摘要
[目的]优化薏米视觉检测中图像采集条件。[方法]在薏米视觉检测中通过图像采集设备采集薏米图像,以灰色、黑色、红色背景,在0～80 lx的光强区间,对采集后的图像进行阈值分割,采用阈值分割算法对图像进行处理。以分割图像薏米粒误差率为评价标准,分析了背景颜色、光照强度2个因素对薏米图像的影响,以确定薏米视觉检测中图像采集的最佳环境。[结果]较好的图像采集环境是黑色背景下50～60 lx的光照强度,其误差控制在6%,且在该光照区间旁光照强度上下浮动10 lx内误差也在10%内,这种环境能有效提高薏米图像的质量。[结论]优化薏米图像采集环境可大幅度降低图像噪声的影响,提高了薏米图像的质量。
[Objective]The study aimed to optimize the image acquisition conditions in the visual inspection of coix seed.[Method] In the visual inspection of coix seed,the coix seed images were collected by the image acquisition equipment.With the gray,black and red as the background,in light intensity interval of 0~80 lx,the collected images were made for the threshold segmentation and treated by using the threshold segmentation algorithm.With the grain error rate of segmentation images as the evaluation standard,the effects of two factors including the background color and light intensity were analyzed to confirm the best environment for the image acquisition in the visual inspection of coix seed.[Result] The good image acquisition environment was 50 ~ 60 lx illumination intensity under the black background,the error control was in 6%,and the error of the illumination intensity in fluctuating about 10 lux beside the illumination interval was within 10%,indicating that this kind of environment could effectively improve the quality of coix seed image.[Conclusion]The optimization of the image acquisition environment for the coix seed could greatly reduce the influence of the image noise and improve the quality of the coix seed image.
出处
《安徽农业科学》
CAS
北大核心
2011年第32期20235-20236,共2页
Journal of Anhui Agricultural Sciences
关键词
薏米
采集条件
光照强度
背景颜色
Coix seed
Acquisition conditions
Light intensity
Background color
