Michael Parra – Update 3

As of now I have been looking into Skin Detection as told by Skyler. I am still working on the Raspberry Pi, however I am trying to switch into using Skyler’s Camera Interface programmed in Verilog.

Here is a website/pdf that I was looking into for the Skin Detection.

http://w3.ualg.pt/∼ftomaz/fr/fr.php

In this tutorial I read that every image uses Blue Red Green to manipulate multiple colors. This if statement shows the colors that are normal for human skin color. This is just one filter to detect the actual color boundaries of white and black and mixed colored people.

// R G and B values are presented in the range 0 to 255
if ( (B > 160 && R < 180 && G 160 && R < 180 && B < 180) || // Too much green
(B < 100 && R < 100 && G 200) || // Green
(R+G > 400) || // Too much red and green (yellow like color)
(G > 150 && B .40) || // Too much blue in contrast to others
(G/(R+G+B) > .40) || // Too much green in contrast to others
(R 100 && B > 110 && G < 140 && B < 160) // Ocean
)

Furthermore what would be needed is to group these objects together to determine if a person or face is there. meaning that this if statement should test just about every pixel and once it is tested it should group the close pixels together to detect the human skin.

I would like to spend more time on researching skin color detection using this pdf/website:

http://www.cs.rutgers.edu/~elgammal/pub/skin.pdf
___________________
aside from this skin detection algorithm I am also working on the PI, as a back up/ last resort case if we do not get the face detection to work on FPGA.

I have just installed and finished downloading a couple of Open Source Files onto my Pi, and am just waiting on it to finish updating.

https://pythonprogramming.net/raspberry-pi-camera-opencv-face-detection-tutorial/

this tutorial seems to be more clear on instructions for getting the face detection to work. I downloaded the 14.7 GB File from the mentioned google drive. and am having some difficulties with operating systems.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s