Towards deep learning based smart farming for intelligent weeds management in crops

Frontiers in Plant Science

Plant Science

Muhammad Naveed Tahir¹  Muhammad Ali Saqib²  Yaser Hafeez²  Muhammad Aqib³ 

[1] Department of Agronomy, Pir Mehr Ali Shah (PMAS)-Arid Agriculture University Rawalpindi, Rawalpindi, Punjab, Pakistan;Pilot Project for Data Driven Smart Decision Platform for Increased Agriculture Productivity, Pir Mehr Ali Shah (PMAS)-Arid Agriculture University Rawalpindi, Rawalpindi, Punjab, Pakistan;University Institute of Information Technology (UIIT), Pir Mehr Ali Shah (PMAS)-Arid Agriculture University Rawalpindi, Rawalpindi, Punjab, Pakistan;University Institute of Information Technology (UIIT), Pir Mehr Ali Shah (PMAS)-Arid Agriculture University Rawalpindi, Rawalpindi, Punjab, Pakistan;National Center of Industrial Biotechnology, Pir Mehr Ali Shah (PMAS)-Arid Agriculture University Rawalpindi, Rawalpindi, Punjab, Pakistan;

关键词: artificial intelligence;    digital agriculture;    object detection;    weed management;    YOLO;   

DOI  :  10.3389/fpls.2023.1211235

 received in 2023-04-24, accepted in 2023-06-29,  发布年份 2023

来源: Frontiers

【 摘 要 】

IntroductionDeep learning (DL) is a core constituent for building an object detection system and provides a variety of algorithms to be used in a variety of applications. In agriculture, weed management is one of the major concerns, weed detection systems could be of great help to improve production. In this work, we have proposed a DL-based weed detection model that can efficiently be used for effective weed management in crops. MethodsOur proposed model uses Convolutional Neural Network based object detection system You Only Look Once (YOLO) for training and prediction. The collected dataset contains RGB images of four different weed species named Grass, Creeping Thistle, Bindweed, and California poppy. This dataset is manipulated by applying LAB (Lightness A and B) and HSV (Hue, Saturation, Value) image transformation techniques and then trained on four YOLO models (v3, v3-tiny, v4, v4-tiny).Results and discussionThe effects of image transformation are analyzed, and it is deduced that the model performance is not much affected by this transformation. Inferencing results obtained by making a comparison of correctly predicted weeds are quite promising, among all models implemented in this work, the YOLOv4 model has achieved the highest accuracy. It has correctly predicted 98.88% weeds with an average loss of 1.8 and 73.1% mean average precision value.Future workIn the future, we plan to integrate this model in a variable rate sprayer for precise weed management in real time.

【 授权许可】

Unknown   
Copyright © 2023 Saqib, Aqib, Tahir and Hafeez