Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Early Diagnosis and Severity Assessment of Weligama Coconut Leaf Wilt Disease and Coconut Caterpillar Infestation using Deep Learning-based Image Processing Techniques

Samitha Vidhanaarachchi, Janaka L. Wijekoon, W. A. Shanaka P. Abeysiriwardhana, Malitha Wijesundara

TL;DR

The paper tackles the urgent need for early diagnosis of Weligama Coconut Leaf Wilt Disease (WCWLD) and Coconut Caterpillar Infestation (CCI) to protect coconut yields. It introduces a deep learning–driven pipeline that combines transfer learning CNNs for WCWLD detection and severity assessment, Mask R-CNN for CCI instance segmentation and progression, and YOLO for accurate caterpillar counting, validated on Sri Lankan field data. Results show WCWLD identification at ~90% accuracy, CCI at ~95% accuracy, severity at ~97% accuracy, and YOLO-based caterpillar counting with a mean Average Precision (mAP) of 96.87% (YOLOv5), outperforming image-processing approaches. The work provides a practical pathway for early intervention in coconut plantations and contributes a publicly accessible dataset on Kaggle to support further research, with plans to extend to additional regions and incorporate explainable AI.

Abstract

Global Coconut (Cocos nucifera (L.)) cultivation faces significant challenges, including yield loss, due to pest and disease outbreaks. In particular, Weligama Coconut Leaf Wilt Disease (WCWLD) and Coconut Caterpillar Infestation (CCI) damage coconut trees, causing severe coconut production loss in Sri Lanka and nearby coconut-producing countries. Currently, both WCWLD and CCI are detected through on-field human observations, a process that is not only time-consuming but also limits the early detection of infections. This paper presents a study conducted in Sri Lanka, demonstrating the effectiveness of employing transfer learning-based Convolutional Neural Network (CNN) and Mask Region-based-CNN (Mask R-CNN) to identify WCWLD and CCI at their early stages and to assess disease progression. Further, this paper presents the use of the You Only Look Once (YOLO) object detection model to count the number of caterpillars distributed on leaves with CCI. The introduced methods were tested and validated using datasets collected from Matara, Puttalam, and Makandura, Sri Lanka. The results show that the proposed methods identify WCWLD and CCI with an accuracy of 90% and 95%, respectively. In addition, the proposed WCWLD disease severity identification method classifies the severity with an accuracy of 97%. Furthermore, the accuracies of the object detection models for calculating the number of caterpillars in the leaflets were: YOLOv5-96.87%, YOLOv8-96.1%, and YOLO11-95.9%.

Early Diagnosis and Severity Assessment of Weligama Coconut Leaf Wilt Disease and Coconut Caterpillar Infestation using Deep Learning-based Image Processing Techniques

TL;DR

The paper tackles the urgent need for early diagnosis of Weligama Coconut Leaf Wilt Disease (WCWLD) and Coconut Caterpillar Infestation (CCI) to protect coconut yields. It introduces a deep learning–driven pipeline that combines transfer learning CNNs for WCWLD detection and severity assessment, Mask R-CNN for CCI instance segmentation and progression, and YOLO for accurate caterpillar counting, validated on Sri Lankan field data. Results show WCWLD identification at ~90% accuracy, CCI at ~95% accuracy, severity at ~97% accuracy, and YOLO-based caterpillar counting with a mean Average Precision (mAP) of 96.87% (YOLOv5), outperforming image-processing approaches. The work provides a practical pathway for early intervention in coconut plantations and contributes a publicly accessible dataset on Kaggle to support further research, with plans to extend to additional regions and incorporate explainable AI.

Abstract

Global Coconut (Cocos nucifera (L.)) cultivation faces significant challenges, including yield loss, due to pest and disease outbreaks. In particular, Weligama Coconut Leaf Wilt Disease (WCWLD) and Coconut Caterpillar Infestation (CCI) damage coconut trees, causing severe coconut production loss in Sri Lanka and nearby coconut-producing countries. Currently, both WCWLD and CCI are detected through on-field human observations, a process that is not only time-consuming but also limits the early detection of infections. This paper presents a study conducted in Sri Lanka, demonstrating the effectiveness of employing transfer learning-based Convolutional Neural Network (CNN) and Mask Region-based-CNN (Mask R-CNN) to identify WCWLD and CCI at their early stages and to assess disease progression. Further, this paper presents the use of the You Only Look Once (YOLO) object detection model to count the number of caterpillars distributed on leaves with CCI. The introduced methods were tested and validated using datasets collected from Matara, Puttalam, and Makandura, Sri Lanka. The results show that the proposed methods identify WCWLD and CCI with an accuracy of 90% and 95%, respectively. In addition, the proposed WCWLD disease severity identification method classifies the severity with an accuracy of 97%. Furthermore, the accuracies of the object detection models for calculating the number of caterpillars in the leaflets were: YOLOv5-96.87%, YOLOv8-96.1%, and YOLO11-95.9%.

Paper Structure

This paper contains 21 sections, 11 equations, 12 figures, 4 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Background Survey
  3. Materials and methods
  4. Data collection
  5. Data Preprocessing
  6. Data Annotation
  7. Classify WCLWD and assess severity level
  8. Identify and classify CCI
  9. Calculate CCI progression level
  10. Detect and count Coconut caterpillars
  11. Counting caterpillars using image processing
  12. Counting caterpillars using object detection (YOLO)
  13. Test setup implementation, results and discussion
  14. Evaluation of WCLWD identification and classification
  15. Evaluation of CCI identification and classification
  16. ...and 6 more sections

Figures (12)

  • Figure 1: Symptoms of WCLWD. Images include (a) Flaccidity, (b) Uneven yellowing, (c) Drying of the leaflets, and (d) Breaking off the leaf tips.
  • Figure 2: The training procedure of Mask R-CNN model for CCI instance segmentation
  • Figure 3: Confusion matrices WCLWD
  • Figure 4: Loss and accuracy plots: a. Loss curve for WCLWD classification, b. Accuracy curve for WCLWD classification, c. Loss curve for symptom severity, d. Accuracy curve for symptom severity
  • Figure 5: Process of calculating the extent of damage: a. Original image, b. Masking, c. Crop segmentation, d. Color segmentation
  • ...and 7 more figures