How to Connecting nearby pixels of binary image

How Connecting nearby pixels of binary image. Because i need to use image region analyzer and need to the image to be as one connected pixels i have thousands of image like that i need to make all pixels connected or how to make image region analyzer as one image not parts.. I work on CT image like this which i need image region analyzer as one part

[KALYAN ACHARJYA]

Is this segmenattion result?

[hussain abdelaziz]

This a slice of CT IMAGE OF HEAD OF A DOLPHIN

Answer 1

[Matt J]

imclose() is the traditional way to close small gaps between nearby pixels.

[hussain abdelaziz]

I will try it thank you

Answer 2

[Image Analyst]

Are there really gaps/openings there or not? If they aren't there, then it looks like you just did the segmentation improperly, like you chose a bad threshold or something. With a good threshold, you should be able to get at least the outer perimeter closed.

[hussain abdelaziz]

When i make image analyzer it read it 20 part!.. And i will try again to make it.

[Image Analyst]

I'm not sure what that means. Do you mean you have 20 images? Or do you mean there were 20 connected components after this app thresholded and labeled the blobs? Please attach the original image.

[hussain abdelaziz]

This is a original images if you convert it to binary and put it on image region analyzer it will give you more than one area.. And i would like to be all image one area to do that image must be all pixels connected to each other

[Image Analyst]

It doesn't look like a binary image. It looks like a gray scale image that was thresholded to create a binary image, and then the binary image was used to mask (erase) the gray scale image outside the mask. Like

mask = grayImage > someThresholdValue;
outputImage = grayImage; % Initialize
outputImage(~mask) = 0; % Erase outside mask.

What I'm saying is that your segmentation algorithm, which might be a simple global threshold like the first line of code I gave above OR it may be something more complicated, is not doing a good job masking the image. So, did you do a global threshold? If so, try changing the threshold value. Or did you do something more sophisticated, like called imbinarize() with the adaptive (locally varying) option?

[hussain abdelaziz]

Yes it's gray image.

Ok I will try thank you very much