Weed_Classification

This research paper presents a baseline approach for classifying weeds growing among millet crops. The dataset used in this study comprises a total of 754 images of weeds and millets with clear and wild backgrounds, primarily grown in Chhattisgarh, India. State- of-the-art deep learning models, including MobileNet, InceptionV3, Resnet152v2, and VGG19, were employed to evaluate the dataset's accuracy. Among these models, MobileNet demonstrated superior performance, achieving a training accuracy of 92.86% when the images where captured in wild background. These results highlight the versatile applications of deep learning models in weed detection.