Can I use IP Cam application to recognize license plates?

Question

Diec Thuan 2021-9-7

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1448079-can-i-use-ip-cam-application-to-recognize-license-plates

评论： Walter Roberson 2021-9-7

Hey guy, I use IP Cam on Xiaomi Redmi Note 9 Pro

I have a code for license plate recognition, but when I take a photo, I can't recognize it, any new ideas?

%chup hinh xe ra va nhan dien bien so xe ra
axes(handles.axes5);
load ('imgfildata.mat')
load('bien.mat')
load('bien1.mat')
load imgfildata;
% cao=webcam;
% preview(cao);
% vid= handles.vid;
camm1 = ipcam('http://192.168.0.102:8080/video');
preview(cam1);
e = getsnapshot(cam1);
imshow(e);
e=imresize(e,[300 500]);
[~,cc]=size(e);
if size(e,3)==3
  e=rgb2gray(e);
end
% se=strel('rectangle',[5,5]);
% a=imerode(picture,se);
% figure,imshow(a);
% b=imdilate(a,se);
threshold = graythresh(e);
e =~im2bw(e,threshold);
e = bwareaopen(e,30);
% imshow(picture)
if cc>2000
    picture1=bwareaopen(e,3500);
else
picture1=bwareaopen(e,3000);
end
% figure,imshow(picture1)
picture2=e-picture1;
% figure,imshow(picture2)
picture2=bwareaopen(picture2,200);
% figure,imshow(picture2)
[L,Ne]=bwlabel(picture2);
propied=regionprops(L,'BoundingBox');
hold on
pause(1)
for n=1:size(propied,1)
  rectangle('Position',propied(n).BoundingBox,'EdgeColor','g','LineWidth',2)
end
hold off
% figure
final_output=[];
t=[];
for n=1:Ne
  [r,c] = find(L==n);
  n1=e(min(r):max(r),min(c):max(c));
  n1=imresize(n1,[42,24]);
%   imshow(n1)
  pause(0.2)
  x=[ ];
totalLetters=size(imgfile,2);
 for k=1:totalLetters
    
    y=corr2(imgfile{1,k},n1);
    x=[x y];
    
 end
 t=[t max(x)];
 if max(x)>.45
 z=find(x==max(x));
 out=cell2mat(imgfile(2,z));
final_output=[final_output out];
% pathname='C:\Users\LHN\Desktop\Vehicle number plate recognition\test images\test images';
% filename = [pathname num2str(cc+k*(size(e,1))) '.png']; 
% imwrite(e,filename);
end
end
biendem1=biendem1+1; 
save('bien1.mat','biendem1')
if biendem<biendem1
a=clock;
s1=sprintf('%d - %d - %d',a(3),a(2),a(1));
s2=sprintf('%d:%d:%d',a(4),a(5),round(a(6)));
load('dulieu.mat'); 
data{biendem,1}=s1; 
data{biendem,2}=s2;
data{biendem,3}=final_output;
save('dulieu.mat','data');
load('dulieu.mat')
biendem=biendem+1;
save('bien.mat','biendem') 
set(handles.uitable3,'Data',data);data
end

Thank all.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2021-9-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1448079-can-i-use-ip-cam-application-to-recognize-license-plates#answer_782219

在 MATLAB Online 中打开

Generally speaking, if your connection to the ipcam is able to bring in images, and the characters in the images are at least as large as about 9 pixels tall, then it is probably possible to read the license plates. More pixels might be requires if non-English characters need to be read.

However, your code has

if cc>2000
    picture1=bwareaopen(e,3500);
else
picture1=bwareaopen(e,3000);
end

which discards image areas that are less than 3000 pixels in area. It would be common for text to be smaller than that, so you might be throwing away too much when you do that.

I notice that you reshape extracted areas to 42 x 24, which is 1008 pixels in area. That suggests that your cut-off of 3000 or 3500 pixels in area might be stricter than is needed.

2 个评论
显示无隐藏无

Diec Thuan 2021-9-7

When I press the button to take a photo, it can only take pictures but not recognize, any new ideas?

Walter Roberson 2021-9-7

在 MATLAB Online 中打开

It should not matter, but you can write your code better.

Change

  x=[ ];
totalLetters=size(imgfile,2);
 for k=1:totalLetters
    
    y=corr2(imgfile{1,k},n1);
    x=[x y];
    
 end
 t=[t max(x)];
 if max(x)>.45
 z=find(x==max(x));
 out=cell2mat(imgfile(2,z));
final_output=[final_output out];
% pathname='C:\Users\LHN\Desktop\Vehicle number plate recognition\test images\test images';
% filename = [pathname num2str(cc+k*(size(e,1))) '.png']; 
% imwrite(e,filename);
end

to

totalLetters = size(imgfile, 2);
x = zeros(1,totalLetters);
for k = 1 : totalLetters
   x(k) = corr2(imgfile{1,k},n1);
end
[maxx, z] = max(x);
t(n) = maxx;
if maxx > 0.45
   final_output{n} = imgfile{2,z};
end