Professional Documents
Culture Documents
Gaussian Mixture Models
Gaussian Mixture Models
Each element in the image is pixel whereas image is a matrix. In that matrix, Pixel value is a number
which has its own entity like intensity, color of image. Consider variable X which is a random in nature
and can take pixel values. Suppose Mixture of Gaussian distribution for probability model is described as
the following
k
f ( x )=∑ p i N ( x∨μi , σ 2i )
i=1
2
2 1 1 −( x−μi)
N ( μi , σ ) =
i exp 2
σ √ 2 pi 2 σi
For a given image X, μi , σ i are mean and standard deviation respectively of class i. the lattice data are
values of Pixels for given image x and GMM is pixel base model. Therefore the parameters are
θ=( p¿¿ 1 , … .., pk , μ 1 …., , μ k , σ 21 , … , σ 2k )¿ the numbers or regions can be identified using histogram of
lattice data in GMM.
GLCM is texture character profile and this profile mention to touch i.e. smooth, silky and rough etc.
First order texture measures are statistics declared from the original image values, like variance, and
pixel neighbour relationship are not implemented. Second order measures defines the relationship
between groups of two (usually neighbouring) pixels in the original image. Third and higher order
textures (noting the relationships among three or more pixels) are theoretically possible but practically/
commonly not implemented due to calculation time and interpretation difficulty. GLCM texture picks up
the relation between two pixels at a time, called the reference and the neighbour pixel. GLCM expounds
the distance and angular spatial relationship over an image sub- region of specific size. GLCM is prepared
from gray scale values. It is taken into account how often a pixel with gray level(gray scale intensity or
gray tone) values come either horizontally, vertically and diagonally to levelled
They are also declared as p135 for top to bottom right, p90 for vertical, p45 for
bottom left to top right and p0 for horizontal.
Firstly fetch angle at 0 degree (horizontal).In the GLCM output, the element (1,1) has value1
because in the input image, there is only 1 opulence where two horizontally near to pixels of
distance 1 having values 1 and 1.GLCM(1,2) has value 2 because in the input image there are
two opulences where two horizontally near to pixels of distance 1 having value 1 and 2.
GLCM(1,3) has value 0 because in the input image there are no opulence where two horizontally
near to pixels of distance 1 having value 1 and 3.The procedure is repeated for the whole GLCM
matrix at different angles.[2]
0 ° 45 ° 90° 135 °
To illustrate, the following figure shows how graycomatrix calculates the first three
values in a GLCM. In the output GLCM, element (1,1) contains the value 1 because
there is only one instance in the input image where two horizontally adjacent pixels
have the values 1 and 1, respectively. glcm(1,2) contains the value 2 because there are
two instances where two horizontally adjacent pixels have the values 1 and 2. Element
(1,3) in the GLCM has the value 0 because there are no instances of two horizontally
adjacent pixels with the values 1 and 3. graycomatrix continues processing the input
image, scanning the image for other pixel pairs (i,j) and recording the sums in the
corresponding elements of the GLCM.