CROP DISEASE NOTIFICATION SYSTEM
Amit Dhage1, Prajwal Wandhare1, Sayali Dhole1, Vaishnavi Nighot1, Prof. Rashmi Kale2
1Student, Department of Computer Science Engineering
2Assistant Professor, Department of Computer Science Engineering
Smt. Kashibai Navale College of Engineering, Pune, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Crop sicknesses are a primary danger to food security; however, their fast identification remains tough in lots of elements of the sector because of the shortage of the essential infrastructure. The aggregate of growing worldwide smartphone penetration and current advances in computer imaginative and prescient made viable with the aid of deep getting to know has paved the way for cellphone-assisted sickness prognosis. The use of a public dataset of 20,306 photos of diseased and healthy plant leaves accumulated underneath controlled situations, we educate a deep convolutional neural network to pick out 15 crop species and 29 diseases (or absence thereof). The educated model achieves an accuracy of 85.35% on a held-out check set, demonstrating the feasibility of this approach. Universal, the method of schooling deep learning models on an increasing number of huge and publicly to be had image datasets gives a clean route in the direction of phone-assisted crop ailment diagnosis on a massive worldwide scale. Notifications offer a completely unique mechanism for increasing the effectiveness of actual-time facts transport systems. However, notifications that demand farmers’ attention at inopportune moments are more likely to have destructive effects and may become a motive of capability disruption in place of proving beneficial to farmers. In order to address these demanding situations a spread of notification mechanism based on tracking and gaining knowledge of crop disease behavior were proposed. The goal of such mechanism is maximizing farmers receptiveness to the added records by means of routinely inferring the proper crop and the proper fertilizers, for assuring accurate yield of crops.
Key Words: Smartphone, Notifications, Crop Disease, Deep Convolutional Neural Network.
