首页> 外文会议>Annual International Meeting of the American Society of Agricultural and Biological Engineers >An algorithm for Distinguishing Potato Tubers on the Conveyor of the Potato Harvester using UV Camera
【24h】

An algorithm for Distinguishing Potato Tubers on the Conveyor of the Potato Harvester using UV Camera

机译:用UV照相机将马铃薯块茎区分马铃薯块茎的算法

获取原文
获取外文期刊封面目录资料

摘要

Mechanical potato harvesters carry tubers through a conveying system which contributes in removing plant residues and fine soil, whereas clods are still being removed manually. This research suggests a new mechanical system for removing the clods based on machine vision. The objective of this paper is to find the ability to distinguish between potato tubers and clods in UV images by developing an algorithm for discrimination between tubers and clods when they are flowing on the conveyor. The discrimination is based on the difference in the UV intensity values between tubers and clods by choosing a threshold value that separate the objects into two groups. However, since the threshold changes according to the conditions of imaging, an algorithm for finding the threshold automatically and processing the raw images was developed. The threshold of each image was found by smoothing the intensity histogram until obtaining a transformed histogram which has peaks representing the background, the clods andthe tubers. The background of the images included the rods of the conveyor of the harvester and the spacing between the rods. The middle point between two peaks was set as the threshold values, according to which, the pixels in the image were segmented.Next, noise was erased and misclassification was corrected by labeling and applying majority rule. A calibration method was applied by placing a stationary potato tuber in the vision range of the camera. This allowed potato tubers to be discriminated with a success rate that exceeded 95% and insured that at least 83% of the clods will be removed in a vision system which separates 3 objects at a time.
机译:机械马铃薯收割机通过输送系统携带块块,这有助于去除植物残留物和细土壤,而Clod仍然被手动去除。该研究表明,用于基于机器视觉去除Clods的新机械系统。本文的目的是通过开发在传送器上流动时,通过开发块块和可靠的差异算法来区分紫外图像中的紫外线图像中的紫外线图像的能力。通过选择将对象分为两组的阈值来基于块茎和CLOD之间的UV强度值之间的差异。然而,由于阈值根据成像条件而改变,因此开发了一种用于自动查找阈值并处理原始图像的算法。通过平滑强度直方图,直到获得具有表示背景的峰的变换的直方图,找到每个图像的阈值。图像的背景包括收割机的输送机的杆和杆之间的间隔。将两个峰之间的中间点设置为阈值,根据该阈值,根据该阈值,图像中的像素被分段。用爆炸,通过标记和施加多数规则来纠正噪声并纠正错误分类。通过将固定的马铃薯块板放置在相机的视觉范围内来施加校准方法。这允许的马铃薯块茎与超过95%的成功率进行区分,并保险成至少83%的Clod在视觉系统中将被移除一次3个物体。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号