基于RGB统计特征的农作物虫害图像感知系统设计
Design of A Crop Pest Image Perception System Based on RGB Statistical Characteristics
周杰 1潘宏侠 2唐明军 1陶涛1
作者信息
- 1. 扬州工业职业技术学院电气与信息工程学院,江苏扬州225127
- 2. 中北大学机械工程学院,山西太原030051
- 折叠
摘要
基于RGB统计特征,设计了一套农作物虫害图像感知系统,以实现对农作物生长状态的实时监测与分析.系统以视频服务器为核心,在连续视频采集周期内通过自相关函数的计算,筛选出消除外部随机干扰的基准图像,对相邻的基准图像进行差分运算,消除相似像素并滤波后统计其RGB特征值.最后,在视频服务器中用VC++开发了人机交互界面,移动客户端可以通过桌面远程控制实现与服务器数据同步.
Abstract
A crop pest image sensing system based on RGB statistical characteristics was designed, so as to realize the real-time monitoring and analysis of crop growth.The video server is used as the core of the system, and the reference image is eliminated by calculating the auto-correlation function in the continuous video acquisition period.The difference of adjacent pixels is calculated, the similar pixels are eliminated and after image filtering, the RGB characteristic values are calculated.The standard deviation of each component of RGB is used to describe the severity of the damage.Experimental results show that the sensitivity and accuracy of the system to different types and different degrees of pest invasion are satisfactory.Finally, we use the VC++to develop the man-machine interface in the video server, and the mobile client can synchronize with the server through the remote control of the desktop .
关键词
农作物/RGB/图像感知/虫害Key words
Crops/RGB/Image perception/Insect pests引用本文复制引用
基金项目
扬州市重点研发计划社会发展面上项目(YZ2015084)
出版年
2017
国家科技期刊平台
NSTL
维普
万方数据