Fruit Classification based on Color and Shape Features in Real Time Video Sequences

Abstract

Fruit classification plays a crucial role in agricultural automation, including fruit harvesting robots, quality control, and crop monitoring. This paper presents a real time computer vision system for classifying fruits in outdoor environemnts using continuous video sequences. It focuses on extracting color and shape features to overcome challenges like varying lighting and motion. In this paper, fruit features were extracted from each frame and utilized machine learning classifiers to classify fruits. Each frame is processed using computer vision techniques to segment fruits and obtain relevant features, which are then classified using a Support Vector Machine (SVM) classifier. The results showed that by combining both color and shape features along with machine learning algorithms for classifying fruits in real time enhances recognition accuracy in an outdoor environments. The proposed system achieved overall classification accuracy of 91.5%. The future work will focus on improving accuracy in case of occlusion and green fruits.