Tomato Fruit Detection Using Modified Yolov5m Model with Convolutional Neural Networks

Fa-Ta Tsai; Van-Tung Nguyen; The-Phong Duong; Quoc-Hung Phan; Chi-Hsiang Lien

doi:10.3390/plants12173067

Tomato Fruit Detection Using Modified Yolov5m Model with Convolutional Neural Networks

Plants (Basel). 2023 Aug 26;12(17):3067. doi: 10.3390/plants12173067.

Authors

Fa-Ta Tsai¹, Van-Tung Nguyen¹, The-Phong Duong², Quoc-Hung Phan¹, Chi-Hsiang Lien¹

Affiliations

¹ Department of Mechanical Engineering, National United University, Miaoli 36002 Taiwan.
² Department of Mechanical Engineering, HCMC University of Technology and Education, Ho Chi Minh City 700000, Vietnam.

Abstract

The farming industry is facing the major challenge of intensive and inefficient harvesting labors. Thus, an efficient and automated fruit harvesting system is required. In this study, three object classification models based on Yolov5m integrated with BoTNet, ShuffleNet, and GhostNet convolutional neural networks (CNNs), respectively, are proposed for the automatic detection of tomato fruit. The various models were trained using 1508 normalized images containing three classes of cherry tomatoes, namely ripe, immature, and damaged. The detection accuracy for the three classes was found to be 94%, 95%, and 96%, respectively, for the modified Yolov5m + BoTNet model. The model thus appeared to provide a promising basis for the further development of automated harvesting systems for tomato fruit.

Keywords: Yolov5; convolutional neural network; tomato fruit detection.

Abstract

Grants and funding