High-speed vision challenge for 100% online on-Strip inspection

2010 34th IEEE/CPMT International Electronic Manufacturing Technology Symposium (IEMT) Pub Date : 2010-11-01 DOI:10.1109/IEMT.2010.5746670

Chen Wanliang, X. Yongrong, Ruan Jianhua

{"title":"High-speed vision challenge for 100% online on-Strip inspection","authors":"Chen Wanliang, X. Yongrong, Ruan Jianhua","doi":"10.1109/IEMT.2010.5746670","DOIUrl":null,"url":null,"abstract":"The paper presents the challenges of effective high-speed detection for screen out incomplete fill (IF) reject via online Vision System at Trim/Form process. In general, most of the assembly factories screen the incomplete fill reject manually or through a vision on Test Handler. The speed of this Unit by Unit inspection is very low. Usually, the effective UPH is less than 30k. Accordingly, a new idea of online Strip by Strip inspection is bring out. By this way, total 36 units (or even more) can be detected every step. And the inspection UPH can reach to more than 500k. To achieve this target, the biggest challenge comes from that large image data/information need to be processed in tens of microseconds. As presented in this paper, based on advanced algorithms, excellent program structure and optimized hardware composition, the system can realize 100% online unit inspection with only 1 camera at a Trim/Form machine. The speed of this vision can freely match with any high-speed running Trim/Form machine, which will gain competitive advantages.","PeriodicalId":133127,"journal":{"name":"2010 34th IEEE/CPMT International Electronic Manufacturing Technology Symposium (IEMT)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 34th IEEE/CPMT International Electronic Manufacturing Technology Symposium (IEMT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IEMT.2010.5746670","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The paper presents the challenges of effective high-speed detection for screen out incomplete fill (IF) reject via online Vision System at Trim/Form process. In general, most of the assembly factories screen the incomplete fill reject manually or through a vision on Test Handler. The speed of this Unit by Unit inspection is very low. Usually, the effective UPH is less than 30k. Accordingly, a new idea of online Strip by Strip inspection is bring out. By this way, total 36 units (or even more) can be detected every step. And the inspection UPH can reach to more than 500k. To achieve this target, the biggest challenge comes from that large image data/information need to be processed in tens of microseconds. As presented in this paper, based on advanced algorithms, excellent program structure and optimized hardware composition, the system can realize 100% online unit inspection with only 1 camera at a Trim/Form machine. The speed of this vision can freely match with any high-speed running Trim/Form machine, which will gain competitive advantages.

查看原文本刊更多论文

100%在线带材检测的高速视觉挑战

本文提出了利用在线视觉系统在裁剪/成形过程中对筛出不完全填充(IF)废品进行有效的高速检测所面临的挑战。一般来说，大多数组装工厂手动或通过Test Handler上的视图筛选不完整的填充拒绝。这种单位检查的速度很低。通常，有效UPH小于30k。据此，提出了在线逐条检测的新思路。通过这种方式，每一步可以检测到总共36个单位(甚至更多)。检测UPH可达500k以上。为了实现这一目标，最大的挑战来自于需要在几十微秒内处理大量图像数据/信息。本系统采用先进的算法、优良的程序结构和优化的硬件组成，在整型机上仅使用1台摄像机即可实现100%的在线单元检测。该视觉速度可与任何高速运行的切边/成型机自由匹配，获得竞争优势。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2010 34th IEEE/CPMT International Electronic Manufacturing Technology Symposium (IEMT)

自引率

0.00%

发文量