OCR with perceptron neural network of Aforge.net answers wrong

Question

I tried to make OCR by perceptrons with Aforge.Net in C#. I learned my network with nine 30*30 pictures in binary. But in the results, it recognizes everything as 'C'. this is the code:

private void button1_Click(object sender, EventArgs e) { AForge.Neuro.ActivationNetwork network = new AForge.Neuro.ActivationNetwork(new AForge.Neuro.BipolarSigmoidFunction(2), 900, 3); 

network.Randomize(); 

AForge.Neuro.Learning.PerceptronLearning learning = new AForge.Neuro.Learning.PerceptronLearning(network); 

learning.LearningRate =1 ; 

double[][] input = new double[9][]; 

for (int i = 0; i < 9; i++) 

{ 

input[i] = new double[900]; } //Reading A images 

for (int i = 1; i <= 3; i++) 

{ 

Bitmap a = AForge.Imaging.Image.FromFile(path + "\\a" + i + ".bmp"); 

for (int j = 0; j < 30; j++) 

for (int k = 0; k < 30; k++) 

{ 

if (a.GetPixel(j, k).ToKnownColor() == KnownColor.White) 

{ 

input[i-1][j * 10 + k] = -1; } 

else input[i-1][j * 10 + k] = 1; } // showImage(a); } 

//Reading B images 

for (int i = 1; i <= 3; i++) 

{ Bitmap a = AForge.Imaging.Image.FromFile(path + "\\b" + i + ".bmp"); 

for (int j = 0; j < 30; j++) 

for (int k = 0; k < 30; k++) 

{ if (a.GetPixel(j , k).ToKnownColor() == KnownColor.White) { input[i + 2][j * 10 + k] = -1; } 

else input[i + 2][j * 10 + k] = 1; } // showImage(a); } //Reading C images for (int i = 1; i <= 3; i++) 

{

 Bitmap a = AForge.Imaging.Image.FromFile(path + "\\c" + i + ".bmp"); 

for (int j = 0; j < 30; j++) for (int k = 0; k < 30; k++) { if (a.GetPixel(j , k ).ToKnownColor() == KnownColor.White) 

{ 

input[i + 5][j * 10 + k] = -1; } 

else input[i + 5][j * 10 + k] = 1; } // showImage(a); } 

bool needToStop = false; 

int iteration = 0; 

while (!needToStop) { 

double error = learning.RunEpoch(input, new double[9][] 

{ new double[3] { 1, -1, -1 },new double[3] { 1, -1, -1 },new double[3] { 1, -1, -1 },//A new double[3] { -1, 1, -1 },new double[3] { -1, 1, -1 },new double[3] { -1, 1, -1 },//B new double[3] { -1, -1, 1 },new double[3] { -1, -1, 1 },new double[3] { -1, -1, 1 } }//C /*new double[9][]{ input[0],input[0],input[0],input[1],input[1],input[1],input[2],input[2],input[2]}*/ ); //learning.LearningRate -= learning.LearningRate / 1000; 

if (error == 0) break; 

else if (iteration < 1000) iteration++; 

else needToStop = true; 

System.Diagnostics.Debug.WriteLine("{0} {1}", error, iteration); } 

Bitmap b = AForge.Imaging.Image.FromFile(path + "\\b1.bmp"); //Reading A Sample to test Netwok double[] sample = new double[900]; 

for (int j = 0; j < 30; j++) 

for (int k = 0; k < 30; k++) 

{ if (b.GetPixel(j , k ).ToKnownColor() == KnownColor.White) { sample[j * 30 + k] = -1; } 

else sample[j * 30 + k] = 1; } 

foreach (double d in network.Compute(sample)) System.Diagnostics.Debug.WriteLine(d);//Output is Always C = {-1,-1,1} }

I really wanted to know why it answers wrong. 

Answer 1

Your calculation for loading your initial 30x30 images into a double[900] array in the input structure is wrong. You need to change 'j * 10 + k to j * 30 + k' or you will get invalid results. Later you can use the correct calculation of the offset while loading the test image, which is why it's not being matched correctly against the corrupted samples.

You should write a function that helps to load a bitmap into a double[900] array and call it for each image, instead of writing the same code multiple times. This helps to reduce problems like this, where different results are given by two pieces of code that should return the same result.

If you want to know more about Neural Network visit this Neural Network Tutorial.