Hello everyone, I hope you’re all doing well!
My name is William Cardoso, and I have a degree in Computer Science. This is my first post on the forum, and I’d like to share a project I’m working on and trying to improve.
Project Overview
My main challenge is counting small black dots, which are Aedes mosquito eggs. These eggs are deposited on an Aucatex panel, a material placed in water where mosquitoes lay their eggs.
Image Characteristics
- Captured using a high-quality scanner
- File sizes range from 4MB to 5MB
urrent Approach
My method involves dividing the image into 512x512 pixel squares and then applying a neural network to count the eggs. However, I recently started exploring Scikit-Image, and I’m fascinated by its potential. I now believe I can achieve similar results using image processing techniques instead of deep learning.
Seeking Advice
I would love to hear your thoughts on:
- Image processing techniques that could be effective for counting these eggs.
- How to optimize the neural network approach if I continue using deep learning.
- Any resources or methods I should study to improve accuracy.
I’m happy to provide more
details if needed. Looking forward to your suggestions!
original image
Example square 512x512
