Weed Detection in Soybean Crop Using Deep Neural Network

The problematic and undesirable effects of weeds lead to degradation in the quality productivity yields. These unacceptable are close competitors crops as they constantly devour water, air, nutrients, sunlight which helpful for maturation crops. For better cultivation good production crops, weed detection at appropriate time is an essential stride. In recent years, various state-of-the-art (SOTA) architectures were proposed detect among crop yields, but lacked computational cost. This paper mainly focuses on proposing a customized architecture comparative study with transfer learning models detecting classifying soybean by concentrating low selected SoTA beneficial large scale very costs. terms selection, Maximum Validation Accuracy (MVA), Least Cross-Entropy Loss (LVCEL), Training Time (TT) considered objective function value system. total, 15 CNNs 18 Transfer analyzed help metric evaluations finding best optimal classification. Experimentation analysis resulted C13 being robust outperformed every model achieving highest accuracy 0.9458 5.9335 ROC-AUC 0.9927 classification from