Measurement of plant height based on stereoscopic vision under the condition of a single camera

Quantitative study of the characteristics of plants plays an important role in the plants´ growth and development research, and non-destructive measurement of the height of plants based on machine vision technology is one of the difficulties. In this paper, a new algorithm which combined three-dimensional computer vision techniques and image processing method, is proposed to solve this problem. We acquired cotton images using one single camera in the non-destructive monitoring system for cotton growth first. Through the pre-processing of the cotton image, the height of cotton is represented by coordinates of two key points on the image. Based on the parameters of the camera and some constraints of the two key points, we can obtain plant height in the world coordinate system. Experimental results show that plant height measurement error range can be controlled in less than 5mm.