MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

# Feature Extraction using edge detection procedure

Asked by kanimozhi suguna on 19 Aug 2012

Hi all! I am working on mammogram image segmentation using MATLAB. I have a doubt. Is there any possibility to segment the original image using its edge detected image? Can the edge detected image can act as filter for the original image for segmentation? For this process do I need to convert the edge detected image to matrix format for segmentation? If possible can someone provide me sample code?

## Products

No products are associated with this question.

Answer by Image Analyst on 19 Aug 2012

I don't know the current best approaches for finding anomalies in mammograms. I suggest you look here http://iris.usc.edu/Vision-Notes/bibliography/contentsmedical.html#Medical%20Applications,%20CAT,%20MRI,%20Ultrasound,%20Heart%20Models,%20Brain%20Models When you've coded up your favorite algorithm into MATLAB, you can come back here with your programming questions. Usually we don't recommend the best algorithm for an application like this where it's not so straightforward, and biological images are usually the toughest to segment, so there is a huge variety of methods to deal with them. Maybe edge detection is the approach to use or maybe it's thresholding or texture analysis or morphology or something way more sophisticated, or a combination of approaches. I'll leave the algorithm selection up to you but we can help with the MATLAB programming part of it.

## 1 Comment

kanimozhi suguna on 20 Aug 2012

Sir you have asked to give details regarding the code.

I = imread('mdb001.bmp'); J = imadjust(I,[0.5 0.8],[0.5 1]); level = graythresh(J); BW = im2bw(I,level); edgenew = edge(BW,'sobel'); imshow(BW) figure, imshow(edgenew)

this is the code i used. from this i will be obtaining two images.. from either of the image i have to extract the needed area from the original image. I am confused in how to proceed. Please suggest.