论文部分内容阅读
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
In view of the disadvantages of manual measurement and photoelectric detection of seeding performance, a method of detecting the filling performance of the seeder with moving image processing is proposed. A computer image processing system for detecting multi-target high-speed moving objects (seeds) is developed. The influence of light on the captured moving objects and the image acquisition process are analyzed. The shooting area of the system and a reasonable image processing window are given. Threshold method to accurately extract the target and background, established based on the seed edge, the area and the distance between the seeds to test the applicator filling performance of the applicable algorithm. Experiments show that: the use of the system error is small, fast