Analysis of Kirlian images: feature extraction and segmentation

Images of high geometrical complexity are found in various applications in the fields of image processing and computer vision. In this paper we utilize general processing techniques, mainly based on image morphology. We focus on Kirlian images, which due to their high complexity, comprise of features appearing in many biomedical images. In this paper, a first approach is given on the extraction of specific features dealing with the size and geometrical structure of Kirlian images. The extraction is implemented with the use of tools provided by the broader field of computer vision, thus providing a multi-faceted description of the images. Furthermore, this paper provides and promotes the use of automatically extracted information. Finally, efficient algorithms for obtaining the information on the size and structure of Kirlian images are presented and a number of conclusions are drawn and discussed that provide an insight on the underlying information within a highly complex image such as Kirlian images.