A novel method for surface defect inspection of optic cable with short-wave infrared illuminance

Abstract

Intelligent on-line detection of cable quality is a crucial issue in optic cable factory, and defects on the surface of optic cable can dramatically depress cable grade. Manual inspection in optic cable quality cannot catch up with the development of optic cable industry due to its low detection efficiency and huge human cost. Therefore, real-time is highly demanded by industry in order to replace the subjective and repetitive process of manual inspection. For this reason, automatic cable defect inspection has been a trend. In this paper, a novel method for surface defect inspection of optic cable with shortwave infrared illuminance is presented. The special condition of shortwave infrared can not only provide illumination compensation for the weak illumination environment, but also can avoid the problem of exposure when using visible light illuminance, which affects the accuracy of inspection algorithm. A series of image processing algorithms are set up to analyze cable image for the verification of real-time and veracity of the detection method. Unlike some existing detection algorithms which concentrate on the characteristics of defects with an active search way, the proposed method removes the non-defective areas of the image passively at the same time of image processing, which reduces a large amount of computation. OTSU algorithm is used to convert the gray image to the binary image. Furthermore, a threshold window is designed to eliminate the fake defects, and the threshold represents the considered minimum size of defects. Besides, a new regional suppression method is proposed to deal with the edge burrs of the cable, which shows the superior performance compared with that of Open-Close operation of mathematical morphological in the boundary processing. Experimental results of 10000 samples show that the rates of miss detection and false detection are 2.35% and 0.78% respectively when equals to 0.5mm, and the average processing period of one frame image is 2.39ms. All the improvements have been verified in the paper to show the ability of our inspection method for optic cable.

A Review and Analysis of Automatic Optical Inspection and Quality Monitoring Methods in Electronics Industry

“…However, this solution is considered costly and needs special alignment. 3D optical [176] Adaptive template Rule-based 100% 0% -ViBe, SDL, SAM& RPCA [199] Systematic matrix Rule-based Up to 100% 0.05% 1.64s (total) [52], [198], [407], [408] [196] Textural Adaptive thresholding Rule-based 78% (precision) ->0.203s/image - [246] ICA Rule-based --8.1ms/image - [249] Optical flow Rule-based --0.054s/image - [248] Modified Hough transform Rule-based -0% 0.696s/image Hough transform [271] Segmentation F-SVDD 95% -7.8s/panel SVDD [271] Segmentation QK-SVDD 96% 7.54% 60ms/panel SVDD [267] PCA [415] can be considered an effective tool to mitigate these problems and to provide full description of the component 3D nature.…”

Three-dimensional defects detection of high-voltage cable joint based on iterative residual fitting

“…With the development of image processing technology, visual inspection methods with the advantages of low cost and flexible application have been widely used in industrial defect detection [8][9][10]. At present, the research on the detection of defects in high-voltage cables by machine vision methods remains in the field of twodimensional (2D) images [11][12][13][14]. This type of method mainly designs image feature vectors based on the texture and edge characteristics of defects, and then selects classifiers such as support vector machines to recognize defects.…”

Damage detection of cables in cable-stayed bridges using vibration data measured from climbing robot

“…Huang et al (2020) proposed an intelligent method for damage detection of steel wire ropes based on a convolutional neural network. Chen et al (2016) presented image processing based method for defect inspection of optic cable using short-wave infrared illuminance. Wang and Xu (2007) developed a robot system for localizing damage of stay cables.…”