• Proceedings of the Korea Database Society Conference
• /
• 1995.12a
• /
• pp.165-168
• /
• 1995

본 논문은 비디오 연속 영상에서 얼굴의 위치를 추적하는 알고리즘에 관하여 기술한다. 컴퓨터 비젼에서 대량의 비디오 연속 영상내 물체 추적은 실시간에 처리되는 빠른 알고리즘이 요구된다. 기존의 방법은 형태에 기반한 알고리즘으로 물체의 회전, 크기 변화, 겹침 등에 대한 문제에 민감하여 여러 가지 어려움이 발생한다. 그러나 칼라를 이용한 알고리즘은 이러한 문제에 대하여 둔감하여 훨씬 효과적이다. 본 논문은 칼라 3D 히스토그램을 이용한 Swain과 Ballard의 역 투사(backprojection) 방법을 적용하여 비디오 연속 영상에서 얼굴의 위치를 빠르고 정확히 추적하는 알고리즘을 제안한다.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.1
• /
• pp.52-60
• /
• 2008

Purpose: To establish the methods for sinogram formation and correction in order to appropriately apply the filtered backprojection (FBP) reconstruction algorithm to the data acquired using PET scanner with multiple scintillation crystal layers. Materials and Methods: Formation for raw PET data storage and conversion methods from listmode data to histogram and sinogram were optimized. To solve the various problems occurred while the raw histogram was converted into sinogram, optimal sampling strategy and sampling efficiency correction method were investigated. Gap compensation methods that is unique in this system were also investigated. All the sinogram data were reconstructed using 20 filtered backprojection algorithm and compared to estimate the improvements by the correction algorithms. Results: Optimal radial sampling interval and number of angular samples in terms of the sampling theorem and sampling efficiency correction algorithm were pitch/2 and 120, respectively. By applying the sampling efficiency correction and gap compensation, artifacts and background noise on the reconstructed image could be reduced. Conclusion: Conversion method from the histogram to sinogram was investigated for the FBP reconstruction of data acquired using multiple scintillation crystal layers. This method will be useful for the fast 20 reconstruction of multiple crystal layer PET data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.6
• /
• pp.780-785
• /
• 2010

This paper introduces GP(Genetic Programming) based color detection model for an object detection and tracking. Existing color detection methods have used linear/nonlinear transformatin of RGB color-model and improved color model for illumination variation by optimization or learning techniques. However, most of cases have difficulties to classify various of colors because of interference of among color channels and are not robust for illumination variation. To solve these problems, we propose illumination robust and non-parametric multi-colors detection model using evolution of GP. The proposed method is compared to the existing color-models for various colors and images with different lighting conditions.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.2-5
• /
• 2022

Human organs in the body have a complex structure, and in particular, the small intestine is about 7m long, so endoscopy is not easy and the risk of endoscopy is high. Currently, the test is performed with a capsule endoscope, and the test time is very long. The doctor connects the removed storage device to the computer to store the patient's capsule endoscope image and reads it using a program, but the capsule endoscope test results in a long image length, which takes a lot of time to read. In addition, in the case of the small intestine, there are many curves due to villi, so the occlusion area or light and shade of the image are clearly visible during the examination, and there may be cases where lesions and abnormal signs are missed during the examination. In this paper, we provide a method of assisting small intestine capsule endoscopic lesion examination using artificial neural networks to shorten the doctor's image reading time and improve diagnostic reliability.