计算机图像智能识别下的割草机器人设计研究  

Research on Design of Mowing Robot Based on Computer Image Intelligent Recognition

作　　者：袁社锋[1] Yuan Shefeng(Henan Vocational College of Agricultural,Zhengzhou 451450,China)

机构地区：[1]河南农业职业学院,郑州451450

出　　处：《农机化研究》2024年第11期136-139,共4页Journal of Agricultural Mechanization Research

基　　金：河南省高等教育教学改革研究与实践项目(2021SJGLX698,2021SJGLX835)。

摘　　要：为了提升割草机器人的工作效率、安全及自主性,基于堆叠降噪自动编码机设计了智能图像识别算法,用于实现割草机器人进行作业时自动化识别环境,以进一步提高工作效率。将割草机器人视觉传感器所采集的草地图像作为输入信号,通过叠加多层自动降噪编码机组成深度神经网络,可以深入挖掘草地图像所携带的信息,识别并提取图像特征。通过训练所建立网络,获得稳定输出,提高了割草机器人识别目标准确率。试验结果表明:本算法可进一步提高割草机器人识别准确率,从而提高工作效率。By studying the operation process and working environment of the mowing robot,in order to improve the working efficiency,safety and autonomy of the mowing robot,an intelligent image recognition algorithm is designed based on the stack noise reduction automatic encoder,which is used to realize the automatic recognition environment of the mowing robot during operation,and further improve the working efficiency.The grassland image collected by the vision sensor of the mowing robot is taken as the input signal,and a depth neural network is formed by superimposing a multi-layer automatic noise reduction encoder.The network can deeply mine the information carried by grassland images,recognize and extract image features.Through the network established by the training institute,stable output can be obtained,and the accuracy of target recognition of the mowing robot can be improved.The experimental results show that the algorithm in this paper can further improve the recognition accuracy of the mowing robot,thus improving the work efficiency.

关 键 词：图像识别 机器学习 特征提取 降噪自动编码机 割草机器人 

分 类 号：S817.111[农业科学—畜牧学] TP242[农业科学—畜牧兽医]