Min-Suk Bang

An efficient selection of HOG feature for SVM classification of vehicle

Support Vector Machine (SVM) classifier with Histogram of Oriented Gradients (HOG) feature become one of the most popular techniques used for vehicle detection in recent years. And the computing time of SVM is a main obstacle to get real time implementation which is important for Advanced Driver Assistance Systems (ADAS) applications. One of the effective ways to reduce the computing complexity of SVM is to reduce the dimension of HOG feature. In this paper, we examine the effect of the number of HOG bins on the vehicle detection and the symmetric characteristics of HOG feature of vehicle. And we successfully demonstrate the speed-up of SVM classifier for vehicle detection by about three times while maintaining the detection performance.

Performance analysis of an OFDM-based method for V2X communication

In an effort to enhance traffic safety by employing advanced wireless communication systems, intensive investigations and research are in progress worldwide. V2X communication is one core solution for governing and advancing future traffic safety and mobility. This paper presents a performance analysis of OFDM-based systems for V2X communication, to guarantee frequency non-selectivity and minimum ICI (Inter-Carrier-Interference). Most importantly, this analysis aims to mitigate the Doppler spread due to vehicles moving at high speeds.

Visual quality improvement for hybrid 3DTV with mixed resolution using conditional replenishment algorithm

This paper proposes the conditional replenishment algorithm (CRA) to improve the visual quality of hybrid stereoscopic 3DTV where the spatial resolutions of left and right views are mismatched. The basic concept of CRA is to determine the better replacement from disparity compensated view and simply enlarged view in generating the enhanced right view. The simulation results show that the proposed CRA can successfully improve the objective quality of the poor view and give positive effect on the final 3D visual quality.

Quality control of conditional replenishment algorithm for hybrid 3DTV with mixed resolution

Conditional replenishment algorithm (CRA) was proposed to improve the visual quality of 3DTV system with mixed resolution. Since CRA generates additional information which is transmitted to a decoder, the bit rate or quality gain needs to be carefully managed. This paper proposes a method to control the quality of images enhanced by CRA as uniform as possible while maintaining the average bit rate. The simulation result shows that the quality can be successfully controlled.

Effects of selection of a reference view on quality improvement of hybrid 3DTV

Conditional replenishment algorithm (CRA) was proposed to improve the quality of a hybrid 3DTV system with mixed resolution. In the hybrid 3DTV system, the role of a reference view is important since not only the overall quality of the hybrid 3DTV depends on the quality of the reference view, but also the reference view directly affects the quality improvement of the hybrid 3DTV using CRA. Moreover, since stereo mismatch between left and right cameras can be occurred, carefully choosing the reference view between the left and right views is very important. In this paper, we observe effects of the selection of a reference view on the quality of a hybrid 3DTV which is improved by CRA.

A method to efficiently encode inter-view vectors in conditional replenishment algorithm

Conditional replenishment algorithm (CRA) was proposed to improve the quality of a hybrid 3DTV system with mixed resolution. Since CRA generates additional information which includes inter-view vectors, it is needed to efficiently encode the information. This paper proposes a method to apply differential pulse code modulation (DPCM) to the inter-view vectors so as to improve the coding efficiency of CRA.

Compensation of Directional and Rotational Mismatch of Separate Stereo Cameras by Vanishing Point and Lines

The rectification of stereoscopic image is indispensable process to implement stereo vision system since the accuracy of estimated disparity strongly depends on the alignment of epipolar line. A new and simple rectification method is proposed in this paper especially when there are vanishing points and lines in two images from separate cameras. In order to compensate the mismatch of viewing directions of separate cameras, we estimate the vanishing points of two images and make common reference point. After that, the mismatch of rotation is compensated by calculating the relative distance between vanishing lines in two images. The proposed method is very efficient for rectification of separate stereo cameras and successfully applicable to the scene of streets or hallway where the vanishing points and lines can be easily obtained.