Re: k-means Clustering

Posted by Richard Han-2 on Dec 10, 2007; 10:02am
URL: http://imagej.273.s1.nabble.com/k-means-Clustering-tp3697801p3697804.html

Hi Toby,

Thank you so much for your reply.
I would like to measure the area of red/blue from RGB images of  
histological slides. I have tried the Threshold color (HSV) as  
suggested to me by Jacqui Ross. It seems the Threshold color is very  
accurate (better than the k-means). If I use RGB split, the red  
channel actually resemble the blue in the original image??
I guess I will stick with the threshold color (HSV) for measurement.

Regards

Richard

>
> Date:    Fri, 7 Dec 2007 10:31:32 -0500
> From:    Toby Cornish <[hidden email]>
> Subject: Re: k-means Clustering
>
> It might be helpful to know what sort of images these are.
>
> I haven't worked a lot with k-means for color discrimination, but  
> AFAIK this is just using kmeans clustering based on pixel RGB value  
> distances, so it doesn't really know "blue."  You could follow up  
> the kmeans with a little algorithm that looks at the pixels in each  
> cluster and evaluates which cluster is the most red and which is  
> the most blue.
>
> Alternatively, you could use other means of segmenting an image  
> into colors, such as HSV colorspace segmentation.
>
> toby
>
>> Date:    Thu, 6 Dec 2007 09:53:04 +0000
>> From:    Richard Han <[hidden email]>
>> Subject: k-means Clustering
>>
>> Dear All,
>>
>> I am using k-means Clustering to measure the blue and red in my
>> images. I use the threshold to move from one cluster to another. In
>> some images the k-means clustering could produce very good matches
>> (I'd say spot on), but in others the results were less desirable. My
>> settings are
>>
>> Number of clusters to: 4
>> Cluster center tolerance: 0.0001
>> Enable randomization seed: ticked
>> Randomization seed: 48
>> The other options were not ticked
>>
>> Could I do anything more to improve the analysis? and... Is it
>> possible to pre-define the red and blue clusters (ie, assign cluster
>> 1 to be blue)?
>>
>> Any help will be appreciated. Thanking you.
>>
>> Richard Han
>> University of Edinburgh
>
>
> Toby C. Cornish, M.D., Ph.D.
> Pathology Resident
> Johns Hopkins Medical Institutions
> [hidden email]
>
> ------------------------------
>
> Date:    Fri, 7 Dec 2007 10:48:08 -0500
> From:    Toby Cornish <[hidden email]>
> Subject: Re: k-means Clustering
>
> Something else I forgot is that you could try kmeans after  
> conversion from RGB colorspace to the LAB or another colorspace  
> (convert to a 3-slice stack in a new colorspace and run kmeans on  
> the stack). Plugins are readily available for that process.
>
> toby
>
>> Date:    Thu, 6 Dec 2007 09:53:04 +0000
>> From:    Richard Han <[hidden email]>
>> Subject: k-means Clustering
>>
>> Dear All,
>>
>> I am using k-means Clustering to measure the blue and red in my
>> images. I use the threshold to move from one cluster to another. In
>> some images the k-means clustering could produce very good matches
>> (I'd say spot on), but in others the results were less desirable. My
>> settings are
>>
>> Number of clusters to: 4
>> Cluster center tolerance: 0.0001
>> Enable randomization seed: ticked
>> Randomization seed: 48
>> The other options were not ticked
>>
>> Could I do anything more to improve the analysis? and... Is it
>> possible to pre-define the red and blue clusters (ie, assign cluster
>> 1 to be blue)?
>>
>> Any help will be appreciated. Thanking you.
>>
>> Richard Han
>> University of Edinburgh
>
>
> Toby C. Cornish, M.D., Ph.D.
> Pathology Resident
> Johns Hopkins Medical Institutions
> [hidden email]
>
>