what is the best method to do iris recognition?

here is the codes:

function[thv,th,tv]=gen_templateVVV(I)
%function[t,x,xv,xgroup,svalue]=cumsumi(I)
    
    %I=adapthisteq(I);
    d=size(I);
    a=(d(1)/3)*(d(2)/10);
    k=0;l=0;
    
    for i=1:3:d(1)                          %i iterates over the rows in steps of 3.
        for j=1:10:d(2)                     %j iterates over cols in steps of 10
            if(i+3<d(1) && j+10<d(2))       %as long as there exists one more step, i.e, till you don't fall off the edge
                k=k+1;
                s(k,:,:)=I(i:i+2,j:j+9);    %s is a 3d matr, k is the indx num, next 2 are the cell.
                x(k)=mean(mean(s(k,:,:)));  %x(k)is the representative sum of cell no. k
            end
        end    
    end
    
    %vertical grouping
    lv=0;
    xvgroup=zeros(300);
    kk=0;
    
    for kv=1:d(2)/10
        lv=lv+1;                                  %lv denotes group number
        kk=0;
        mm=0;
        for ver=kv:30:kv+870                      %there are total 928 cells to group,we have 100/3=33 rows and 300/10=30 columns so for every column 33/=6(approx.) groups
            if(ver==kv+kk+150)                    %kk is used to determine next group number after successive  5 cells visited. first group=x(1),x(31),x(51)...x(121).
                kk=kk+150;
                lv=lv+1; 
                mm=0;
            end
            xvgroup(lv)=xvgroup(lv)+x(ver);       %xvgroup(lv) has the average of the 5 cells in group lv.
            mm=mm+1;
            xvv(lv,mm)=x(ver);                    %xvv(l,1:5) has the rep values of each of the 5 cells of group lv.
        end
    end
    xvgroup=xvgroup/5;
    
    %vert grouping ends
    %horizontal grouping
    for i=1:5:k                            % make a group out of 5 cells. l denotes the total no of groups.
        if(i+4<k)
            l=l+1;
            xgroup(l)=sum(x(i:i+4))/5;     %xgroup(l) has the average of the 5 cells in group l.
            xv(l,1:5)=x(i:i+4);            %xv(l,1:5) has the rep values of each of the 5 cells of group l.
        end
    end
    
    %horizontal grouping ends
    %Computing Si=Si-1+(Xi-X|) for horizontal
    z=0;
    for i=1:l                              % i iterates over the groups
        for j=1:5                          % j iterates over each cell in the group
            if(j==1)
                m=0;                       %for the first cell, mean=0, ie S0=0
            else
                m=svalue(i,j-1);           %m is Si-1 in paper
            end
            z=z+1;
            svalue(i,j)=m+(xv(i,j)-xgroup(i)); %svalue(i,j) stores the cum sum of jth cell in ith group. Si=Si-1+(Xi-X|)
        end
    end
    %Si=Si-1+(Xi-X|) for horizontal ENDS
    %Computing Si=Si-1+(Xi-X|) for vertical
    for i=1:lv                             % i iterates over the groups
        for j=1:5                          % j iterates over each cell in the group
            if(j==1)
                m=0;                       %for the first cell, mean=0, ie S0=0
            else
                m=svaluev(i,j-1);          %m is Si-1 in paper
            end
            z=z+1;
            svaluev(i,j)=m+(xvv(i,j)-xvgroup(i)); %svalue(i,j) stores the cum sum of jth cell in ith group. Si=Si-1+(Xi-X|)
         end
    end
    %Si=Si-1+(Xi-X|) for Vertical ENDS
    
    %Generating irisode for horizontal
    for i=1:l                              % i iterates over the groups
        [maxg,p]=max(svalue(i,1:5));       % maxg is max value in group, p is its index
        [ming,q]=min(svalue(i,1:5));       % ming is min value in group, q is its index
        if(p>q)
            r=q;s1=p;                      %adjust so that greater of p and q is assigned to s, and the other, to r.
        else
            r=p;s1=q;
        end
        for j=1:5
            if(j>=r && j<=s1)
                prev=j-1;
                if prev==0
                   prev=1;  
                end
                if(svalue(i,j)>=svalue(i,prev)) % cell is on upward slope. set iriscode to 2.
                    iriscode(i,j)=255;
                else
                    iriscode(i,j)=128;          %cell is on downward slope.set iriscode to 1.
                end
            else
                iriscode(i,j)=0;                %cell is not between minidx and maxidx, set iriscode to 0.
            end
        end
    end
    %Generating irisode for horizontal  ENDS
    
    %Generating irisodeV for Vertical
    for i=1:lv                              % i iterates over the groups
        [maxg,p]=max(svaluev(i,1:5));       % maxg is max value in group, p is its index
        [ming,q]=min(svaluev(i,1:5));       % ming is min value in group, q is its index
        if(p>q)
            r=q;s1=p;                       %adjust so that greater of p and q is assigned to s, and the other, to r.
        else
            r=p;s1=q;
        end
        for j=1:5
            if(j>=r && j<=s1)
                prev=j-1;
                if prev==0
                    prev=1;
                end
                if(svaluev(i,j)>=svaluev(i,prev))  % cell is on upward slope. set iriscodeV to 2.
                    iriscodev(i,j)=255;
                else
                    iriscodev(i,j)=128;         %cell is on downward slope.set iriscodeV to 1.
                end
            else
                iriscodev(i,j)=0;               %cell is not between minidx and maxidx, set iriscodeV to 0.
            end
        end
    end
    %Generating irisode for Vertical ENDS
    
    b=1;
    i=1;
    for j=1:5:l                               %j is from 1 to no_of_groups, steps of 5
        th(i,b:b+4)=iriscode(j,:);            %t stores the iriscode  generated for each step.
        if(b+5>d(2)/10)
            i=i+1;
            b=1;
        end
        b=b+5;
    end
    b=1;
    i=1;
    %vert
    
    for j=1:5:lv                               %j is from 1 to no_of_groups, steps of 5
        tv(i,b:b+4)=iriscodev(j,:);            %tv stores the iriscodeV  generated for each step.
        if(b+5>d(2)/10)
            i=i+1;
            b=1;
        end
        b=b+5;
    end
    %vert ends
    ths=size(th);
    tvs=size(tv);
    
    for i=1:ths(1)
        for j=1:ths(2)
            thv(i,j)=th(i,j);                         %thv is the horizontal and vertical combined iris (vertical below horizontal)
        end
    end
    for p=1:tvs(1)
        i=i+1;
        for q=1:tvs(2)
            thv(i,q)=tv(p,q);
        end
    end
end

[Walter Roberson]

This appears to be from https://github.com/chandravadans/IrisRecognition from 2013.

[Rik]

The OP had already edited away the question in 2017. Why the title needed to be changed, why the comments needed to be deleted, or why the username should be changed, I don't know. I will respect the OP's privacy and leave out the original username below, but I would like to voice my opinion that I think that is much more curtesy than they deserve.

The original title seems to be this:

hello, i am a newbie in matlab programming, i found codes for generating template, but i do not know what algorithm that have been implemented in this codes. can anyone tell me is it has specific name or just matlab standart?

Deleted comment:

OP on 25 Jun 2017:

yes sir, and i've emailed him and no response.

[Rena Berman]

(Answers Dev) Restored edit

Answer 1

[Image Analyst]

I have no idea other than it's "Generating irisodeV" from the very poor comments. I guess you don't know what it does either or you wouldn't be asking us. So why do you even care? Why do you want to run some random program that you have no idea what it does?

[Image Analyst]

Is the original author not available for questions? I mean, I'd have to spend maybe a half hour delving into that code, and I guess you would too. Perhaps a discussion with the author would speed that up.

[Jan]

@Alfio Rezika: Are you sure that you are allowed to publish the code? If the documentation of the code does not explain clearly, what is performed, a professional programmer would treat this program as unreliable - which is a good reason to delete it. Guessing details about foreign code is extremely prone to errors and bugs. I'm sure even the members of the forum cannot guess the "name" of the algorithm and I cannot imagine what "standard Matlab" is. So my advice is: If there is no exhaustive documentation and you cannot contact the author, delete the code. Seriously.

[Jan]

@Alfio Rezika: Deleting questions after an answer has been given is disliked in the forum. With "delete the code" I meant: on your computer. I've suggested this, because it is a known effect, that working with foreign codes can only be reliable, if they are shipped with a very exhaustive documentation. In other cases understanding the and modifying the code takes more time than rewriting it, in general.

[Rik]

Comments restored from search engine cache:

OP on 19 Jun 2017

Edited: OP on 19 Jun 2017

it generates image template sir. i use this piece in my iris recognition project on template based feature extraction process . i need to know so i can write the report about what algorithm is used in this codes. thanks for a nice answer and guess. really helpful. why you are so smart?

% Comment from Image Analyst

OP on 19 Jun 2017

no chance sir, i've emailed him. no response.

% Comment from Jan

OP on 25 Jun 2017

by matlab standart i mean like the common way to code some process in matlab sir. in this case for generating image template. i will delete it soon.

% Comment from Jan

OP on 25 Jun 2017

Edited: OP on 25 Jun 2017

ya ok thanks alot sir. FYI not every question is belongs to me. there are friends who also want to ask but dont have an account yet. so i offer mine to them. im so grateful there are alot of expert who trying to help. thank you.