Paper Title
Shape-Based Fruit Counting: Innovations in Yield Estimation Using Machine Learning Techniques

Abstract
The paper presents the development of an automated orange fruit counting system using shape descriptors and machine learning algorithms. The paper aims to address the need for efficient and accurate fruit counting methods in the agricultural industry. The system utilizes computer vision techniques to analyze images of orange fruits and employs shape descriptors, namely HOG and SIFT, to extract relevant features. These features are then used as inputs for machine learning algorithms, including SVM, KNN, and K-means, to classify and count the orange fruits. The paper involves a comprehensive experimental setup, implementation of the proposed system, and evaluation of its performance using a dataset of orange fruit images. The results demonstrate the effectiveness of the automated counting system of two best models with an accuracy of 100% and 75% respectively, quantifying orange fruits and provide insights for potential applications in crop management. Keywords - Machine Learning, Fruit Counting, Computer Vision, Shape Descriptors, Histogram of Gradient, Scale Invariant-Feature Transform