Deep learning-based real-time detection and classification of tomato ripeness stages using YOLOv8 on raspberry Pi

Deep learning-based real-time detection and classification of tomato ripeness stages using YOLOv8 on raspberry Pi Md Nahiduzzaman Rusab Sarmun Amith Khandakar http://orcid.org/0000-0001-7068-9112 Md Ahsan Atick Faisal Munshi Sajidul Islam Mohammad Kaoser Alam Tawsifur Rahman Nasser Al-Emadi M Murugappan http://orcid.org/0000-0002-5839-4589 Muhammad E H Chowdhury http://orcid.org/0000-0003-0744-8206 Abstract

The automated detection of tomato ripeness is critical in crop management and harvesting. In most earlier works, tomato image ripeness detection has been based upon a limited set of images and binary classification (ripe and unripe). This study uses the cutting-edge YOLOv8 object detection algorithm and a comprehensive dataset to propose an accurate real-time system for detecting and classifying tomato ripeness (multi-class). Based on two open-source datasets (Kaggle and Internet-sourced), we developed and tested the proposed system. In this method, a comprehensive tomato image dataset is curated, YOLOv8 models are built, seamlessly integrated into an embedded system (Raspberry Pi4), then evaluated and validated. The model shows exceptional performance in detecting three distinct classes of ripeness: unripe, partially ripe, and ripe. It surpasses existing state-of-the-art models in both accuracy and efficiency. Based on the Kaggle dataset, our model achieves an average precision at 50 of 0.808, with F1-scores of 0.80, 0.65, and 0.796 for unripe, partially ripe, and ripe classes, respectively. It achieves mAP at 50 of 0.725 and F1-scores of 0.747 (unripe), 0.652 (partially ripe), and 0.72 (ripe) for the corresponding classes of the Internet-sourced Dataset, exceeding current state-of-the-art models. Finally, the proposed tomato ripeness detection algorithm is implemented on the Raspberry Pi 4 system and exhibits notable performance. With the integration of YOLOv8 into an embedded system (Respbeery Pi4), it can be used to improve efficiency and reduce labor costs in tomato-picking robots, helping to revolutionize agricultural practices.
01 16 2025 03 31 2025 015219 10.1088/crossmark-policy iopscience.iop.org Deep learning-based real-time detection and classification of tomato ripeness stages using YOLOv8 on raspberry Pi Engineering Research Express paper © 2025 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved. 2024-10-29 2025-01-07 2025-01-16 https://iopscience.iop.org/page/copyright https://iopscience.iop.org/info/page/text-and-data-mining 10.1088/2631-8695/ada720 https://iopscience.iop.org/article/10.1088/2631-8695/ada720 https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf https://iopscience.iop.org/article/10.1088/2631-8695/ada720 https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf https://iopscience.iop.org/article/10.1088/2631-8695/ada720 https://iopscience.iop.org/article/10.1088/2631-8695/ada720/pdf Comput. Electron. Agric. Zheng 10.1016/j.compag.2022.107029 198 2022 Research on tomato detection in natural environment based on RC-YOLOv4 Comput. Electron. Agric. Wan 10.1016/j.compag.2018.01.011 146 43 2018 A methodology for fresh tomato maturity detection using computer vision Zhao 347 2019 Healthy and diseased tomatoes detection based on YOLOv2 Comput. Electron. Agric. Chen 10.1016/j.compag.2021.106450 190 2021 Detecting ripe fruits under natural occlusion and illumination conditions Agronomy Su 10.3390/agronomy12071638 12 1638 2022 Tomato maturity classification based on SE-YOLOv3-MobileNetV1 network under nature greenhouse environment AgriEngineering Chowdhury 10.3390/agriengineering3020020 3 294 2021 Automatic and reliable leaf disease detection using deep learning techniques Rupanagudi 7 2014 A cost effective tomato maturity grading system using image processing for farmers PLoS One Abisha 10.1371/journal.pone.0284021 18 2023 Brinjal leaf diseases detection based on discrete shearlet transform and deep convolutional neural network Comput. Electron. Agric. Kurtulmus 10.1016/j.compag.2011.07.001 78 140 2011 Green citrus detection using eigenfruit, color and circular gabor texture features under natural outdoor conditions Bioengineering Martis 10.3390/bioengineering11080799 11 799 2024 Novel hybrid quantum architecture-based lung cancer detection using chest radiograph and computerized tomography images 2009 Innovative Technologies in Intelligent Systems and Industrial Applications Murugappan 10.1109/CITISIA.2009.5224237 70 2009 Appraising human emotions using time frequency analysis based EEG alpha band features Karthikeyan 10.1007/978-3-642-35197-6_26 vol 330 2012 EMG signal based human stress level classification using wavelet packet transform Phys. Eng. Sci. Med. Hemalakshmi 10.1007/s13246-024-01410-3 47 863 2024 PE-Ynet: a novel attention-based multi-task model for pulmonary embolism detection using CT pulmonary angiography (CTPA) scan images Hum-Cent. Intell. Syst. Ahamed 10.1007/s44230-024-00067-1 4 299 2024 Automated colorectal polyps detection from endoscopic Images using MultiResUNet framework with attention guided segmentation Adv. Eng. Inf. Wang 10.1016/j.aei.2023.102280 59 2024 A high-accuracy and lightweight detector based on a graph convolution network for strip surface defect detection Meas. Sci. Technol. Wang 10.1088/1361-6501/ad63c2 35 2024 High-accuracy and lightweight weld surface defect detector based on graph convolution decoupling head IEEE Trans. Instrum. Meas. Wang 10.1109/TIM.2023.3302372 72 1 2023 YOLO-MSAPF: multiscale alignment fusion with parallel feature filtering model for high accuracy weld defect detection Sensors Liu 10.3390/s20072145 20 2145 2020 YOLO-tomato: a robust algorithm for tomato detection based on YOLOv3 Sensors Magalhães 10.3390/s21103569 21 3569 2021 Evaluating the single-shot multibox detector and YOLO deep learning models for the detection of tomatoes in a greenhouse Multimedia Tools Appl. Lawal 10.1007/s11042-021-10933-w 80 26751 2021 Development of tomato detection model for robotic platform using deep learning Agriculture Egi 10.3390/agriculture12091290 12 1290 2022 Drone-computer communication based tomato generative organ counting model using YOLO V5 and deep-sort Machines Ge 10.3390/machines10060489 10 489 2022 Tracking and counting of tomato at different growth period using an improving YOLO-deepsort network for inspection robot Frontiers in Plant Science Wang 10.3389/fpls.2021.634103 12 2021 Tomato anomalies detection in greenhouse scenarios based on YOLO-Dense International Journal of Fuzzy Logic and Intelligent Systems Nugroho 10.5391/IJFIS.2022.22.3.223 22 223 2022 Comparison of deep learning-based object classification methods for detecting tomato ripeness Frontiers in Plant Science Liu 10.3389/fpls.2022.942875 13 2022 TomatoDet: Anchor-free detector for tomato detection Measurement Turečková 10.1016/j.measurement.2022.111977 204 2022 Slicing aided large scale tomato fruit detection and counting in 360-degree video data from a greenhouse Appl. Artif. Intell. Zhang 10.1080/713827180 17 375 2003 Data preparation for data mining Larxel 2020 Glenn 2023 Glenn J Ayush C Jing Q 2023 YOLO by Ultralytics (version 8.0. 0) GitHub https://github.com/ultralytics/ultralytics (accessed on 10 Mar 2024 ) Machine Learning and Knowledge Extraction Terven 10.3390/make5040083 5 1680 2023 A comprehensive review of yolo architectures in computer vision: from yolov1 to yolov8 and yolo-nas Lin 740 2014 Microsoft coco: common objects in context Waxmann 2023 Solawetz 2023 Curtin 82 2019 Deep learning for inexpensive image classification of wildlife on the Raspberry Pi Bhosale 1 2022 IoT deployable lightweight deep learning application for COVID-19 detection with lung diseases using RaspberryPi IEEE Sens. J. Dewangan 10.1109/JSEN.2020.3027097 21 3570 2020 Deep learning-based speed bump detection model for intelligent vehicle system using raspberry Pi Science Swets 10.1126/science.3287615 240 1285 1988 Measuring the accuracy of diagnostic systems arXiv preprint arXiv Powers 2020 Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation Padilla 237 2020 A survey on performance metrics for object-detection algorithms Rezatofighi 658 2019 Generalized intersection over union: a metric and a loss for bounding box regression Eng. Struct. Wu 10.1016/j.engstruct.2022.114962 272 2022 Autonomous surface crack identification of concrete structures based on an improved one-stage object detection algorithm Jin 10.1109/CVPRW59228.2023.00461 4385 2023 Glass wool defect detection using an improved YOLOv5 Agronomy Zhang 10.3390/agronomy13082012 13 2012 2023 DCF-Yolov8: an improved algorithm for aggregating low-level features to detect agricultural pests and diseases NanoCode012 10.5281/zenodo.4154370 2020 ultralytics/yolov5: v3.1 - bug fixes and performance improvements Jocher G Stoken A Borovec J NanoCode012 2020 ultralytics/yolov5: v3.1 - bug fixes and performance improvements Zenodo (accessed on 5 Mar 2024) 10.5281/zenodo.4154370 Zhao 16965 2024 Detrs beat yolos on real-time object detection Wang 1 2025 Yolov9: learning what you want to learn using programmable gradient information