Bolt looseness detection based on Canny edge detection algorithm
Daoyuan Song, Xinghua Xu, Xiaopeng Cui, Yangbin Ou, Weiming Chen
Abstract
Summary Aiming at current serious problems that the bolt looseness can cause heavy damages and the manual bolt looseness detection is ineffective, this study presents a bolt looseness detection method based on Canny edge detector. First, HSV (hue, saturation, value) color space is used to extract the area of mark bar. Then, after expanding, eroding and median filtering the image, Canny edge detector is used to segment the marker area to extract its edge. Last, the fitted slope of edge is used to judge whether the bolt is loose. The experimental results show that the proposed method has F1 score of 88.37%, contributing to the automated detection of bolt looseness.
Topics & Concepts
Canny edge detectorDeriche edge detectorEdge detectionEnhanced Data Rates for GSM EvolutionArtificial intelligenceComputer visionHueDetectorCorner detectionComputer scienceHSL and HSVMathematicsImage (mathematics)Image processingBiologyVirologyTelecommunicationsVirusImage and Object Detection TechniquesIndustrial Vision Systems and Defect DetectionAdvanced Neural Network Applications