Artificial Intelligence in Agriculture: Machine Learning Based Early Detection of Insects and Diseases with Environment and Substance Monitoring Using IoT

Abstract

Agriculture industry plays a crucial role in providing employment and food to the people. The major problem in agriculture industry is the attack of diseases in the plant leaves since the early stage. Machine learning becomes one the most important platforms for the detection of plant disease. The automatic leaf characteristics detection which is essential in monitoring large fields of crops, automatically detects the symptoms of leaf characteristics as soon as they appear on the plant leaves. The issues related to good crop production is varied in different areas based on the fertilizers and its quantity, water supply and its pH value, rainfall distribution at the uneven interval, soil absorption and other issues. Considering all the challenges in the agricultural sector in Oman, the need to introduce new technologies and methods has become urgent, to comply with Oman Vision 2040 which targets to increase the contribution of this sector to the achieve 90% of the overall GDP along with other sectors. Agriculture has always been depending on specific static understanding and actions. In this work, the vegetable crops largely found in and around Muscat Governorate have been listed, and based on image acquisition and pre-processing, the detection of plant disease is performed. The research provides a comprehensive intelligent farming prototype using the machine learning algorithm to detect the disease and the causing factors. Sensors, actuators, control units -Raspberry Pi, database system, AI and machine learning software running in central controlling server are used to provide the desired results.