Comparative analysis of mature tomato detection by feature extraction and machine learning for autonomous greenhouse robots

Accurate detection of tomatoes grown in greenhouses is important for timely harvesting. In this way, it ensured that mature are collected by distinguishing them from the unripe ones. Insufficient light, occlusion, and overlapping adversely affect tomatoes. addition, time consuming people to detect at certain periods large greenhouses. For these reasons, high-performance automatic greenhouse robots has become an increasingly studied area today. paper, two feature extraction methods, histogram oriented gradients (HOG) local binary patterns (LBP), which effective object recognition, commonly used classifiers machine learning, support vector machines (SVM) k-nearest neighbor (kNN), comparatively count The HOG LBP features classified separately together SVM or kNN, success each case compared. Performance improved eliminating false positive results postprocessing stage using color information.