MATLAB: How to Remove Minor background details from the image

background removalImage Processing Toolboximage segmentation

For the object extraction i need to enhance the image contrast, which not only enhance foreground object but also the minor background details of gray image.

How to remove the background as i need the background clear in order to detect the edges of salient object?

Best Answer

It's hard to say without seeing the image. Perhaps thresholding? Perhaps something else first? But after all the processing steps are done, the last step in segmentation is always thresholding. You do whatever you need to do to get to a point where you can say "this is foreground" and "this is background" and that is thresholding.

Related Solutions

MATLAB: How can i separate all the connected regions in a binary image and display them individually.

Use bwlabel() and ismember

[labeledImage, numRegions] = bwlabel(binaryImage);
props = regionprops(labeledImage, 'Area')
% Compute all the areas.
allAreas = [props.Area]
% Display each blob in its own figure.
for k = 1 : numRegions
    thisRegion = ismember(labeledImage, k);
    figure;
    imshow(thisRegion);
    drawnow;
end

MATLAB: IMAGE ANALYSIS: Do I have to convert the grayscale AFM images into binary before doing edge detection/segmentation

"Is imbinarize(image) a necessary step to find connected objects and such?" Yes. Well, not necessarily imbinarize, but binarization from some method, like manual thresholding like

binaryImage = grayImage > someThreshold;

"Is it important to binarize grayscale images to do edge detection, segmentation of different foreground shapes, etc?" No Edge detection can be done on gray scale images. Segmentation is the process of applying various operations until you get to a point where you can threshold to produce a binary image. The binary image is the "segmented" image. If you then apply a label (ID number) to each connected blob, for example with the bwlabel() function, then that is now called a "labeled" image.

See my File Exchange for a variety of ways to segment gray scale and color images.

Best Answer

Related Solutions

MATLAB: How can i separate all the connected regions in a binary image and display them individually.

MATLAB: IMAGE ANALYSIS: Do I have to convert the grayscale AFM images into binary before doing edge detection/segmentation

Related Question