The project deals with graph based image partitioning using spatial data. Traditionally spatial data has been stored and presented in the form of a map. The images are segmented and then the segmented data is transformed by a threshold value given as input. The number of objects or the size is controlled thereby segmentation becomes effective. The algorithm is suitable for both gray scale as well as color images of different types such as bitmap or jpg. The method affords an effective

more »

... e to complex modeling of the original image data while taxing advantage of the computational benefits of graph cuts. The project segments image by keying in any point location inside the automatically starting from the center point. In addition, the statistical data such as number of objects found during segmentation and similar objects with in the image are also calculated. The gray scale conversion of the particular segments is also carried out so that the output image partitions the image into different objects. Suppose the image may be x-ray images which could identify the cell information.