car license plate character recognition using neural network

sir again i have one big query...iwant to recognize the license plate charactyer using neural network...but really i m too much confuse that how could i find the link between segmented characters and neural network....means i m unable to find syntax of giving segmented characters as i/p of neural...plese ans me..if any idea...please.........any one help me.....

Answer 1

[Geoff]

I have not done plate recognition, and have only dabbled in neural networks, but I would expect that you need to translate your segmented characters into a vector. That means you first need to standardise the width and height of each segment, which will involve resampling. I would also expect some kind of image normalisation.

Then you can supply each segment as a 1-dimensional vector. The number of pixels in that vector is the number of inputs in your neural network. From memory, the neural network expects inputs along the rows, and each individual training case in the columns. So you would insert each segment-vector into a unique column of your input matrix. The corresponding training output vector, of course, is the character that each segment-vector represents.

Something like that, anyway =)

[sneha]

hello sir how can i convert my segmented char into vectors??any matlab syntax?

Answer 2

[Swati]

This one is the code for you,,,of Number plate recognition,,,I have earlier work on this one,,,,in this take a image of car,,then take seperate image of its no plate (for ex,,if no is MP 09-T_456,,,,so take seperate image of M then P,,,then,,,T,,,so on,,,all will b in a single folder n in .bmp format)

may ths one helps you,,,

 {I = imread('car.jpg'); 
I2 = rgb2gray(I);
I4 = im2bw(I2, 0.2);
bw = bwareaopen(I4, 500);
se = strel('disk',15);
bw = imclose(bw,se);
bw = imfill(bw,[1 1]);
[B,L] = bwboundaries(bw,4);
imshow(label2rgb(L, @jet, [.5 .5 .5]))
hold on
for k = 1:length(B)
 boundary = B{k};
 plot(boundary(:,2),boundary(:,1),'w','LineWidth',2)
end

stats = regionprops(L,'Area','Centroid');

for k = 1:length(B)

boundary = B{k};

delta_sq = diff(boundary).^2;    
perimeter = sum(sqrt(sum(delta_sq,2)));

area = stats(k).Area;

metric = 27*area/perimeter^2;

metric_string = sprintf('%2.2f',metric);

if metric >= 0.9 && metric <= 1.1
  centroid = stats(k).Centroid;
  plot(centroid(1),centroid(2),'ko');

    goalboundary = boundary; 
    s = min(goalboundary, [], 1);
    e = max(goalboundary, [], 1);
  goal = imcrop(I4,[s(2) s(1) e(2)-s(2) e(1)-s(1)]); 
  end

text(boundary(1,2)-35,boundary(1,1)+13,...
  metric_string,'Color','g',...
'FontSize',14,'FontWeight','bold');
end
goal = ~goal;
goal(256,256) = 0;
figure;
imshow(goal);
w = imread('P.bmp');
w = ~w;
C=real(ifft2(fft2(goal).*fft2(rot90(w,2),256,256)));
thresh = 240;
figure;
imshow(C > thresh);

[sneha]

thanks swati i m trying on your code...if any query will b generate then will ask u...

[sneha]

hi swati by running your code i got this error,

Error in ==> mathworkrecogchr at 51

C=real(ifft2(fft2(goal).*fft2(rot90(w,2),256,256)));

can u help me plz???

[Archit Save]

i am getting an error...'Undefined function or variable goal'

[Image Analyst]

You probably renamed or misspelled a variable somewhere. What variable is given in the call to imshow() just prior to that? If goal weren't a variable, then the imshow() would have thrown an error before the fft line.

Answer 3

[sneha]

thank you very much swati and geoff..........i will try on both of your ways.....thanks a lot....

Answer 4

[sneha]

hey rakshita can u mail me ur code...i have run your code given above but it gives me errorlik this... Undefined function or variable 'y_cor1'.

Error in ==> mathworklpr at 276 lp=d(y_cor1:x_cor2,x_cor1:y_cor2);

i cant go ahead....i want to recognize charctrs with neural net...plz help

Answer 5

[sneha]

hi now i m succesfully segment license plate chara..now m going 2 recognize them using OCR....in MATLAB.... plz anyone can help me....its urgent...pleaseeee..........

[Derick kundukulam]

did the code work.. even i'm gettin the error C=real(ifft2(fft2(goal).*fft2(rot90(w,2),256,256)));

Answer 6

[Image Analyst]

Have you checked here http://iris.usc.edu/Vision-Notes/bibliography/motion-f726.html#License%20Plate%20Recognition,%20Extraction,%20Analysis? Does neural nets seem to be the most common method?

Answer 7

[Imtiaz Hussain]

this is my research topic plz can you help me?

[Image Analyst]

Sure. We'll try, after you read this link.