Jae-Soo Cho

Effective Lane Detection and Tracking Method Using Statistical Modeling of Color and Lane Edge-Orientation

This paper proposes an effective lane detection and tracking method using statistical modeling of lane color and edge-orientation in the image sequence. At first, we will address some problem of classifying a pixel into two classes(lane or background) and detecting one exact lane. Generally, the probability of a pixel classification error conditioned on the distinctive feature vector can be decreased by selecting more distinctive features. A proposed pixel classifier model(Bayes decision rule for minimizing the probability of error) uses two distinctive features, lane color and edge-orientation, for classifying a lane pixel from background image. By estimating PDFs of each feature and continuously updating the estimated PDFs, we can effectively adapt the various road conditions and the different types of lane. The histogram of edge magnitudes with respect to edge-orientation will be used as the PDF for the lane edge orientation feature. Similarly, the color histogram of the HSV color model will be used as the PDF of the color feature. And, for the postprocessing, we will use the LMS algorithm in order to exclude misclassified pixels and decide one optimal lane position. Various comparative experimental results show that the proposed scheme is very effective in the lane detection and can be implemented in real-time.

Intelligent video tracking based on fuzzy-reasoning segmentation

In our previous work [J. Cho, D. Kim, D. Park, Robust centroid target tracker based on new distance features in cluttered image sequences. IEICE Transactions on Information and Systems, Vol. E83-D, No. 12, December, 2000.], we presented a novel centroid target tracker based on new distance features in cluttered image sequences. A real-time adaptive segmentation method based on new distance features was proposed for the binary centroid tracker. The target classifier by the Bayes decision rule for minimizing the probability of error should properly estimate the state-conditional densities. In this correspondence, the proposed target classifier adopts the fuzzy-reasoning segmentation instead of the estimation of the state-conditional probability densities. Comparative experiments show that the performance of the proposed fuzzy-reasoning segmentation is superior to that of the conventional thresholding methods. The usefulness of the fuzzy-reasoning segmentation for practical applications is demonstrated by considering two sequences of real target images. The tracking results are good and stable without difficulty of the probability densities estimation.

Study on Effective Visual Surveillance System using Dual-mode(Fixed+Pan/Tilt/Zoom) Camera

An effective dual-mode camera system(a passive wide-angle camera and a pan-tilt-zoom camera) is proposed in order to improve the performance of visual surveillance. The fixed wide-angle camera is used to monitor large open areas, but the moving objects on the images are too small to view in detail. And, the PTZ camera is capable of increasing the monitoring area and enhancing the image quality by tracking and zooming in on a specific moving target. However, its FOV (Field of View) is limited when zooming in on a specific target. Therefore, the cooperation of wide-angle and PTZ cameras is complementary. In this paper, we propose an automatic initial set-up algorithm and coordinate transform method from the wide-angle camera coordinate to the PTZ one, which are necessary to achieve the cooperation. The automatic initial set-up algorithm is able to synchronize the views of two cameras. When a moving object appears on the image plane of a wide-angle camera after the initial set-up positioning, the obtained values of the wide-angle camera should be transformed to the PTZ values based on the coordinate transform method. We also develope the PTZ control method. Various in-door and out-door experiments show that the proposed dual-camera system is feasible for the effective visual surveillance.

A New Vertex Selection Scheme Using Curvature Information

A new vertex selection scheme using the curvature information of contour points for polygonal contour approximation is presented. The proposed method consists of two-step procedure. At first, we compute curvature values for contour points on the curvature scale space (CSS) and select high curvature points as principal vertices. Selected principal vertices, thereby, divide an overall contour into several contour segments. In the second step, the vertices for each contour segment are selected using the progressive vertex selection (PVS) method in order to obtain minimum number of vertices under the given maximum distance criterion (D*max).

A Learning-based Visual Inspection System for Part Verification in a Panorama Sunroof Assembly Line using the SVM Algorithm

Abstract: This paper presents a learning-based visual inspection met hod that addresses the need for an improved adaptability ofa visual inspection system for parts verification in panorama sunroof assembly lines. It is essential to ensure that the manyparts required (bolts and nuts, etc.) are properly installed in the PLC sunroof manufacturing process. Instead of humaninspectors, a visual inspection system can automatically perform parts verification tasks to assure that parts are properly installedwhile rejecting any that are improperly assembled. The proposed visual inspection method is able to adapt to changinginspection tasks and environmental conditions through an efficient learning process. The proposed system consists of two majormodules: learning mode and test mode. The SVM (Support Vector Machine) learning algorithm is employed to implement partlearning and verification. The proposed method is very robust for changing environmental conditions, and various experimentalresults show the effectiveness of the proposed method.Keywords: visual inspection, part verification, panorama sunroof, support vector machine

Effective application of statistical multiplexing in advanced T-DMB System

This paper deals with an enhancement method on the simultaneous multimedia broadcast capacity over advanced T-DMB(AT-DMB) broadcast channel which is constrained by wireless capacity and service rate variation. Especially we focus on the statistical multiplexing, which can effectively mitigate the individual programs service rate variation without service quality degradation. In the current T-DMB system, as each program encoder produces a fixed data rate for each single program, a statistical multiplex could enhance each programs achievable video quality. This paper proposes an efficient statistical multiplexing scheme for an AT-DMB system and shows illustrative simulation results of the proposed method.

Camera Calibration Method for an Automotive Safety Driving System

This paper presents a camera calibration method in order to estimate the lane detection and inter-vehicle distance estimation system for an automotive safety driving system. In order to implement the lane detection and vision-based inter-vehicle distance estimation to the embedded navigations or black box systems, it is necessary to consider the computation time and algorithm complexity. The process of camera calibration estimates the horizon, the position of the car`s hood and the lane width for extraction of region of interest (ROI) from input image sequences. The precision of the calibration method is very important to the lane detection and inter-vehicle distance estimation. The proposed calibration method consists of three main steps: 1) horizon area determination; 2) estimation of the car`s hood area; and 3) estimation of initial lane width. Various experimental results show the effectiveness of the proposed method.

Real-time Temporal Synchronization and Compensation in Stereoscopic Video

In this paper, we propose a real-time temporal synchronization and compensation algorithm in stereoscopic video. Many temporal asynchronies are caused in the video editing stage and due to different transmission delays. These temporal asynchronies can degrade the perceived 3D quality. The goal of temporal alignment is to detect and to measure the temporal asynchrony and recover synchronization of the two video streams. In order to recover synchronization of the two video streams, we developed a method to detect asynchronies between the left and the right video streams based on a novel spatiogram information, which is a richer representation, capturing not only the values of the pixels but their spatial relationships as well. The proposed novel spatiogram additionally includes the changes of the spatial color distribution. Furthermore, we propose a block-based method for detection of the pair frame instead of one frame-based method. Various 3D experiments demonstrate the effectiveness of the proposed method.

Study on Viewpoint Estimation for Moving Parallax Barrier 3D Display

In this paper, we present an effective viewpoint estimation algorithm for the Moving parallax barrier method of 3D display mobile device. Moving parallax barrier is designed to overcome the biggest problem, the limited view angle. To accomplish it, the position of the viewer`s eyes or face should be estimated with strong stability and no latency. We focus on these requirements in the poor performance of mobile processors. We used a pre-processing algorithm in order to overcome the various illumination changes. And, we combined the conventional Viola-Jones face detection method and Optical-flow algorithm for robust and stable viewpoint estimation. Various computer simulations prove the effectiveness of the proposed method.

Robust Viewpoint Estimation Algorithm for Moving Parallax Barrier Mobile 3D Display

This paper presents a robust viewpoint estimation algorithm for Moving Parallax Barrier mobile 3D display in sudden illumination changes. We analyze the previous viewpoint estimation algorithm that consists of the Viola-Jones face detector and the feature tracking by the Optical-Flow. The sudden changes in illumination decreases the performance of the Optical-flow feature tracker. In order to solve the problem, we define a novel performance measure for the Optical-Flow tracker. The overall performance can be increased by the selective adoption of the Viola-Jones detector and the Optical-flow tracker depending on the performance measure. Various experimental results show the effectiveness of the proposed method.