Machine learning algorithm for Avocado image segmentation based on quantum enhancement and Random forest

Abstract

Precision agriculture (PA) represents the use of new technologies, specially computer vision, to increase agricultural productivity, where image segmentation plays a crucial role in several PA applications. This paper presents a Machine learning algorithm for Avocado image segmentation based on quantum enhancement and Random forest. In order to show the performance of the proposed method in term of segmentation, which represents one of the most sensible computer vision technics to noise and low illumination images, a set of experimentations based on synthetic and real images devoted to agricultural applications (avocado fruit detection and localization) are done. The Segmentation accuracy (SA) and the mean intersection over Union (MIoU) metrics are adopted to evaluate its performance against other algorithms presented in the literature. The proposed method shows good results in terms of segmentation quality, sensibility to noise and low illumination conditions, outperforming the existing and widely used binarization methods.