Items filtered by date: October 2019

- The Head of Computer Sciences Department Prof. Dr. Hala Bahgat Abdel Wahab received a thanks and appreciation letter from the Faculty of Information Technology / University of Babylon for her extraordinary efforts in the committee to discuss the thesis of a doctoral student (Basim Sahar Yassin), hoping from her to give more to serve the public interest.





Source: Department's Media

Published in Dept. News

- The teachers (Dr. Yousra Hussein Ali) from the Department of Computer Science with the participation of researcher Maysaa' Saadoun Mohsen published their scientific research titled (the removal of reflections using the smallest iterative squares method) in the Iraqi Journal of Science issued by the College of Science / University of Baghdad. The idea of ​​the research is summarized in the widespread effects in the images captured through a transparent glass medium. These reflections represent a major problem in photography and strongly affect the performance of image processing algorithms. This research deals with the problem of separating reflection from two images taken at two different angles in front of a transparent glass medium and suggests the use of the smallest iterative squares method in the separation process, which has proven superior to performance over the modern algorithm through quantitative and visual comparisons.





Source: Department's Media

Published in Dept. News

- Two teachers (Dr. Abdul Amir Abdullah Karim and Dr. Ikhlas Faleh Nasser) from the Department of Computer Science published a research titled (Distinguishing the English letter from the video series based on a bag of visual words) in the Journal of the College of Madinat Al-Elem University. the research states that there are a lot of digital images available for the printed documents and the search is still on to reach the best algorithm for recognizing the English letters The proposed method consists of four steps. First the wavelet transformation is applied to literal images using the filter. Second, the important points are discovered using the angle detector. Third, those important points are described using the strong Acceleration Descriptor. Fourth, used an algorithm to group attributes into clusters to get a bag of visual words. The experimental results showed that the proposed method is more efficient, fast and distinctive compared to using the method of seven moments and with less time so the bag of visual words is considered the optimum accuracy in the process of distinguishing letters.





Source: Department's Media

Published in Dept. News

    Developers              Dept. Media         

Top