计算机视觉运用于棉花水分养分供给系统设计  

Cotton Automatic Water and Nutrition Supply System Design Based on Computer Vision

作　　者：耿宾涛[1] 王现君[2] 

机构地区：[1]平顶山工业职业技术学院网络教育学院,河南平顶山467000 [2]平顶山工业职业技术学院计算机与软件工程学院,河南平顶山467000

出　　处：《农机化研究》2017年第11期179-182,共4页Journal of Agricultural Mechanization Research

基　　金：河南省科技厅科技项目(20160375)

摘　　要：针对目前棉花水分检测存在的缺陷和过量施肥的问题,提出了一种基于计算机视觉的棉花水分养分亏缺实时检测和水分养分自动供给系统。运用配置有近红外光发射器的CCD照相机采集得到棉花叶片图像信息,对图像进行去噪音等预处理得到植物叶片的图像,对叶片水分亏缺和缺素信息相关的颜色特征、纹理的时域特征、纹理的频域特征及缺素初期斑点的时频特征,分别运用相对差值百分率直方图、差分算子,以及傅里叶变换和小波包提取等新方法进行提取,并利用遗传算法对提取的众多特征项进行优化选择,建立棉花叶片图像信息与水分养分亏缺状况的对应关系,并通过物联网技术连接微喷灌终端实现水分养分定时定量自动供给。将系统于湖北进行试用试验,试验表明:本系统大幅度提高了棉花产量,降低了灌溉水消耗量、肥料消耗量和土壤肥料残留量,减轻了农业生产对环境的污染,有较好的应用前景,值得推广。A automatic water and agricultural fertilizer supply system based on computer vision is designed to solve the problem exits in cotton moisture detection and the excessive fertilization problem. A CCD camera equipped with a nearinfrared transmitter is applied to get cotton leaves image information which is processed through relative difference percentage histogram to extract color feature, difference operator is applied to extract the time domain features; Fourier transformer is used to extract frequency domain characters of leaves texture ; wavelet packet is employed to extract the time-frequency characteristics, and genetic algorithm is used to optimized to features to established the corresponding relationship between the cotton leaf image information and nutrient and water deficit of the cotton. The designed system is build up and works in Hubei province for testing its performance, and the results shows that this system greatly improved cotton production, reduced the amount of irrigation water, the amount of fertilizer and the residual amount of soil fertilizer , such reduce the pollution to the environment and the application prospect is good.

关 键 词：计算机视觉 图像分析 水分诊断 养分诊断 微喷灌系统 棉花 

分 类 号：S126[农业科学—农业基础科学]