Abstract: Identification of weeds is a task in agriculture fields for farmers. it is helpful for the farmers to identify the type of weed present in the agricultural lands. This method will save the time and we need some knowledge to identify it. if we see the recent advances in deep learning, weeds has been possible to identify automatically. In this paper we will propose a deep learning-based approach for identification of weeds using a Mass Regional-convolutional neural network (MR-CNN). The proposed system has used large set of datasets for various images and previously trained MR-CNN models for achieving high accuracy of weed species. This model has been trained on different dataset of images collected from different plants and fields. The result shows that the proposed system has different methods in identifying of weeds and high accuracy, it will be more accurate for identification of weeds in agriculture

 
Keywords: weed identification, agriculture, crop yield, image datasets, Deep learning, Machine learning.

| DOI: 10.17148/IARJSET.2023.107129