Iot-Assisted Deep Learning Approach For Tomato Leaf Disease Monitoring In Greenhouses Using Faster R-Cnn

Abstract

Diseases Affecting Tomato Leaves Significantly Prevent The Yield Of Global Agriculture. Quick And Accurate Identification Is Necessary To Reduce Crops. Conventional Manual Examination Is Laborious And Susceptible To Errors, While Previous Automated Techniques Often Show Deficiencies In Robustness And Scalability. This Research Introduces A Deep Framework Combined With A Greenhouse With The Support Of Iot For Continuous Plant Health Assessment. The Pictures Of The Sheets Obtained In The Controlled Settings Are Investigated Using A Faster R-CNN With The Spine Cspresnet-50, Making It Easier To Extraction Of Elements And Accelerated Convergence. Compared To Conventional CNN, Our Methodology Is Achieved By Higher
Detection, Reduced Detection Of False Detection And Increased Mean Average Precision (Map), Which Facilitates Reliable Identification Of The Early Phase Disease. This Technology Shows Scalability And Efficiency For Intelligent Agricultural Applications.