Prediction of Crop and Weed Growth Stages using Neural Network in Machine Learning

This paper presents the structure, usage and assessment of a convolutional neural system-based methodology has been applied for the arrangement of various phenological phases of plants. Our CNN design can consequently characterize distinctive phenological phases of eleven sorts of plants. So as to assess the presentation and productivity of our profound learning-based methodology, an old-style AI approach dependent on physically removed highlights is additionally executed. Textural highlights dependent on GLCM highlights have been removed and joined arrangement of highlights are taken care of into an AI calculation. The arrangement pace of the methodology dependent on physically extricated highlights is contrasted with those of our CNN based methodology. Trial results demonstrate that CNN put together methodology is fundamentally powerful with respect to the eleven sorts of plants we investigated. There are a wide range of ways profound learning can be applied on a dataset relying upon the size of the dataset. There are many research headings that we are intending to take for arranging phenological stages. Future work may comprise of building our own CNN design without any preparation especially prepared for arranging phenological phases of plants, just as trying different things with other pre-prepared CNN models and making sense of an approach to recognize infections in crops.