Unsupervised crop anomaly detection using machine learning for automated crop monitoring

Name: Satida Sookpong

LCSH: Kasetsart University -- Theses. M.Eng. (Artificial Intelligence and Internet of Things) 2022

Classification :.LCCS: TK5105.59

LCSH: Kasetsart University. -- Department of Electrical Engineering -- Theses

LCSH: Anomaly detection (Computer security)

LCSH: Neural networks (Computer science)

LCSH: Machine learning

LCSH: Computer vision

Abstract: In this work, we investigated the use of Generative Adversarial Networks (GANs) for the unhealthy plant identification in crop images for the automatic crop monitoring system. Here, we trained the GAN to generate healthy crop images from a healthy plant image dataset. An unhealthy crop image can be detected by comparing the best-fit healthy crop image generated by GAN, and actually observed image. If the Mahalanobis distance between the observed image and model-generated one is larger than a pre-defined threshold, an image can be considered an unhealthy plant image. The results from the ROC and Precision Recall curves show that our model can detect unhealthy crop images with high accuracy.

Kasetsart University. Office of the University Library

Address: Bangkok

Email: tdckulib@ku.ac.th

Name: Teerasit Kasetkasem

Role: Thesis Advisor

Name: Teera Phatrapornnant

Role: Thesis CO-Advisor

Name: Jaehoon Yu

Role: Thesis CO-Advisor

Created: 2022

Modified: 2025-07-20

Issued: 2025-07-20

วิทยานิพนธ์/Thesis

application/pdf

URL: https://www.lib.ku.ac.th/KUthesis/2564/satida-soo-all.pdf

CallNumber: TK5105.59 .S28

eng

DegreeName: Master of Engineering

Level: Master's Degree

Descipline: Artificial Intelligence and Internet of Things

Grantor: Kasetsart University

©copyrights Kasetsart University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	satida-soo-all.pdf	4.36 MB

ใช้เวลา

0.028247 วินาที