Chang-Rak Yoon

Augmented reality and representation in vehicle for safe driving at night

To reduce traffic accidents at night driving, we introduce an augmented reality system using head-up display (HUD) in a vehicle and discuss the effectiveness and usefulness of the proposed system with representation strategy. The proposed system embraces recognition of dynamic object by radar-vision fusion, forward collision warning by threat assessment, and representation strategy using HUD. For warning information to prevent collision at night driving, an augmented reality information aligned at drivers eye and overlapped with real world is shown through the proposed system, and the usefulness of the system is validated in the experiments.

Development of lane-level guidance service in vehicle augmented reality system

A driver should always pay attention to various situations which he faces inside and outside the car while driving and always keep eyes forward for safe driving. A vehicle augmented reality system provides the driver with safe and convenience information which matched with the real world through HUD. It makes the driver keep eyes forward road and other cars. We have developed the vehicle augmented reality system and various services and tested them on an indoor testbed This system provides driver with the forward situation awareness service, the lane change service, the lane departure service and so on displayed as the vehicle augmented reality. We describe the development contents and results of the developed vehicle augmented reality system focused on the lane-level guidance technology which determines the lane change or the lane departure using route information, road properties, the driving lane information, position, and so on.

Development of Augmented Forward Collision Warning System for Head-Up Display

In this paper, we present an augmented forward collision warning system for Head-Up Display (HUD). The convergence of HUD and Augmented Reality (AR) needs challenge and make innovative application in automobile industry. We focus on an advance of Forward Collision Warning System (FCWS) by the fusion of HUD and AR. The proposed system detects the frontal vehicles and pedestrians, assesses the imminent danger caused by the detected results, and alerts the augmented warnings to support safe driving practices.

Fuzzy inference and logical level methods for binary graphic/character image extraction

Thresholding is one of the most important approaches to image segmentation. It has been widely used to characterize many images containing some objects of reasonably uniform brightness against a background of differing brightness. Typical examples include handwritten/typewritten text and microscope bio-medical samples. Even though it can be applied to the image processing widely, there is no robust thresholding technique to circumvent noisy image. In this study, firstly, the published character/graphic image extraction techniques were reviewed and investigated and new thresholding technique such as fuzzy inference and modified logical level are proposed. In fuzzy inference technique, new methods of fuzzification, fuzzy rule, and defuzzification are introduced for lower error and high speed image binarization.

Augmented reality information registration for head-up display

In this paper, we present an information registration method for an Augmented Reality Head-Up Display (AR-HUD) system. The AR-HUD system superimposes driving assistance warnings and instructions that correctly match the real world according to the drivers view. AR information registration is a necessary procedure prior to the presentation of these warnings and instructions. This need is important in the field of safe driving, where the driver should keep his attention to specific situations and decide the upcoming maneuvers. This paper presents an automated AR information registration method that can establish correspondences between cameras and HUD on the assumption that the view of the driver are fixed toward the front. We also verify some of the use cases in the implemented AR-HUD system based on a test-bed vehicle.

Augmented reality based on driving situation awareness in vehicle

In this paper, we introduce a vehicle augmented reality (AR) system to present information of driving situation awareness in a vehicle. Today, manufacturers related with vehicles have been pointing to AR as a next-generation visualization technology for in-car driving displays. Such in-vehicle AR-based display systems are helpful in reducing driver distractions, thereby increasing driver safety, and provide intelligent interactions for enhancing driver convenience. The proposed system offers information of driving situation and warning to a driver through the augmented reality using head-up display. The system consists of several sub-modules such as sensor, vehicle/pedestrian recognition, vehicle state information, driving information, time to collision (TTC), threat assessment, warning strategy, and display modules. We have defined the threat level and the presentation of AR information based on TTC values and drivers preference throughout experiments. The proposed system have been installed to a test vehicle with a vehicle AR information system prototype and carried out in the real road environments. The proposed system demonstrates to offer intuitively danger information according to the presentation rules to a driver on real road.

Colorimetric characterization for digital camera by using multiple regression

Digital cameras provide a very useful way of capturing images for use in image processing and color communications. Many of techniques have developed to obtain high-quality color digital images. But the RGB signals generated by digital cameras are device dependent and not colorimetric. We have extracted the device independent characteristics of digital cameras and made transfer matrices to transform the RGB color signals into the CIE XYZ tristimulus values. The least square fitting technique was used to derive the coefficients of 3/spl times/m polynomial transfer matrices. We have investigated the degree of the polynomial and the number of color samples which produce the best result. The parameters affecting the characterization were also described.

Characterizing the desktop color printer with polynomial regression

In many application areas, the printer has played a key role of reproducing the digital color images. According to the characteristics of output devices, the uncorrected distortion has arisen in the reproduced color. This paper proposed the color correction based on polynomial regressions that represented the characteristics of the desktop color printer. In addition, the proper number of sample used in obtaining the regression equation was investigated. The experiments showed the possibilities of the proposed methods by comparing the /spl Delta/Eab value between the original color and the corresponding reproduced color.

A Study on Augmented Reality of a Vehicle Information Using Head-Up Display

A vehicle Augmented Reality Information System is introduced for augmented reality (AR) information such as collision warning and path guide in a vehicle. The proposed system decides information to be offered through forward situation awareness service and lane-level navigator service and provides AR information on windshield using head-up display in a vehicle for a driver. It is help to improve a drivers cognition.

Augmented reality for collision warning and path guide in a vehicle

Today, the automotive industry is focused on human-vehicle interaction related with safety and convenience while driving. One of those is head-up display (HUD) to implement augmented reality (AR) in a vehicle for information offering [Park et al. 2013]. In this study, we discuss a vehicle Augmented Reality Information System (vARIS) to offer information such as collision warning and path guide through augmented reality (AR) in a vehicle. This system realizes AR on windshield using projection type HUD and provides AR information on 50 inch HUD display for safety and convenience of a driver. A currently commercialized HUD has 3~5 inch display.