摘要
为了提高瓶盖检测效率、降低成本,该文提出了一种基于形态学的饮料瓶瓶盖缺陷检测算法,在TI公司的DaVinci系列DSP处理器TMS320DM6467上实现缺陷检测算法,并利用Ping-Pong双缓存和流水线优化等技术进行算法优化。实验结果表明,有缺陷的瓶盖被成功剔除,误检率为2.32%。经代码优化后,瓶盖缺陷检测所需时间从437.14Mcycles减少到9.72Mcycles,提高了45倍,实现瓶盖缺陷实时检测。
In order to improve the efficiency of detecting the caps and reduce the cost,an algorithm of cap defects detection is proposed in this paper.The algorithm is completed on the hardware platform DM6467,and optimized in accordance with the characteristics of DM6467.Experimental results show that the defective caps are removed success,error-rate reaches 2.32percent.After code optimizing,the time required for detecting the caps defects reduced from 437.14Mcycle to 10.88Mcycles,increased nearly 45 times,detecting the cap defects real-time.
出处
《杭州电子科技大学学报（自然科学版）》
2012年第6期81-84,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
基金
浙江省重大科技专项基金资助项目(C11108-1)
关键词
缺陷检测
达芬奇
代码优化
defect detection
DaVinci
code optimization
作者简介
江晨晓（1988-），男，浙江台州人，在读研究生，计算机应用技术．
