[image detection] recognition of the front and back of a coin based on texture features with matlab code attached

1 brief introduction

In the banking industry, self-service industry and other financial service industries involving the identification of paper money , The paper money sorter plays a very important role , It has replaced a lot of boring and error prone manual work , And it has greatly improved the speed and efficiency of this business . However, there is still a big gap in the field of banknote sorter manufacturing between China and other countries with advanced banknote sorter manufacturing technology , China's large banks and other financial institutions use almost all the banknote sorters imported from abroad , The price is more than one million RMB . The denomination recognition of banknotes is the core function of the banknote sorter , Therefore, it is necessary to break the monopoly of foreign countries in China's sorter market , It is necessary to develop efficient paper currency identification technology with independent intellectual property rights . After fully investigating the domestic and foreign RMB banknote recognition market and widely reading the relevant literature , This paper presents a method for detecting the front and back images of RMB banknotes .​

2 Complete code

clc,clear,close all;

% The drawing reading regulations are 200*400 size

f1=imresize(imread('5.jpg'),[200 400]);

f2=imresize(imread('5f.jpg'),[200 400]);

%imshow(f);

%figure,imshow(f1);

%figure,imshow(f2);

% take rgb Turn to grayscale

%g1=rgb2gray(f);figure,imshow(g1);

%g2=rgb2gray(ff);figure,imshow(g2);

% look for g Of Graphs h and w

for n=1:2

    a=strcat('f',int2str(n));

    aa=eval(a);

    aa1=rgb2gray(aa);

    [h,w]=size(aa1);

    % take g The picture is divided into 9 block , high h Three equal parts

    h1=1:round(h/3);

    h2=round(h/3):round(h*2/3);

    h3=round(h*2/3):h;

    %w according to 1:2:1 branch

    w1=1:round(w/4);

    w2=round(w/4):round(w*3/4);

    w3=round(w*3/4):w;

    %9 The size of a small piece

    g1=aa1(h1,w1);g2=aa1(h2,w1);g3=aa1(h3,w1);

    g4=aa1(h1,w2);g5=aa1(h2,w2);g6=aa1(h3,w2);

    g7=aa1(h1,w3);g8=aa1(h2,w3);g9=aa1(h3,w3);

    % Define a 9 All zero matrix with row and column

    m=zeros(9,1);

    figure;

    % Do a cycle

    for i=1:9

        % adopt strcat Concatenate the contents before and after commas （ take int Integer to str String type ）

        s=strcat('g',int2str(i));

        % Execute string statements as statements , decimal

        xx=eval(s);

        % count xx Average value

        mx=mean2(xx);

        %m The number of lines is mx value

        m(i,1)=mx;

        subplot(3,3,i);

        imshow(xx);

    end

    %m The maximum of ,k Is the number of components

    [c,k]=max(m);

    s=strcat('f',int2str(n));

    xxx=eval(s);

    figure,imshow(xxx);

    if(k==2||k==8)        

        title(' positive ')

    end

    if(k==9||k==1)        

        title(' The reverse ')

    end

end

3 Simulation results

4 reference

[1] Peng Hui . Image classification and recognition based on texture features [J]. Computer and information technology , 2007(10):3.

About bloggers ： Good at intelligent optimization algorithms 、 Neural networks predict 、 signal processing 、 Cellular automata 、 The image processing 、 Path planning 、 UAV and other fields Matlab Simulation , relevant matlab Code problems can be exchanged by private letter .