Lung segmentation on x-ray images with neural validation
Dawid Połap, Marcin Woźniak · 2017
Lung segmentation on x-ray images is an important part in the process of feature extraction for recognition purposes. Using it we can extract specific data from the input image. Segmentation allows to remove unnecessary elements such as bones and spine, leaving in the image only the lungs. This solution reduces the area of the image subjected to further analysis in terms of disease detection. In this paper, segmentation technique based on graphics processing methods and swarm algorithm was presented. A swarm methodology was used for extraction of particular portions of the information for which we have applied convolutional neural network as a detector. For the composed method we have performed tests to show and discuss the results.