AgRobots (A Combination of Image Processing and Data Analytics for Precision Pesticide Use)

摘要

India is mainly an agricultural country. Agriculture plays a vital role in the Indian economy. Over 70% of the rural households depend on agriculture as their principle means of livelihood. Though there is growth in other sectors, the overall share of agriculture on GDP of the country has decreased. Usually the yield is affected by many factors such as soil moisture, temperature, water salinity and by pests such as insects, fungal/bacteria/viral diseases. This problem caused by pests is overcome by using pesticides. Pesticides are the substance used for destroying insects and other organisms that are harmful for cultivated plants but the excess use of pesticides is harmful not only for surrounding environment but also for the human health. Thus the pesticides must be used in controlled manner. Here this paper mainly deals with mechanism that uses image processing technique to analyze the ill part of the plant and to provide medicine i.e. the pesticide to that part only and not to the entire plant. Initially photos of plant are clicked from different angle and these images are analyzed one by one to check if any part of the plant is attacked by any virus or bacteria. In this analysis if any part of the plant is found infected then pesticide is sprayed only to that part not to the entire plant. This way pesticide used is confined only to infected part and the excess use of pesticide is controlled. Here open CV-Python is used. Deep learning is also employed. Deep learning is a part of a broader family of machine learning based on learning data representation and learning can be semi or fully supervised. Here the general purpose computer used is Raspberry Pi. As the result of this project the amount of pesticide used is reduced by spraying pesticide to only the affected part of the plant.