Search

Automatic prediction of diseases in plants is a challenging task with technology intervention. In addition, proper maintenance of plant growth requires several steps, including research on environmental factors and management of water supply for proper plant growth. Traditional irrigation procedures stay inefficient additionally unpredictable. Every year, about 18% of the yield is lost to pests around the world. Identifying plant viruses remain the key to avoiding yield losses on agricultural products that are difficult to do manually. This paper presents an architecture that allows users to perform prediction of plant diseases using machine learning and climatology based data in real time, allowing farmers to view farm details automatically without manual intervention. Disease detection of visually visible symptoms of plants using image processing is proposed and treatment to reduce the level of damage is proposed to farmers. Therefore, the proposed system will increase productivity and benefit the irrigation sector. The main problem facing farmers in agriculture is the lack of water due to the effective irrigation separators on small farms. However, this technique consumes several weaknesses that can be improved, a method called the Automatic Irrigation System with weather forecasts for the efficient use of water resources. This overcomes the shortcomings of the AISPF process. The method uses available aquatic assets extra capacity by measuring atmospheric pressure, or barometric pressure (in inches), apart from measuring the water present-day in the topsoil and actually predicting the weather. This system uses a soil moisture sensor and a pneumatic sensor for performing measurements to be processed by the algorithm and water is discharged accordingly. This enables efficient use of water automatically. The system provides better accuracy than state of the art techniques for plant disease prediction and water flow control, which could be useful for irrigation.