摘要
介绍了一种采用CMOS图像传感器来实现智能机器视觉识别的方法,通过LM9638采集图像数据,经过复杂可编程逻辑器件(CPLD)(XC95144-TQ144)进行总线传输和转换,然后ARM以直接内存存取(DMA)的传输方式将图像数据保存到存储器中。同时ARM将采集到的图片数据进行预处理,通过模板对比算法,将采集到的数据与事先存储的数据进行分析和比较,完成机器智能视觉识别。通过CPLD对采集时序进行了分析,并进行了测试仿真,经验证后速度和效果均达到设计要求。
An intelligent machine vision recognition method with a CMOS image sensor is intro- duced. The image data are captured through the LM9638, and the bus transport and conversion are implemented with CPLD （XC95144-TQ144） . Then the image data are saved to the memory in DMA mode with an ARM. In the ARM, the collected image data are pre-processed, and ana- lyzed and compared with the pre-stored data through the template comparison algorithm to com- plete the machine visual recognition. The timing of the acquisition is analyzed with a CPLD and the testing simulation is performed. It is demonstrated that the speed and effectiveness of the design meet the requirements.
出处
《电子产品可靠性与环境试验》
2012年第3期65-69,共5页
Electronic Product Reliability and Environmental Testing
