Image segmentation for lung region in chest X-ray images using edge detection and morphology

Studies of medical image segmentation have long been done as a mean to distinguish object region from one to another for further image analysis. The segmentation of lung region in chest X-ray (CXR) based on object edge detection is one of the popular method applied. Early edge detection algorithms like Sobel, Prewitt and Laplacian have been used to segment the lung however, none of them can successfully generate a truly satisfied segmentation output. The reason for this fail is because they are high pass filter that is sensitive to image noise. Hence, the requirement for better edge detection algorithm that can cope with reasonable lower and upper threshold value for image noise like canny edge should be highlighted. Moreover, combining this algorithm with morphology method (dilation and erosion) will produce better outcome. Therefore, this paper has proposed method for segmenting lung region in CXR images using canny edge filter and morphology. Although the filter can detect the lung edge, unfortunately, the final edges lines produce are still unsatisfied. To solve the problem, Euler number method is applied to extract the lung region before executing the edge detection using the filter. The implementation produced convincing result as most of the segmented image is almost similar to the ground truth image.