Enhancing Precision Agriculture with Machine Learning and Image Processing: A Comparative Evaluation of YOLO and RCNN for Weed Identification and Detection
摘要
Agriculture is one of the origins of mortal food in this world. Currently due to the growing population, we need the lesser productive capability of the husbandry to meet the demands. In olden days, humans employed natural styles as a kind of industrial poison, just like they did with cow manure in fields. This resulted in a rise in output sufficient to suit the needs of the populace. However, subsequently, people were permitted to generate more profits through further development. A revolt known as the “Green Revolution” therefore began. After this period operation of deadly venoms as dressings and fungicides have increased to a drastic position. By doing so we were successful in adding productivity, but we've forgotten the damage done to the terrain, which will raise a mistrustfulness in our food on this beautiful earth. So, in this design, we're going to apply a model using image processing and deep literacy that will be suitable to separate whether a crop is weed or not and lets us to spot the dressings on only weed crops and not prompt the surroundings where good crops may be present, by this we can enhance crop yield as we’re scattering the dressings on only weed crops and not on other areas of the field.