Plot a DET curve for bio-metric verification

[ncomputing]

Hi guys,
I plot a DET curve based on following steps: First, I change the threshold and count the number of false rejection and false acceptance. Second, I use plot MATLAB function to draw FAR and FRR.

However, national institute of standard and technology (NIST) releases a MATLAB function called "[P_miss,P_fa] = Compute_DET(True_scores,False_scores)". When I used Compute_DET(FRR,FAR) gives me bellow figure.

But, in guidance of this function it has applied as following:
Ntrials_True = 1000;
True_scores = sqrt(var(FAR1))* randn(Ntrials_True,1)+mean(FAR1);

Ntrials_False = 1000;
mean_False = mean(FRR1);
stdv_False = sqrt(var(FRR1));
False_scores = stdv_False * randn(Ntrials_False,1) + mean_False;

%-----------------------
% Compute Pmiss and Pfa from experimental detection output scores
[P_miss,P_fa] = Compute_DET(True_scores,False_scores);

%-----------------------
% Plot the detection error trade-off
figure;
Plot_DET (P_miss,P_fa,'r');
title('A DET plot');

Now, my question is which DET curve is true? I mean when I count False acceptance and False rejection rate based on different threshold, what should I have to do for plotting correct DET curve ?
I will really appreciate, if anyone can guide and explain to me. Thanks.