Anemia blood cell localization using modified K-means algorithm

Abstract EN

In this project segmentation of image strategy based on K-means clustering calculation is displayed.

The proposed strategy utilizes clustering to allocate the dominant colors in medical tissue images for purpose of segmentation with high performance.

The initialization step of the system is the selection of suitable color model used for segmentation.

A set of inter and intra-class measures are used to evaluate the degree of model suitability.

The method is able to make segmentation at different classification resolutions.

For purpose of performance evaluation the comes about of the proposed strategy, standard K-Means and as of late altered K-Means are compared.

The exploratory comes about appeared that the proposed strategy gives superior result.