This repository presents a code that aimed to add a new Image classifier, using Random Forests to Levis an Image classification software coded by Mines Paristech's Robotics lab : CAOR
The package is composed of two sub_modules :
- hog : Histogram of Oriented Gradients
- randomForests : Random Forest Calssifier
This sub-module handles the images preprocessing, it takes images as inputs and returns the associated Histogram of Oriented Gradients
this folder contains the follower java files :
- HOG : Realize the preprocessing
- HOGPARAM : stores and returns the parameters of the trees
- Tree : Builds and store a tree
As the name suggests it, this sub-module builds and trains the Random Forest.
this folder contains the follower java files :
- RandomForest : Builds and Store the random Forest
- RandomForestLearner : class that trains and update the Random Forest
- RF : root class for RandomForest handles simple methods to leave only machine learning related methods to the RandomForest class.