摘要
作物籽粒计数是作物研究中重要的工作之一。本文在总结国内外常用的籽粒计数方法及原理的基础上,对基于计算机视觉的图像分析法、碰撞声音识别法以及基于红外线的颗粒计数方法进行综合比较、分析。结果表明,图像分析法和碰撞声音法操作简单,但易受环境影响;红外计数法被广泛使用,但籽粒下落速度过快时,受光电信息处理电路限制。针对不同的研究需求,采用不同的研究方法是比较合理的选择。
Grain count is one of the important tasks in crop research.On the basis of summarizing the common methods and principles of grain counting at home and abroad,the image analysis method based on computer vision,collision sound recognition method and the particle counting method based on infrared ray were comprehensively compared and analyzed in this paper.The results showed that the image analysis method and the impact sound method were simple but easy to be affected by the environment;the infrared counting method was widely used,but when the grains fell too fast,it was limited by the photoelectric information processing circuit.According to different research needs,it is a reasonable choice to adopt different research methods.
作者
周麟
武威
刘涛
孙成明
ZHOU Lin;WU Wei;LIU Tao;SUN Cheng-ming(College of Horticulture and Plant Conservation,Yangzhou University,Yangzhou Jiangsu 225009;Agricultural College,Yangzhou University)
出处
《现代农业科技》
2020年第12期18-20,22,共4页
Modern Agricultural Science and Technology
基金
江苏高校优势学科建设工程资助项目(PAPD)。