Scribd
Upload
2 views1 page

USTC_CSE_POSTER_TEMPLATE22

This document outlines the development of a deep learning system for detecting and classifying insect pests in agriculture fields in Bangladesh. The proposed system achieves around 99% accuracy in identifying harmful insect species using a mobile application. The research employs various machine learning techniques and aims to assist farmers in managing crop pests effectively.

Uploaded by

nawaz0x1scribd
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views1 page

USTC_CSE_POSTER_TEMPLATE22

This document outlines the development of a deep learning system for detecting and classifying insect pests in agriculture fields in Bangladesh. The proposed system achieves around 99% accuracy in identifying harmful insect species using a mobile application. The research employs various machine learning techniques and aims to assist farmers in managing crop pests effectively.

Uploaded by

nawaz0x1scribd
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 1

DEVELOPMENT OF A SYSTEM FOR DETECTION OF INSECTS IN AGRICULTURE FIELDS

In this picture, A farmer using


Introduction mobile app to capture insect
image and detection that
 In Agriculture field, detection of crop pests is a challenging task for farmers. insect pest is harmful or not in
the crops.

10,000 Directly responsible for abolition of


insect's crop yield worldwide.

26 insect pest species occur in the Bangladesh


rice ecosystem.

2
23 rice insect pest species are identified in
Bangladesh.

 This research basically for 2


Bangladesh agriculture fields,
developing a system which is
detect & classified insects in the
crops. Different Types of Insect pest Mobile Application

Objective Result
The proposed system can give the following results:
 To develop a deep learning model that can accurately identify the presence
of insects in agriculture fields in Bangladesh.  Meanwhile, we have successfully trained in TensorFlow model total of 22
insect images of three classes.
 To evaluate the accuracy and reliability of the deep learning model in  Around of Accuracy 99% for 5 classes of insect pests.
detecting insects in real-world agriculture fields in Bangladesh.  Five classes of insects, will be classified and detect, is it harmful or not for
crops.

Methods & Materials


Discussion
 Insect classification with image processing and machine learning techniques.
In this project, different insect datasets will be classified and Detected by
 Insect classification involves various steps to be performed. The flow of applying machine learning and insects detection algorithm, and the results
steps for insect classification is illustrated in Figure will be compared. several Insect images were rescaled, pre-processed and
enhanced To increase the dataset to improve accuracy. Comparison of
classification accuracy different machine learning techniques include ANN,
SVM, KNN, Naive Bayes, and CNN models. Which will be used in this
research through Python programming language.
Image Test
Collection &Train
Split
Conclusions
Image
Augmentation
Insect Validation
Train Image
Image
Test Image  The main goal of this research on detecting harmful insect in
Dataset
Image Bangladesh agriculture field.
Annotation Detection &  This system can be control through mobile app or web, which can help
Train Model
Classification the people who are cultivate corn, rice, soybean, and wheat etc.
Image
Read
Performance
Evaluation References
[1] Insect classification and detection in field crops using modern machine
learning techniques, Published: 26 June 2020 .
[2] Denan Xia, School of Computer Science and Technology, Anhui
University, China. Insect Detection and Classification Based on an Improved
Convolutional Neural Network. Published online 2018 Nov 27.
Materials : [3] Tiago Domingues , Tomás Brandão, Ricardo Ribeiro and João C.
Python Google Colab Ferreira, Insect Detection in Sticky Trap Images of Tomato Crops Using
Machine Learning . Published online: 21 November 2022

You might also like