基于图像处理的粮食虫害自动分级虚拟系统研究  

Automatic classification virtual system of grain pest based on image processing

作　　者：颜丙生[1] 陈富恒 郭军宏[2] 

机构地区：[1]河南工业大学机电工程学院,河南郑州450007 [2]国家知识产权局专利局专利审查协作北京中心,北京100086

出　　处：《粮食与饲料工业》2014年第6期18-20,26,共4页Cereal & Feed Industry

基　　金：河南省教育厅自然科学重点基金(13B430946);河南工业大学高层次人才基金(2011BS047)资助

摘　　要：针对粮食害虫检测自动化程度不高的问题,研究开发了基于图像处理的粮食虫害程度自动分级虚拟系统,该系统采用LabVIEW的机器视觉工具包,对采集到的粮食害虫图像进行处理,提出了利用像素比例的方法来实现粮虫计数和虫害自动分级报警。利用仿真图像对所设计的虚拟系统进行了试验研究,研究结果表明,粮虫计数误差小于5%,此系统可以用于某种特定粮食虫害的在线监测。For the automatic detection level of grain pest was not high, the automatic classification virtual system of grain pest based on image processing was developed. Using machine vision toolkit LabVIEW, the grain pests images collected were processed, and grain pests counting and automatic grading alarm were achieved using the pixeI ratio methods. The virtual sys- tem design was studied by means of simulation image, and the results showed that the grain pest counting error was less than 5%, so this system could he used in the on line monitoring for a particular grain pest.

关 键 词：储粮害虫 虚拟系统 自动分级 图像处理 

分 类 号：S379.5[农业科学—农产品加工] TP391.41[农业科学—农艺学]