Pest Prediction in Rice using IoT and Feed Forward Neural Network

Abstract

Rice is a cereal grain, and in its domesticated 
form is the staple food for over half of the world’s human 
population. Rice is the seed of the grass species Oryza sativa 
(Asian rice) or, much less commonly, O. glaberrima (African rice). 
It is cooked by boiling, or it can be ground into flour. It is 
eaten alone and in a great variety of soups, side dishes, and 
main dishes in Asian, Middle Eastern, and many other cuisines. 
Other products in which rice is used are breakfast cereals, 
noodles, and such alcoholic beverages as Japanese sake. Rice has 
become commonplace in many cultures worldwide; in 2021, 787 
million tons were produced, placing it fourth after sugarcane, 
maize, and wheat. Stem borers are moths that attack rice crops. 
Sam Kattiveettil James 
Dept. of Computer Sscience 
Amal Jyothi College of Engineering 
Kanjirapally, Kerala, India 
samkattiveettiljames2024@cs.ajce.in
 striped stemborer, gold-fringed stemborer, dark-headed striped 
stemborer, and the pink stemborer. 
They feed upon tillers and causes deadhearts or drying of the 
central tiller, during vegetative stage and causes whiteheads at 
reproductive stage. Environmental factors such as relative 
humidity, rainfall, and temperature can influence the growth of 
stem borers in rice fields. This study aims to identify specific 
changes in environmental conditions, such as temperature, 
humidity, and rainfall, that may trigger outbreaks of stem borers. 
By pinpointing these factors, the study aids in identifying hotspots 
of insect pests in rice fields and provides insights for farmers. Our 
proposed system is a machine learning model which takes in data 
from temperature, humidity and rainfall sensors in fields and uses 
it to make predictions, whether pest attack will occur or not, so 
that necessary precautions can be taken.