|
A Turn Based Algorithm In An Urban Road Transportation Network |
|
ByuhngMunn Suhng, Heeduk Jeong, Wangheon Lee(HANSEI UNIVERSITY, Republic of Korea) |
| A turn based algorithm is a generalized link based algorithm so as to solve the turn penalty problem. The recursive visit-unit changing of the turn based algorithm builds up hierarchical node-link data layers. This paper propose a turn based algorithm in an urban road transportation network by using hierarchical node-link data layers. |
|
|
Target Size Prediction and Verification by Geometrical Analysis and SE-WORKBENCH for Ground Target Detection |
|
Sungho Kim(Yeungnam University, Republic of Korea), Sohyun Kim(Agency for defense development, Republic of Korea) |
| Conventional target detectors use filter kernel to enhance target signature and depress background clutter. It is important to use a predefined kernel size to enhance target detection performance and speed. This paper presents a geometrical analysis of ground target size using a set of target acquisition parameters and verifies predicted the target size by comparing with the output of commercial simulator (OKTAL SE-WORKBENCH). |
|
|
Object Tracking using KLT –aided Mean-shift algorithm |
|
Sunho Kim, Jungho Kim, Youngbae Hwang, Byoungho Choi, Juhong Yoon(Korea Electronics Technology Institute(KETI), Republic of Korea) |
| In this paper, we present a new object tracker using color-based mean-shift and feature-based optical flow methods. By fusing two approaches, we show the improved performance of object tracking for partial occlusion and severe appearance changes. For this purpose, we iteratively compute the mean-shift vector based on color histograms and tracked features by KLT. |
|
|
Performance Improvement in complex environment Based on Ensemble Learning Algorithm By Combining 2D DNF Weak Classifier |
|
Hyeon-Gyu Min, Dong-Joong Kang(Pusan National University, Republic of Korea) |
| The object detection method employing the 1D feature has benefit that the calculating speed is fast. However, detection accuracy and performance is low in complex background. Therefore, in this paper, we propose an ensemble learning algorithm that combines 1D feature classifier and 2D DNF cell classifier to improve the performance of object detection in single input image. The reason for selecting 2D DNF classifier is that the classifier is able to classify the object not categorized in traditional weak classifier. And we proposed method to choose the feature for reducing the time of learning. |
|
|
Real-Time Range Image Segmentation on GPU |
|
XINHUA JIN, Mun-Ho Jeong(Kwangwoon University, Republic of Korea) |
| In this paper propose a GPU-based parallel processing method for real-time image segmentation with neural oscillator network. Range image segmentation methods can be divided into two categories: edge-based and region-based. Edge-base method is sensitive to noise and region-based method is hard to extracting the boundary detail between the object. However, by using LEGION (Locally Excitatory Globally Inhibitory oscillator networks) to do range image segmentation can overcome above disadvantages. The reason why LEGION is suitable for parallel processing that each oscillator calculate with ... |
| |