How to divide image and calculating quantized mean values of hsv and lab color channel.

3 次查看（过去 30 天）
s malik
s malik 2016-4-13
评论： s malik 2016-4-14
I want to divide an image into 8* 16 overlapping blocks,
blocks are overlapping of 50% in horizontal and vertical.
then from each block, I want to compute quantized mean values
of HSV and lab color channel.

请先登录，再回答此问题。

采纳的回答

Image Analyst
Image Analyst 2016-4-13
Easily accomplished with blockproc(). See my attached demos for about 6 or 8 different ways to call blockproc(). Find the one you want and adapt it to your situation. Post your adapted code back here if you still need help.
  1 个评论
s malik
s malik 2016-4-14
I have read your tutorial and It is quite helpful. Specifically I have changed the code as under.
Img=imread('v1.tif');% specifically the size of the image is 128*48
hsvImg=rgb2hsv(Img);% as I want the mean HSV values
% meanFilterFunction = @(theBlockStructure) mean2(theBlockStructure.data(:)) ;
blockSize = [8 16];
overlap=[4 8];
blockyImage = blockproc(hsvImg, blockSize, meanFilterFunction,'BorderSize', overlap,'TrimBorder', 0);
[rows, columns] = size(blockyImage);
please tell me whether overlapping is correct as i need 50% blocks overlapping.

请先登录，再进行评论。

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Neighborhood and Block Processing 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by