<< Chapter < Page Chapter >> Page >

We have chosen a variety of features from literature review to explore. [1] These can be divided into two categories, binary and grayscale, which refer to the types of images that the code operates on. The binary features are based on morphological properties including cell and nucleus area and perimeter. The grayscale features describe the texture and contrast of the cells. We implemented a total of 12 features, 5 binary and 7 grayscale. All the features below were calculated for both the cell and the nucleus, except for the ratio of nucleus area to cell area.

We used 150 test images in order to extract features for our matrices. Again, our images were 250 x 250 pixels which we believed to be large enough to minimize errors from segmentation.

Binary features

Ratio of nucleus to cell area

The ratio of nucleus to cell areas can be calculated easily using a set of binary nucleus and cell images.

We calculate area by summing all of the white pixels in an image into a scalar number, and compute the ratio by dividing these numbers.

Ratio of area to perimeter

The ratio of area to perimeter is calculated for both the cell and the nucleus. Perimeter can be calculated using the MATLAB image processing function regionprops 'Perimeter', which sums continuous adjacent pixels of objects in an image.

Circularity

Circularity is calculated as a function of area and perimeter (ref paper, insert eq + variables). We calculated circularity for both the nucleus and cell. The closer the value is to 1, the more circular an object is.

Circularity formula.

Compactness

Compactness is calculated for both the cell and the nucleus. It is another measure of how circular an object is. Compactness is defined as the square root of area of the nucleus divided by the area of a circle with the same perimeter.

Compactness Equation

Grayscale features

The grayscale features are based on the gray-level co-occurrence matrix (GLCM), which can be used to extract statistical measures of texture. The GLCM is a matrix with elements p(i,j) that are equal to the number of times in the image a pixel with grayscale intensity i appears adjacent to a pixel with grayscale intensity level j. [3]

Contrast

Contrast measures the difference in grayscale intensity between adjacent pixels over the entire image. The greater the difference in intensity values, the higher the value of contrast.

Contrast equation.

Homogeneity

Homogeneity measures the distances of GLCM elements from the GLCM diagonal. Homogeneity ranges from 0 to 1. If adjacent pixels always have very similar values of grayscale intensity, the homogeneity will be close to 1.

Homogeneity equation.

Entropy

Entropy is a measure of the randomness of grayscale intensity values of pixels. Entropy is based off the grayscale histogram of the image. The histogram can be created from the GLCM by summing across the rows to find the total number of pixels p(i) for each grayscale intensity value i. [4]

Entropy equation.

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now




Source:  OpenStax, Automatic white blood cell classification using svm and neural networks. OpenStax CNX. Dec 16, 2015 Download for free at http://legacy.cnx.org/content/col11924/1.5
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Automatic white blood cell classification using svm and neural networks' conversation and receive update notifications?

Ask