AI Researcher & Builder · Dhaka, Bangladesh
Samin Sadaf
I work on applied deep learning, most recently on detecting disease in corn leaves from images. My focus is on models that can be evaluated honestly — documented data, reproducible splits, and results reported as they are.
Current research
All researchThesis · Deep learning
Corn Leaf Disease Detection Using Deep Learning
A comparative study of DenseNet121 and EfficientNet-B0 for classifying corn leaf disease from images, built as my undergraduate thesis at Daffodil International University.
Focus
01
Computer vision for agriculture
Detecting plant disease from leaf imagery, where class imbalance and field conditions make clean benchmarks hard to trust.
02
Transfer learning under constraint
What holds up when compute and labelled data are both scarce — pretrained backbones, augmentation, and honest validation splits.
03
Reproducible evaluation
Documenting datasets, seeds and splits so a reported result can actually be re-run by someone else.
Get in touch
Open to research collaboration and conversations about applied deep learning.
Contact