I need a C# 2008 code for this project : "NO USE OF OPEN CV or EMU CV or any DLL allowed"
Project: Skin Likelihood based on Gaussian Model
I have a face data set, for this set I want to design a Gaussian model that based on the attached papers. The model takes the images for training.
- Image converted into pure colors , R=R/R+G+B,G=G/R+G+B,B=B/R+G+B
1- Convert each image from RGB to YCbCr
2- It calculates the mean and covariance for each image Cb and Cr, then
3- Find the Gaussian joint probability density function (pdf)
pdf = exp(-0.5* x'*inv(brcov)* x) ,where x = [(cb-bmean); (cr-rmean), "bmean&rmeam) are calculated from the point 2.
4- Apply Low pass files 3x3 on the result.
5- Divide each pixel on max pdf value.
6- Apply the Finding the adaptive threshold algorithm as this matlab code
% Adaptive Thresholding
previousSkin2 = zeros(i,j);
changelist = ;
for threshold = 0.55:-0.1:0.05
skin2 = zeros(i,j);
skin2(find(skin1>threshold)) = 1;
change = sum(sum(skin2 - previousSkin2));
changelist = [changelist change];
previousSkin2 = skin2;
7-Finally, optimal threshold is calculated for segmentation as matlab code below:
% Finding the optimal threshold
[C, I] = min(changelist);
optimalThreshold = (7-I)*0.1
skin2 = zeros(i,j);
skin2(find(skin1>optimalThreshold)) = 1;
All of above are in these matlab codes:
[url removed, login to view]~robles/ee368/detect.m
[url removed, login to view]~robles/ee368/ChromaDist.m
[url removed, login to view]~robles/ee368/skinmodel.m
[url removed, login to view]~robles/ee368/SegmentSkin.m
I'll provide you with the database after acceptance
My project similar to the skin detection equations in this page :
My project output show be similar to this
12 freelancer bu iş için ortalamada 653$ teklif veriyor
Hi Can you provide the paper? or any similar program that give the same output. Also please define what will be the input of the application and what's the expected outcome. Thanks
I can deliver a complete application in VS 2008 or VS 2010 using C#. with source code and installer. Application will calculate Gaussian joint probability density model, and can be displayed in 2d or 3d model.
Hi I am interested in your project. I am expert in image processing. I have rich experience in image color space, segment. I think I will be helpful to you. Thanks for your regards.