Spontaneous Insect Pest Detection and Identification System (SIPDIS)

Abstract

Insect pests damage crops worldwide. They have a direct impact on agricultural food production by chewing the leaves of crop plants, sucking out plant juices, boring within the roots, stems or leaves, and spreading plant pathogens. They feed on natural fibers, destroy wooden building materials, ruin stored grain, and accelerate the process of decay This problem greatly impacts agricultural products, resulting in low quality and quantity of crops and other agriculture related items like wheat, maize, rice, potatoes, tomatoes etc. It not only destroys the product but also wastes precious time and farmer empowerment. This issue is very critical and has a direct impact on reducing the economy and food security worldwide. However, due to a large number of insect pests and their types throughout the globe we need agriculture domain experts to easily identify these insect pests. It is important because they then must use the right pesticide or chemical to get rid of them. In this project, our aim to is to optimize and manage the agricultural crops from insects and pests by developing an automatic insect pest detection and identification system. This system will be an application that can easily be used by the farmers and other relevant people of this field to identify the pest using captured image through their smart devices. This application will be built using Artificial Intelligence and deep learning techniques. The identification and classification will be done using convolution neural network-based models of deep learning.