MATLAB: Classical image segmentaion problem

digital image processingimage processingImage Processing Toolboximage segmentation

hi I am trying to extract the fields in the image for example the whole field against first anme and so on. Now i have achieved partial results from vertical and horizontal edge detection but i want to explore any new algorithm or ideas hints etc. Attached is my image any suggestions are more than welcome guys. See attached image

Best Answer

This is an obvious case for a template . No edge detection needed. You have a form! Take advantage of that fact! Just have a master image that is perfectly aligned with the rows and columns of the image. Then determine the box coordinates for all the fields on that perfect master form. Now, for a "test" form, use imregister() to align it to the master one, then simply use the known coordinates of the fields to extract the sub-images. There are examples for aligning images in the help for imregister.

Related Solutions

MATLAB: Segmentaion of dermoscopy image

Take the blue channel and threshold it to select dark things:

binaryImage = rgbImage(:,:,3) < 150; % Or whatever threshold works.

Then call bwareafilt() to take the largest blob

binaryImage = bwareafilt(binaryImage, 1);

MATLAB: Classical orbital elements Vectors

Use the norm( ) function instead of mag( ). E.g., norm(v) instead of mag(v).

Calculate evec before you calculate e.

Your evec code appears correct, but your description of this is missing the mu term in the denominator. E.g.,

e⃗ =((v2−μ/r)r⃗ −(r⃗ ⋅v⃗ )v⃗ )/μ

Best Answer

Related Solutions

MATLAB: Segmentaion of dermoscopy image

MATLAB: Classical orbital elements Vectors

Related Question