| Summary: | 碩士 === 國立雲林科技大學 === 電子工程系 === 103 === Tire inner linear is used in every tire to prevent air from escaping to make tire pressure stable. So defects on tire inner linear which can not be always avoid in tire manufacturing process will affect traffic safety seriously. At present tire companies still mostly depend on inspect personnel to complete defect inspection and the final result is affected by personnel's experience or proficiency, for the reason that human power inspection is lack of reliability.
In the study, the main purpose is to do pre-detect for tire before excluding defects from quality control (QC). We use ring light source for system illumination and we do image processing using OpenCV library. We utilized histogram equalization, smooth processing, image binary threshold, and morphological processing to segment the defects from background. And we present two methods of inspection for blob defects and skew line defect. Blob analysis is for blob defects and Hough transform is for skew line defect.
We can determine the type of blob defects by using parameters of perimeter, area and circularity by blob analysis. And we also can determine the skew line defects by using parameter of slope from Hough transform.
In this study, we realized relevant solutions for blob defects and skew line defect on tire inner linear. The experimental results shows that methods which we presented have 100% of identification success rate in reproducibility test and it has the advantages over other methods such as simple, low cost, pc base, and with open source.
|
|---|
