Prior art keywords
image
line
region
wire clamp
straight line
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
Application number
CN201811180568.0A
Other languages
Chinese (zh)
Inventor
杨飏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Institute of Railway Technology
Original Assignee
Nanjing Institute of Railway Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Institute of Railway Technology filed Critical Nanjing Institute of Railway Technology
Priority to CN201811180568.0A priority Critical patent/CN109410186A/en
Publication of CN109410186A publication Critical patent/CN109410186A/en
Pending legal-status Critical Current

Links

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
    • G06T7/10—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
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The present invention provides a kind of contact net positioner wire clamp image detecting method, with the development of contact net fault detection technique, malfunction elimination under the line of the magnanimity high-definition image of the available positioning device of overhead contact line device, in order to improve the efficiency and accuracy of fault detection, the present invention is based on image processing methods, by the detection algorithm in the structure feature Design Orientation wire clamp region to registration clamp region, this method real image live for background complexity, the even, Noise of uneven illumination etc. has good practicability.

Description

A kind of contact net positioner wire clamp image detecting method
Technical field
The present invention relates to field of image detection, and in particular to a kind of contact net positioner wire clamp image detecting method.
Background technique
In the overhead contact line device of high-speed railway, registration clamp is connected to contact line and supports device, subjects bow It is also most weak components that net maximum impact force and contact line tension, which are most important, repeatedly occurs to loosen in actual operation, take off The failures such as fall, catastrophe failure even causes contact line to fall off to cause accident between pantograph, and the detection mode of registration clamp is main at present It is artificial line inspection and manually checks the mode of high-definition image, it is time-consuming and laborious.
The advantages that high-precision of image procossing, high efficiency, automation, makes it in overhead contact line intellectualized detection Using more and more extensive, locator slope detection, locator identification etc. based on image procossing are had been realized at present.However state Internal and external pin studies seldom the target detection of registration clamp and accident analysis, also there is no precedent in Practical Project.The present invention provides A kind of automatic positioning and extraction topography's method of the registration clamp based on image procossing, this method greatly improve wire clamp event The detection efficiency of barrier simultaneously provides some basis for for consequent malfunction identification.
Summary of the invention
Technical problem: in order to solve the defects of prior art, the present invention provides a kind of contact net positioner wire clamp images Detection method.
Technical solution: a kind of contact net positioner wire clamp image detecting method provided by the invention, the present invention be by with What lower technical method was realized:
Step 1, Image Edge-Detection and Morphological scale-space are carried out to Registration Device of OCS, divides foreground and background；
Step 2, straight-line detection is carried out on image after singulation and selects end region for candidate region；
Step 3, longitudinal straight-line detection is carried out in each candidate region, if there is no longitudinal straight line, then the region is non-fixed Bit line clip area domain, then further determines if it exists；
Step 4, if longitudinal straight line and lateral straight line are non-intersecting, the region non-locating wire clamp region；As intersected, with intersection point It 1/2 is that radius takes circle for the center of circle, lateral straight line, the circle and two straight lines have 3 intersection points, then are determined as T-shaped structure, the region For registration clamp region, returning to straight line endpoint is registration clamp coordinate.
Pretreatment is denoised using LoG operator by Gaussian filter smoothed image, with single order local derviation finite difference formulations ladder Amplitude and direction are spent, and inhibits non-maximum, using dual threashold value-based algorithm, steps are as follows for calculating:
Step 1, the corresponding gray value of normalization histogram peak value is sought:
Step 2, minimum length distributed area (α, β) is sought, so that α≤gmax≤ β, gray probability and >=50% in section；
Step 3, mathematical expectation of probability is sought:
Image preprocessing, characterized in that specific step is as follows:
Step 1: reducing original image；
Step 2: calculating normalization histogram mathematical expectation of probability, Canny operator Grads threshold is arranged in by stages, carries out to image Edge detection；
Step 3: morphological erosion, expansion process being carried out to edge image, connect adjacent area.
The utility model has the advantages that a kind of contact net positioner wire clamp image detecting method provided by the invention, compared with prior art, Present invention real image live for background complexity, uneven illumination, Noise etc. all has good applicability, and algorithm detection is quasi- True rate reaches 90% or more, greatly improves detection efficiency.
Specific embodiment
A kind of contact net positioner wire clamp image detecting method, comprising the following steps:
(1) edge detection for carrying out LoG method to the image of contact net registration clamp first, by operator extraction edge, Ash can be calculated according to the corresponding relationship of each gray value of image and the probability of appearance in the normalization histogram for calculating image Mathematical expectation of probability is spent, and threshold value is set according to mathematical expectation of probability distributed area, steps are as follows for the calculating of mathematical expectation of probability:
Step 1: the corresponding gray value of normalization histogram peak value is sought:
Step 2: minimum length distributed area (α, β) is sought, so that α≤gmax≤ β, gray probability and >=50% in section； Step 3: seek mathematical expectation of probability:
Symbol meaning in above-mentioned format is as follows: g is each gray value of normalization histogram；PgIt is each in normalization histogram The corresponding probability of gray value；gmaxFor the corresponding gray value of normalization histogram peak value；gmFor gray probability mean value.
(2) preprocessing process is carried out to image, its step are as follows:
Step 1: reducing original image, and locator relative size in figure is larger, and locator inspection can not be influenced by reducing original image Operation is improved on the basis of survey.
Step 2: calculating normalization histogram probability value, and Grads threshold is arranged in by stages
Step 3: to handling at edge image progress morphological erosion, expansion, adjacent area is connected.
(3) in the edge image after Morphological scale-space, each component linear feature is more obvious, carries out on this basis global Straight-line detection further carries out cluster screening according to length, position and angle information, obtains meeting the straight of locator feature Line, and extracting straight line end region is candidate region, the specific steps are as follows:
Step 1: according to locator mounting technique standard, the value of slope of locator in a certain range, is arrived actual acquisition Image in locator straight line angle analyzed, detect and screen the straight line in angular range.
Step 2: being clustered, merged and extracted class center line to the straight line after screening, opposite in the picture comprising component Position, angle and length relation.
Step 3: extraction straight line end region is candidate region, according to the positive positioning of straight line foundation location information differentiation and instead Positioning, and select left end point and right endpoint region as registration clamp candidate region.
(4) to the identification of registration clamp, its step are as follows:
Step 1: longitudinal straight-line detection carries out longitudinal straight-line detection in each candidate region, if there is no longitudinal straight Line, then the region non-locating wire clamp region, and if it exists, then carry out next step judgement
Step 2: there is longitudinal direction straight line in region in the detection of V-arrangement feature, and with lateral straight line intersection, be with two straight-line intersections The center of circle, lateral straight line 1/2 take circle for radius, and the circle and two straight lines have 3 intersection points, two straight line forming V-shape intersection constructions.

Claims (3)

1. a kind of contact net positioner wire clamp image detecting method, characterized in that comprise the steps of:
Step 1, Image Edge-Detection and Morphological scale-space are carried out to Registration Device of OCS, divides foreground and background；
Step 2, straight-line detection is carried out on image after singulation and selects end region for candidate region；
Step 3, longitudinal straight-line detection is carried out in each candidate region, if there is no longitudinal straight line, then the region non-locating line Clip area domain, then further determines if it exists；
Step 4, if longitudinal straight line and lateral straight line are non-intersecting, the region non-locating wire clamp region；It is circle with intersection point such as intersection The heart, lateral straight line 1/2 take circle for radius, and the circle and two straight lines have 3 intersection points, then are determined as T-shaped structure, which is fixed Bit line clip area domain, returning to straight line endpoint is registration clamp coordinate.
2. image preprocessing according to claim 1, characterized in that pretreatment passes through Gaussian filter using LoG operator Smoothed image denoising, with single order local derviation finite difference formulations gradient magnitude and direction, and is inhibited non-maximum, is calculated using dual threshold Method, steps are as follows for calculating:
Step 1, the corresponding gray value of normalization histogram peak value is sought:
Step 2, minimum length distributed area (α, β) is sought, so that α≤gmax≤ β, gray probability and >=50% in section；
Step 3, mathematical expectation of probability is sought:
3. image preprocessing according to claim 1, characterized in that specific step is as follows:
Step 1: reducing original image；
Step 2: calculating normalization histogram mathematical expectation of probability, Canny operator Grads threshold is arranged in by stages, carries out edge to image Detection；
Step 3: morphological erosion, expansion process being carried out to edge image, connect adjacent area.
CN201811180568.0A 2018-10-09 2018-10-09 A kind of contact net positioner wire clamp image detecting method Pending CN109410186A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811180568.0A CN109410186A (en) 2018-10-09 2018-10-09 A kind of contact net positioner wire clamp image detecting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811180568.0A CN109410186A (en) 2018-10-09 2018-10-09 A kind of contact net positioner wire clamp image detecting method

Publications (1)

Publication Number Publication Date
CN109410186A true CN109410186A (en) 2019-03-01

Family

ID=65467575

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811180568.0A Pending CN109410186A (en) 2018-10-09 2018-10-09 A kind of contact net positioner wire clamp image detecting method

Country Status (1)

Country Link
CN (1) CN109410186A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111723794A (en) * 2020-06-24 2020-09-29 西南交通大学 Real-time flexible contact net positioning point identification method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111723794A (en) * 2020-06-24 2020-09-29 西南交通大学 Real-time flexible contact net positioning point identification method
CN111723794B (en) * 2020-06-24 2022-07-22 西南交通大学 Real-time flexible contact net positioning point identification method

Similar Documents

Publication Publication Date Title
US11443437B2 (en) Vibe-based three-dimensional sonar point cloud image segmentation method
WO2021114508A1 (en) Visual navigation inspection and obstacle avoidance method for line inspection robot
CN106683076B (en) The method of train wheel tread damage detection based on textural characteristics cluster
CN111814686A (en) Vision-based power transmission line identification and foreign matter invasion online detection method
CN109993099A (en) A kind of lane line drawing recognition methods based on machine vision
CN103198705B (en) Parking place state automatic detection method
CN114693610A (en) Welding seam surface defect detection method, equipment and medium based on machine vision
CN112232222B (en) Bullet train axle box end cover bolt loss fault detection method based on image processing
CN105678737A (en) Digital image corner point detection method based on Radon transform
CN115690012A (en) Detection method for wrong connection line of electric energy meter
CN110782411A (en) Anti-vibration hammer slip identification method based on image processing
CN109410186A (en) A kind of contact net positioner wire clamp image detecting method
CN112288674A (en) Vehicle-mounted contact network component defect online identification method
CN114266895A (en) Fan blade image segmentation and splicing method and device
CN113723314A (en) Sugarcane stem node identification method based on YOLOv3 algorithm
CN113639685A (en) Displacement detection method, device, equipment and storage medium
CN112418210B (en) Intelligent classification method for tower inspection information
JP2019003575A (en) Image analysis device and maintenance management method of railroad facility component
Cheng et al. Power pole detection based on graph cut
CN114219818A (en) Rigid contact net positioning wire clamp clamping stagnation detection method and device, computer equipment and storage medium
CN111062959B (en) Extraction and characterization method for bottom edge burr cutting characteristics of aviation thin-wall micro-structural part
CN115619813A (en) SEM image foreground extraction method and device, computer equipment and storage medium
CN112330633B (en) Jumper wire adhesive tape damage fault image segmentation method based on self-adaptive band-pass filtering
CN115100443A (en) Cable defect identification method based on high-speed template matching calculation
CN110310276B (en) Method for positioning pipe cap of supporting device of electrified railway based on geometric characteristics

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20190301