Segmented Embedded Rapid Defect Detection Method for Bearing Surface Defects

Abstract

The rapid development of machine vision has prompted the continuous emergence of new detection systems and algorithms in surface defect detection. However, most of the existing methods establish their systems with few comparisons and verifications, and the methods described still have various problems. Thus, an original defect detection method: Segmented Embedded Rapid Defect Detection Method for Surface Defects (SERDD) is proposed in this paper. This method realizes the two-way fusion of image processing and defect detection, which can efficiently and accurately detect surface defects such as depression, scratches, notches, oil, shallow characters, abnormal dimensions, etc. Besides, the character recognition method based on Spatial Pyramid Character Proportion Matching (SPCPM) is used to identify the engraved characters on the bearing dust cover. Moreover, the problem of characters being cut in coordinate transformation is solved through Image Self-Stitching-and-Cropping (ISSC). This paper adopts adequate real image data to verify and compare the methods and proves the effectiveness and advancement through detection accuracy, missing alarm rate, and false alarm rate. This method can provide machine vision technical support for bearing surface defect detection in its real sense.

Research on defect detection method of bearing dust cover based on machine vision and multi-feature fusion algorithm

“…The second method is to identify defect features in bearing images through continuous iterative training of deep learning [9,10]. Lei et al [11] proposed a segmented embedded rapid defect detection method for surface defects, which realized the bidirectional fusion of image processing and defect detection and can accurately detect surface defects such as dents and scratches. Liu et al [12] proposed a detection system for automatically synthesizing tiny defects on bearing surfaces.…”

A novel CNN ensemble framework for bearing surface defects classification based on transfer learning

“…Generally, a machine vision inspection system comprises four modules, namely, image acquisition, image processing, feature extraction, and decision making [5]. Many vision-based methods have been developed for bearing surface defect detection [6][7][8][9]. For instance, Lei et al [6] detected notch, abnormal dimension, non-character region, and character defects in the four image processing stages of bearing position, bearing segmentation, character extraction, and character recognition.…”

“…Many vision-based methods have been developed for bearing surface defect detection [6][7][8][9]. For instance, Lei et al [6] detected notch, abnormal dimension, non-character region, and character defects in the four image processing stages of bearing position, bearing segmentation, character extraction, and character recognition. Liu et al [7] utilized a coaxial light (CL) source image to detect gaps and stains on the bearing surface based on the Otsu binary threshold segmentation method.…”

Endüstriyel uygulamalarda güvenilir bir kalite kontrolü için yapay sinir ağı kullanan gerçek zamanlı bir desen eşleştirme algoritmasının geliştirilmesi

“…Thus, human errors in industrial applications have been minimized and made more stable. Many studies have been done in the literature on the subject (Kamtongdee et al, 2013;Hengdi et al, 2011;Panoiu et al, 2015;Hryniewicz et al, 2015;Patil and Ingle, 2020;Koniar et al, 2014;Rouget et al, 2018;Kalina and Golovanov, 2019;Zhou et al, 2020).…”

“…The normalized crosscorrelation model is considered for silkworm gender identification. In another study (Hengdi et al, 2011), it was aimed to find defective characters in the bearing production process with the bearing character recognition system. Character recognition was done using normalization processes; character traits have been extracted and recognized.…”