05-20-2020 02:20 PM
hello, am doing a project on brain tumour detection i already segmented out tumour part from the background but i stucked in how to figure out the glcm features like contrast, skewness, kurtosis and so on to differentiate malignant and beinign images
05-20-2020 08:58 PM
Wow! If you wrote this code, I recommend finding someone who is an expert at Image Analysis and knows what GLCM (I bet you spell LabVIEW wrong, as well) is and means. If you want them to code in LabVIEW, ask to see a sample of their code -- if they use any sequence structures, or if no Block Diagram fits on a single screen, reject them and keep looking.
You do know, I presume, that the "G" stands for "Grey" (or "Gray"), and you know you are reading in images in Grayscale. So why are there functions "Color to RGB" in this VI? If you want to develop an algorithm whose first name is "Gray" (or "Grey"), why don't you forget all about "colorizing" your Image and implement the algorithm? The method is pretty straight-forward -- just follow the instructions. When in doubt, take a constant Image and analyze it. When you've done that, and have a simple VI that takes a given grayscale image (which you've attached) and implements your attempt at the GLCM algorithm, attach the Image and the Algorithm if you are still having problems and we'll critique your code. But focus on the GLCM.
Bob Schor
05-20-2020 09:25 PM
color to rgb is needed to convert original image in to arrays for segmentation process ....... but as a i attached below segmentation with k-means is perfect but i want to get features of segmented image again .......
05-20-2020 09:29 PM
the main problem am facing is i cant get pixel information to go through glcm method