基于杂草图像处理的阈值分割算法研究

Research on Threshold Segmentation Algorithm Based on Weed Image Processing

杂草在生长过程中会和作物争抢土壤当中的养分,造成作物的质量和产量降低。而目前存在的除草方式:人工除草要耗费人力,而且效率一般;机械除草的机器一般较大,会造成土壤的板结问题;由于使用过量的除草剂,会给作物和环境带来了严重的影响,而且作物会产生抗药性。为了解决这个问题,研究人员不得不重新研究一种除草方式,随着自动化、智能化的推进,除草机器人也随之出现。运用MATLAB对采集到的杂草图片进行超绿灰度化、直方图均衡化和阈值分割处理,为之后运用智能算法对杂草进行分类识别奠定基础。

Weeds can compete with crops for nutrients in the soil during growth,resulting in lower crop quality and yield.And the weeding methods that currently exist:manual weeding is labor-intensive and generally efficient,mechanical weeding machines are generally larger and can cause soil sloughing problems.The use of excessive amounts of herbicides can have serious effects on crops and the environment,and crops can develop resistance to them.To solve this problem,researchers had to re-examine a way of weeding,and with the advance of automation and intelligence,weeding robots emerged.MATLAB is used to process the collected weed images with super green grayscale,histogram equalization and threshold segmentation to lay the foundation for the subsequent classification and recognition of weeds using intelligent algorithms.