after haar wavelet how i can detect left extreme and right extreme pixel locations of eyes , mouth,nose.

I have applied Haar wavelets on face image, one which represents the two eyes, then a brighter bar nose and cheeks, and then another dark bar where the mouth is. After that, i want to find out left extreme and right extreme points (pixel locations)of eye, mouth ,nose. how i can findout it please help.

Answer 1

[Image Analyst]

Use regionprops() to get the bounding box of all the blobs. See my image segmentation tutorial in my File Exchange. http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862

[sweety]

Thank you for reply... I applied this whole method on face image but it is not giving desired results.. beacause i want to find out pixel locations of left extreme and right extreme points after that i want to calculate distance between them ...please give me proper steps or any link . Thanks.

[Image Analyst]

Maybe your Haar wavelet code is no good. Try the face detector built in to the Computer Vision System Toolbox like Dima suggested. Otherwise post your accurately and correctly segmented image showing the blobs for the eyes, nose, etc. Or look in Vision Bib for more algorithms. We can't do long involved algorithm development for you so you'll have to do that yourself but we can help with simple things that take just a few minutes, such as spotting syntax errors or errors in program flow, etc.

[sweety]

Thanks for reply.

Answer 2

[Dima Lisin]

If you have the Computer Vision System Toolbox, you can use vision.CascadeObjectDetector to detect faces, eyes, nose, and mouth.

[sweety]

Thanks for reply.

[Image Analyst]

And be sure to see the Mathworks Webinar on Face detection/recognition on Nov. 13, 2014: Click Here