A Recognition Model For Identification And Classification Of Insect Species Using Machine Learning

Abstract

XIII newlineAbstract newlineInsect pest infestation is a significant problem faced by the agriculture sector newlineworldwide, resulting in crop damage, quality degradation, and global economic newlinesetbacks. For better diagnosis, computer vision and machine learning models can be newlineused to accurately identify the different insect pest. In our research different fruit fly newlinespecies such as Bactrocera zonata, Bactrocera dorsalis, Zeugodacus cucurbitae, and newlineZeugodacus tau has been identify using machine learning. The specimens were newlinecollected monthly from four different locations of Punjab, the collected species were newlinecounted and sorted up to species level and photographed has been taken for further newlinestudies. Fresh lures were replenished every two months. Weather parameter data were newlineprovided by the Department of Agronomy at Lovely Professional University, newlinePhagwara to know the effect of abiotic on fruit fly population. The result shows that newlinethe peak activity periods in 2021 and 2022 occurred in August and September, newlinerespectively, with Armaan Nagar and Hardaspur having the highest number of fruit newlineflies in both years of the study newline