Investigating Computer Vision and Robotics to Autonomously Identify and Manage Weeds in Crop Fields

Abstract

Eliminating weeds from the crop fields is highly key to the preservation of farming and sustainability. Similarly to traditional methods, they are really labor-demanding and sometimes also can be considered harmful to the surrounding environment. The idea of building a system that includes computer vision and robotics is a persistent prospect for autonomous weed management. Through the computer vision detection algorithms weeds are identified, whereas robots lift them from the tender roots avoiding all damage to the crops. The development stands at producing image samples, imparting algorithms with information and a robotics base. Practical tests for weed detection approved high precision and fast weed management. Fewer chemicals might be seen in the environment and better incomes for farmers are the major impacts. These have size and costs as the major problems. The key priority into the future remains how to improve machine learning, that is, sensor integration and interdisciplinary collaboration of sustainable agriculture towards its improved sustainability.