采用运动图像处理检测排种器充填性能
Application of Kinetic Image Processing to Measurement of Seed Distribution
-
摘要: 针对人工测量和光电检测排种器性能的弊端,提出了采用运动图像处理检测排种器充填性能的方法。研制了检测多目标高速运动物体(种子)计算机图像处理系统,分析了光照对抓拍运动物体的影响和图像的采集过程,给出了系统的拍摄区域和合理的图像处理窗口,并运用最佳熵门限法准确提取目标和背景,建立了根据种子边缘、面积及种子之间的距离检测排种器充填性能的适用算法。试验表明:采用该系统误差小,速度快Abstract: In connection with the malpractice of the property of the seed distribution by using the photoelectrical checking and the human measuring, this paper proposed the application of kinetic image processing to measurement of seeds distribution. The computer image processing system for measuring multi target high speed mass objects(seeds) was established, and the effect of optical source on kinetic objects and the acquisition process of image were also analyzed in detail. The suitable window of image processing was determined. The object and background were exactly found out by optimum entropy threshold. According to the edges and area of the seeds and the distances between the seeds, the general algorithm of detecting seeds filling property was successfully presented. The experiment proves that the error is small and the speed is high by using this system. It makes a foundation for checking the property of the seeds distribution. The experiment showed that the measurement precision and velocity are satisfying.