Early Identification of Major Pest attacks Caused to Crop Loss in Paddy Fields: A Case Study
No Thumbnail Available
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Uva Wellassa University of Sri Lanka
Abstract
Unrecognized or late recognized pest attacks are one of the major problems which lead to crop loss in
paddy cultivation. According to the most recent season harvest Data of Rice research institute, it was
recorded 40% of crop loss in the paddy cultivation in the Hambanthota district. Under a preliminary
survey, it was identified Aphids, Brown-planthoppers, and Thrips were the major pests that caused
the crop loss in paddy in the selected area. Due to the lack of proper knowledge in identification and
lack of timely and accurate information, farmers are struggling to identify and control these pest
attacks in their paddy fields. Due to crop loss, most of them are losing their money, interest, time, and
confidence in paddy cultivation. During the study, domain experts revealed that early identification
and early-stage of controlling these pests can save the majority of the crop loss and save lots of
money which were spent on pesticides in paddy cultivation. This Case study was conducted to
address the issues identified above, in the selected paddy fields in the Gonnoruwa area in
the Hambanthota district. The proposed model use image processing techniques in combination with
Convolutional Neural Networks to detect the pest's attacks in paddy cultivation by analyzing the
symptoms. A set of self-captured images which were labeled with the help of domain experts were
used to build and train the proposed model. The model has achieved 95% of accuracy while
testing. The proposed model will be further improved to identify more pests and disease attacks in the
future while delivering it as a handheld portable device where farmers can use it in real-time in their
paddy fields which will lead to saving their time and money while increasing the paddy yield.
Keywords: Crop loss; Paddy; Pest and diseases; Pest identification model
Description
Keywords
Agriculture, Pest Control, Paddy Fields, Computing and Information Science