// THESIS GUIDANCE PORTAL · Computer Science
Machine Learning Based Crop Disease Detection in Nepal
1. Introduction & Problem Statement
Overview: Build a CNN-based image classification system to detect crop diseases in Nepali agricultural context — rice, wheat, potato. Train on local disease datasets.
Background Context (Nepal): Nepal's agriculture sector employs 65% of the population but faces significant crop losses due to disease. Traditional detection methods are slow and require expert knowledge. Machine learning offers an automated, scalable solution accessible to farmers via mobile apps.
2. Research Objectives
- ›Collect and annotate crop disease image dataset from Nepal
- ›Train CNN model (ResNet/MobileNet) for disease classification
- ›Achieve 85%+ accuracy on local dataset
- ›Build mobile-friendly inference interface
- ›Compare performance with traditional detection methods
3. Proposed Methodology
- Literature review on plant disease detection using ML
- Dataset collection from agriculture stations, Kaggle, field visits
- Image preprocessing — augmentation, normalization
- CNN model training with transfer learning
- Model evaluation — precision, recall, F1-score
- Mobile app prototype using TensorFlow Lite
4. Thesis Chapter-by-Chapter Outline
Chapter 1: IntroductionTU/KU standard
Background, problem statement, objectives, scope and limitations, significance of the study
Chapter 2: Literature ReviewTU/KU standard
Review of ML in agriculture, existing disease detection systems, CNN architectures, gap analysis
Chapter 3: Research MethodologyTU/KU standard
Dataset collection, preprocessing pipeline, model architecture, training strategy, evaluation metrics
Chapter 4: System Design & ImplementationTU/KU standard
System architecture, CNN model design, training results, mobile interface design
Chapter 5: Results & DiscussionTU/KU standard
Model performance, accuracy analysis, comparison with existing work, limitations
Chapter 6: Conclusion & Future WorkTU/KU standard
Summary of findings, contributions, recommendations, future improvements
5. Recommended Tools & Technologies
To implement the practical, technical, or analytical portions of this thesis topic, the following software tools, libraries, or APIs are recommended:
6. Core References & Citations
- [1]Mohanty, S.P. et al. (2016). Using Deep Learning for Image-Based Plant Disease Detection. Frontiers in Plant Science.
- [2]Liu, B. et al. (2018). Plant Diseases and Pests Detection Based on Deep Learning. Frontiers in Plant Science.
- [3]NARC Nepal — National Agriculture Research Council datasets