Aleksandrs Sisojevs

A Surface Smoothing Method for a 3D Model of a Medical Object

This paper describes a method to smoothen the surface of a Medical object’s 3D model. This method is intended to be used on model that was obtained by a triangulation algorithm, but it also can be used on a model that was obtained by a marching cubes algorithm. The basic principle behind this algorithm is that it adjusts the position of the vertices of a 3D model relative to the neighboring vertices, thus evening the rough edges. This method was tested on the model of human head, which was acquired by computer tomography and it showed considerable visual improvement of the model.

Medical Image Region Extraction and 3D Modeling Based on Approximating Curves

This work describes several methods that intend to solve such medical image processing tasks as extraction and 3D visualization of the region of interest (ROI). The proposed methods were tested on the medical images of a brain acquired by computer tomography and proven to be applicable to different types of ROI, resulting in a possible visualization of several ROI at once, i.e. pathology and the head of a patient. The results can be used to provide practical improvements to the reliability of medical diagnostics.

Face recognition method for cases of an insufficient training set, using 3D models of face that were created using two facial images

The face recognition method is proposed for cases of an insufficient training set, when the input data consists only of two facial images (full face and profile). The 3D model of a face is created semi-automatically using the input data (two images), which is then used for the recognition process. The training set for the recognition process consists of these created 3D models of faces. The basic problem of face recognition is the insufficient information about the proportions of the unidentified persons face, images can also contain some artefacts, for example eyeglasses, beard, moustache that can decrease the precision of the recognition process and make the image analysis more difficult. Another important aspect is illumination, which can practically change the results of the classification. The proposed recognition method consists of several steps: unknown image face alignment, facial reference points estimation using gradient maps using dlib (http://dlib.net/) and OpenCV (https://opencv.org/) open source computer vision libraries. After features extraction it is necessary to perform thresholding on some facial reference points, which is most important for recognition process. For this purpose, several important features are selected and distances between them are calculated. The training set consists of early created 3D models of faces that could be used to get the missing information about the proportions of the persons face. The proposed algorithm is used for classification. Using this method classification results are approximately 90% positive compared to when using only the insufficient training set that contains only two images.