Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
A monocular distance measurement method of orange based on the changes of target pixels
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2021 ASABE Annual International Virtual Meeting 2100315.(doi:10.13031/aim.202100315)
Authors: Yan Li, Yingqi Li, Dingke Li, Dianzhuo Zhou, Jie Liu
Keywords: monocular distance measurement; orange; contour occlusion; image processing
Abstract. In mountain orchard, the influence of light and the occlusion of branches and leaves always play an interfere role for machine vision to measure distance between the target fruit and the camera. To tackle with this common situation easily and reduce the effects of inference effectively, a method of monocular distance measurement was proposed based on the change rate of the number of pixels occupied by the target fruit in the images obtained at different shooting distances. In this work, based on the principle of monocular distance measurement and the imaging characteristics of oranges, a model between shooting distance and the number of pixels of effective area fenced by feature points which were extracted from the unobstructed part of the fruit images was established after a mass of tests through collecting images that were shot at different distance with 2.5cm fixed interval distance within 25cm to 150cm range. An algorithm according to the model was designed to achieve rapid prediction of the distance between the target fruit and the camera, and two verification sets of used images—one was shot when the target fruit was inside of camera central field, and the other one belonged to outside of that were made to verify the accuracy. The result showed that the algorithm had high accuracy with less 6% relative error for both validation sets, which proved this monocular distance measurement method based on the changes of target pixels could be well applied to obtaining depth information in fruit harvesting industry.
(Download PDF) (Export to EndNotes)