摘要
为实现机械手对物体的自主抓取,采用Kinect传感器采集RGB图像,运用HSI和Lab色彩空间分割算法对物体进行图像识别,并计算物体的空间坐标,然后根据物体的坐标值反求机械手运行角度,驱动基于S3C2410的ARM控制器的机械手,实现对目标的抓取。以温室内的成熟番茄为抓取对象,实验结果表明该系统能准确有效地实现对物体的识别、定位及抓取。
For the realization of the manipulator to fetch the object,the Kinect sensors were used to get the RGB images.HSI and Lab color space segmentation algorithm were used to realize the recognition of objects,and then the space coordinates of the object were calculated. And then according to the coordinates of the object,manipulator operation angle was gotten reversly.Driving the manipulator ARM controller based on S3C2410,the goal of fetching was achieved. Based on the mature tomato in greenhouse to grab objects,the experimental results show that the system can accurately and effectively realize the object recognition,positioning and fetching.
出处
《仪表技术与传感器》
CSCD
北大核心
2016年第7期123-126,共4页
Instrument Technique and Sensor
基金
中国博士后科学基金资助项目(2015M571783)
江苏省农机基金资助项目(GXZ14002)