摘要
作物识别技术广泛应用于杂草精准对靶施药、果实采摘机器人、植物病虫害识别等方面。从机器视觉和激光探测两方面分析了国内外作物识别的研究现状,机器视觉识别作物主要是利用作物的颜色、纹理、形状、位置特征;激光探测识别作物利用激光的测距信息。分析了国内外融合激光雷达和摄像机信息识别作物的研究现状,总结了激光雷达和摄像机联合标定的方法,并指出其在作物识别中的重要性。
Crop recognition technology is widely used in precise targeted spray of weed, fruit picking robot, plant pest identification and others. Laser acquires the depth information of crops, while the image information is captured by the camera. The actual size and position can be obtained by fusion of depth information and image information of crops.Domestic and international crop recognition researches were analyzed from the two aspects of machine vision and laser detection. Crop recognition based on machine vision used color, texture, shape and location features, while crop recognition based on laser detection used the depth information. The current situation of the fusion of laser radar information and camera information to identify crops was analyzed. Meanwhile the methods of combined calibration of laser radar and camera were summed up and the importance of combined calibration of laser radar and camera in crop recognition was pointed out.
出处
《广东农业科学》
CAS
CSCD
北大核心
2014年第24期161-165,178,共6页
Guangdong Agricultural Sciences
基金
国家林业局"948"项目(2013-4-02)
2013年留学人员科技活动项目择优经费
关键词
激光
机器视觉
联合标定
识别
作物
laser
machine vision
combined calibration
recognition
crop
