See_Urchin

A tool made in Python that detects and measures the size of Paracentrotus lividus in its different evolutionary stages.
The tool analyzes pictures, extrapolates the larvaes/eggs and uses a neural network to detect the different stages of the evolutionary process.

Scripts

getLarvae.py

This is the main file. It extracts the images; applies the image processing techniques; classifies the larvae; saves the individual larvae images (as .jpg) and builds a report. Its usage depends on the directory the user provides in the filename variable and the images must be numbered from 0 to the numberOfImages variable value. The destination folder can also be modified by changing the pathName variable.

imgClass.py

This file has the image processing functions in them bundled within a convenient class that allows for easy and intuitive image manipulation.

larvaeFunc.py

This file has the larvae-oriented functions in it, which are mainly for extracting the information from the processed images.

report.py

This file has the functions that write in the report.html file.

Other Files

report.html

This file displays the larvae images and their information organizedly.

Images

The larvae images must follow this model:

Any image that doesnt have dark larvae on a bright background most likely won't work.

Dependencies

● Python;
● OpenCV;
● Numpy;
● Ludwig;
● Pandas.

Scripts developed by Davi Mello, student of UNIVALI and participant of the LIA team.