• Ms.Mayuri SOni
• Mr. Rajneesh Pachouri
• Mr. Anurag Jain

Weeds are a significant challenge in modern agriculture, leading to reduced crop yields and increased resource competition. Traditional weed management methods, including manual weeding and indiscriminate herbicide application, are labor-intensive, time-consuming, and often environmentally harmful. This thesis presents a novel approach to weed and crop image classification using a deep learning-based system, employing the ResNet50 Convolutional Neural Network (CNN) architecture. The proposed system aims to automate the process of distinguishing between weeds and crops in agricultural fields, enhancing precision and efficiency in weed management. By leveraging the ResNet50 model, pre-trained on large datasets and fine-tuned for the specific task, the system achieves a remarkable classification accuracy of 99.12%. The integration of advanced preprocessing techniques ensures the uniformity of input data, further improving model performance. This work contributes to the advancement of agricultural technology by providing an automated, scalable, and environmentally sustainable solution for weed management, setting a foundation for future innovations in precision farming.

Weed Classification, Crop Identification, Deep Learning in Agriculture, ResNet50 Model, Precision Farming, Automated Weed Management

"Enhancing Agricultural Productivity Through Automated Weed and Crop Classification Using ResNet50 ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.g369-g374, January-2025, Available :http://www.jetir.org/papers/JETIR2501639.pdf

"Enhancing Agricultural Productivity Through Automated Weed and Crop Classification Using ResNet50 ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppg369-g374, January-2025, Available at : http://www.jetir.org/papers/JETIR2501639.pdf