Robust Approach for Detection and Identification of Pest Diseases in Green Farming

Abstract

Plants are an important part of our eco system. They play a vital role in human lives, as they are being used as source of food, medicine, resins, oils, and many other products. With the rising demand for quality fruits and vegetables on the market, pest assaults on plants and their products are becoming more common. Pakistan is an agricultural country, and our economy directly depends on agricultural lands and products. Pest attacks are on the rise as market capacity grows, causing farmers to lose a lot of money. A system should be developed to detect and identify the pests which cause a great harm to crop and vegetables. To overcome these situations, we have developed an application that can help farmers and home gardeners to identify what type of pest infest their plants, orchards, crops, vegetables, fruits, and plant’s stem. This thesis provides complete details about the approaches used to develop an application that can detect and identify pests using Image Processing techniques (IPTs) and Machine Learning Algorithm. The mobile application is designed in such manner that it can be used by farmers in agricultural fields and home gardeners. The mobile application provides a user friendly and easy to operate user interface, through which user can register him/herself and can identify and detect pest using phone’s camera. The application can tell the type of pest and name of pesticide for that pest to avoid further infestation on plant. In this digital era almost, every business has shifted towards digitalization and have reduced human workload and resources. Farmers can now examine what type of insect poses a threat to their fields and orchards with this app, and green farming can be improved.