← Selected Work
Computer Vision · Deep Learning
Plant Disease Detection
An earlier deep-learning project exploring computer vision for greenhouse monitoring and plant disease detection.
CNNImage ClassificationGreenhouse Monitoring
System flow
01Leaf Image
02CNN Classifier
03Disease Prediction
Metrics
- Task
- Multi-class classification
- Model
- CNN
- Domain
- Greenhouse monitoring
Context
An early step in Haben's computer-vision path, before the shift toward efficiency- and language-focused AI work.
Problem
Early identification of plant disease in greenhouse settings typically depends on manual inspection, which is slow and inconsistent at scale.
Why it matters
Automated visual monitoring can catch disease earlier and more consistently than periodic manual checks, reducing crop loss.
Constraints
- Limited labeled image data across disease categories
- Visual similarity between early-stage disease symptoms and healthy variation
Approach
A convolutional neural network was trained to classify leaf images into healthy vs. disease categories, framed as a standard supervised image-classification pipeline.
Experiments
- Baseline CNN architecture trained and evaluated on a labeled leaf-image dataset
Results
- Established a working classification baseline for greenhouse monitoring use cases
Lessons
- This project was an early foundation for the applied computer-vision thinking that later shaped the jump-estimation work