Перегляд за автором "Vlasenko, N."

Зараз показано 1 - 2 з 2
  • Зображення мініатюри
    Публікація
    Aggregate  Parametric  Representation  of  Image  Structural  Description in Statistical Classification Methods
    (2022) Gadetska, S.; Gorokhovatskyi, V.; Stiahlyk, N.; Vlasenko, N.
    Finding effective classification solutions based on the study of the processed data nature is one of the important tasks in modern computer vision. Statistical distributions are a perfect tool for presenting and analyzing visual data in image recognition systems. They are especially effective when creating new feature spaces, particularly, by aggregating descriptor sets in some appropriate way, including bits. For this purpose, it is natural to apply the number of criteria designed to compare the distribution parameters of the analyzed samples. The article develops a speed-efficient method of image classification by introducing aggregate statistical features for the composition of the description components. The metric classifier is based on the use of statisticalcriteria to assess the significance of the classification decision. The developed classification method based on the aggregation of the feature image set is implemented; the workability of the proposed classifier is confirmed. On the examples of the application of variants ofthe method for the system of the real images features, its effectiveness was experimentally evaluated.
  • Зображення мініатюри
    Публікація
    The Research of Image Classification Methods Based on the Introducing Cluster Representation Parameters for the Structural Description
    (Seventh Sense Research Group, 2021) Ahmad M. Ayaz; Gorokhovatskyi, V.; Tvoroshenko, I.; Vlasenko, N.; Mustafa Syed Khalid
    The results of the development of high-speed methods for classifying images in computer vision systems using the description as a set of keypoints descriptors are presented. Classification methods based on the system of cluster centers parameters, which are independently constructed for etalon descriptors set, are researched. The competitive voting of the descriptors of object being recognized on a set of etalon centers is proposed. An optimal way of comparing the sets of cluster centers for an object and etalons is applied. Experimental estimation of the efficiency for the two presented classification methods in terms of computation time and classification accuracy based on the results of applied dataset processing are shown. Based on the research, a conclusion about the effectiveness of classification technologies using cluster centers for structural descriptions of images to ensure decision-making in real-time is made.