The segmentation process helps to identify the specific regions in the images. The segmented regions were used for computer vision for further process. The required regions were previously saved and trained by the system. The seed regions were selected for each portions. The images were segmented by grouping the regions that are belonging to the seed regions. The segmented pixels were then outlined by applying contour for the images. The Image Foresting Transform (IFT) is a tool primarily to the design of image processing operators based on connectivity. The image is interpreted as a graph whose nodes are the pixels and the arcs are defined by an adjacency relation. A path π t is a sequence of adjacent nodes with terminus at some node t. The input images were initially preprocessed. The preprocessing step includes resizing and denoising using median filter. The denoising step removes the unwanted pixels in the image. The images were trained by obtaining the seed regions for the images. The seed regions for the input image is taken. The required region from the images is chosen by growing the seed region. The region growing process is done by Image Foresting Transformation which grows the initial seed region by finding the nearby pixels related to the seed region. For the input image given the 14 feature values were calculated. The distance between the features extracted for the test image and the features of the training images were calculated. The seed point corresponding to the minimum distance obtained is got from the dataset. The obtained seed points were then growed to obtain the needed lung region. The seed points were grown. The image pixels that have values related to the obtained seed point values were got. The image pixels were identified the values were placed in the same position in another image matrix. The boundary points were identified and the boundary were detected and marked. The identified image pixels were given separate colors to differentiate it from other regions. Here the IFT process is combined seed region calculation process which increases the segmentation accuracy of the process. The segmentation accuracy indicated the proposed IFT based segmentation process is helpful in the identification of the regions in the images with a greater reliability which is due to the capacity of the algorithm to choose the optimal path in the images so that the irregular shapes in the images can be also identified and segmented. The proposed system overcomes the disadvantages of the graph based approached for the identification of the specific shapes in the images. The performance of the process is measured by comparing the segmented pixels with the ground truth.