A Review of Optimization Strategy-Based Image Processing Techniques in Apple Picking

College of Humanities & Information Changchun University of Technology, Computer Science and Engineering, 130000, Changchun, China

^* Corresponding author: mgq@ldy.edu.rs

Abstract

This paper reviews the application of optimization strategy- based image processing techniques in apple harvesting, focusing on the important role of various image processing methods in automated agriculture. These image processing methods include convolutional neural networks (CNN), support vector machines (SVM), artificial neural networks (ANN), and deep learning (DL) techniques. In apple target detection, localization, segmentation, ripeness, and quality detection. These techniques can effectively improve the accuracy and efficiency of automated picking systems, reduce the dependence on labor, and at the same time reduce the risk of fruit damage. However, there are still many challenges in practical applications, such as environmental factors like fruit occlusion, light variations, and complex backgrounds, which pose requirements on the adaptability and robustness of image processing techniques. To address these challenges, this paper explores various optimization strategies to improve the performance in complex environments. This paper looks forward to future research directions and proposes that the lightweight model should be further optimized, the real-time and adaptability of the algorithms should be improved, and low-cost and high-efficiency sensors and data processing techniques should be developed in order to promote the application and popularization of image processing technology in large- scale automated apple picking.