计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (24): 144-148.

• 图形、图像、模式识别 • 上一篇    下一篇

KLT织物疵点检测算法研究及FPGA实现

朱学亮,柴志雷,梁久祯,钟传杰   

  1. 江南大学 物联网工程学院,江苏 无锡 214122
  • 出版日期:2012-08-21 发布日期:2012-08-21

Study and implementation of FPGA-based KLT algorithm for fabric inspection system

ZHU Xueliang, CHAI Zhilei, LIANG Jiuzhen, ZHONG Chuanjie   

  1. College of Internet of Things, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Online:2012-08-21 Published:2012-08-21

摘要: 提出了一种基于FPGA的织物疵点检测系统的实现方案,设计在充分利用FPGA的并行体系架构和丰富的块存储资源的情况下引入KLT(Kanade-Lucas-Tomasi)特征点检测算法对织物疵点进行实时检测,以灵活的硬件可编程来满足KLT检测算法的调整以适应不同的检测要求。经验证,系统可以在实际的应用中准确地检测出织物上的疵点,且检测灵活方便。

关键词: 现场可编程门阵列(FPGA), 图像处理, KLT算法, 织物疵点检测

Abstract: A new fabric inspection system based on FPGA is presented. Making use of  FPGA’s parallel structure and rich memory resource design KLT detection algorithm is exploited to do real-time detection and the flexibility of programmed hardware can satisfy the adjust of KLT detection algorithm so as to adapt to different detection requirements. The verification indicates that the design can satisfy the requirements of practical application well and make the detection flexible and convenient.

Key words: Field-Programmable Gate Array(FPGA), image processing, KLT, fabric defect detection