---

eye_locator.cpp File Reference

#include "common.h"
#include "eye_locator.h"
#include <dirent.h>
#include <stdlib.h>
#include <math.h>

Functions
void	getEyeInputData (GrayImage &image, int x, int y, double *input_pixels)
bool	eye_training_done (annie::TrainingSet &ts, annie::TwoLayerNetwork &other_net)
void	trainEyeNN (string NonEyeDir, annie::TrainingSet &ts, string fileName, int hidden)
annie::TrainingSet *	make_eye_training_set (string EyeDir, string fileName)

---

Function Documentation

bool eye_training_done (  annie::TrainingSet &    ts, annie::TwoLayerNetwork &    other_net )

void getEyeInputData (  GrayImage &    image, int    x, int    y, double *    input_pixels )

this is a special function that will do all of the nessisary image preprocessing for the neural network eye locator. This is based off of ImageMagick`s histogram equalization code Parameters: image  the image to get this info from x  the x value of the upper left hand corner of the image to be processed y  the y value of the upper left hand corner of the image to be processed input_pixels  a pointer to the array of input values to the neural network, note that this is how values are returned Exceptions: string

annie::TrainingSet* make_eye_training_set (  string    EyeDir, string    fileName )

the first step in training the network is to compile all of the eyes into a training set to train the network with Parameters: EyeDir  a directory containing image files of eyes. the eyes should be EYE_WIDTH by EYE_HEIGHT fileName  the name of the file to save the training set data to Returns: true if no errors were encountered else false

void trainEyeNN (  string    NonEyeDir, annie::TrainingSet &    ts, string    fileName, int    hidden )

train the network in an interesting way

---