Potato volume measurement based on RGB-D camera

Abstract

The volume is one of the important parameters to evaluate the potato quality. In order to detect the volume of potato tuber automatically, this paper presents a potato volume measurement method based on RGB-D camera. A RGB-D camera established based on binocular stereo vision was used. A novel method was developed to measure potato volume and built utilizing a new image processing algorithm for depth images. The experiment was conducted with the depth image collection and volume detection of 120 potatoes. According to the shape observation, the sampled potatoes were divided into regular and irregular shape group by experience. The 60 potatoes with the best appearance and spheroid shape were marked as regular potatoes, and the 60 potatoes with the worst appearance and lopsided shape were marked as irregular potatoes. By verifying the algorithmic feasibility of the system and the accuracy of the measurement results, the results show that: the prediction error of 60 normal potatoes is 9%; the prediction error of 60 irregular potatoes is 30%, which could be used to measure the potato volume without damage.