Jang Woon Baek

Mono-camera based side vehicle detection for blind spot detection systems

This paper proposes a vision-based side vehicle detection for blind spot detection systems. The proposed algorithm uses a HoG cascade classifier in order to detect vehicles, and tracks the detected vehicles with Kalman filter. The proposed algorithm performs a periodical vehicle detection instead of every frame vehicle detection. And the proposed algorithm reduces the detecting image size by downscaling the original image and setting the region of interest where vehicles can exist. As a result, we can reduce the processing time for vehicle detection. Also, the proposed algorithm uses a false alarm reducing methods by control the reliability points at vehicle tracking. We evaluated the performance of the proposed algorithm in terms of processing time and detection ratio. At target board, the proposed algorithm has 40 frames per second, which meets the real time requirements of the ADAS systems. The detection ratio of the proposed algorithm is over 96 % at both original image size and downscale image size.

Mono-camera based vehicle detection using effective candidate generation

This paper proposes a novel on-road vehicle detection algorithm for Advanced Driver Assistance Systems. The proposed algorithm uses shadow and edge information of original image to generate the candidate vehicles. The proposed algorithm extracts the shadow region darker than the road, and creates an edge image using the Canny edge detector. It performs AND operation between the shadow image and the edge image. The candidate vehicles are verified by SVM classifier based on the HOG feature vector. The experimental results show that the proposed algorithm provides fast and accurate vehicle detection. And the number of candidates is reasonable for the vehicle verification.

Fast and reliable tracking algorithm for on-road vehicle detection systems

In this paper, we proposes a novel tracking algorithm combining Kalman Filter with mean-shift. Kalman Filter predicts the vehicle position in the next frame. Mean-shift finds the best candidate which has maximum similarity with the tracked vehicle in the predicted area. Kalman Filter updates its state value of vehicle position with the position of the best candidate from the mean-shift tracker. As a result, the proposed algorithm tracks the vehicle without local maximum problem of mean-shift tracker. The proposed algorithm is very fast because it does not perform the redetection process, and it has no detection misses because it finds the best candidate which has maximum similarity with the tracked vehicle in the predicted area. Also, the proposed algorithm has deleting and adding policies for the tracking list management. If a vehicle was consecutively detected in the previous frames, the proposed algorithm assumes that the vehicle exists although the vehicle is not detected in the predicted region at the current frame. If a vehicle was not detected in the previous frames consecutively, the proposed algorithm assumes that the vehicle does not exist although the vehicle is detected in the current frame. We evaluated the performance of the proposed algorithm in terms of processing time and detection ratio. At target board, the proposed algorithm has 40 frames per second, which meets the real time requirements of the ADAS systems. The detection ratio and processing time of the proposed algorithm outperformed our former work.

FFS-FR: Flash File System for Fast Recovery

A Flash File System for Fast Recovery (FFS-FR) is proposed in this paper. It uses the links between inodes in a flash file system based on the CFFS (Core Flash File System). The FFS-FR is able to restore the consistency of the metadata quickly by reading only the inodes instead of reading the whole flash area. The experimental results show that the recovery time of the proposed technique is 450, 150, 450 times faster than that of the YAFFS, JFFS2 and CFFS respectively.

Runtime debugging and monitoring tool for automotive software development

This paper prsesents a runtime debugging and monitoring tool for development of automotive applications based on OS EK/VDX op erating system. The tool indicates the debugging information for de velopers by controlling the Background Debug Mode interface which provides in-circuit debugging functionality in microcon trollers. Also, the tool provides special display functions for monitoring internal operating system data and properties of the OSEK objects by using the ORTI (OSEK Run-Time Interface) file that is a universal in terface for development tools to the OSEK/VDX operating system.

Software prober for power consumption analysis of embedded linux applications

This paper presents a software prober for power consumption analysis of embedded Linux applications. The software prober as a loadable Linux kernel module, which collects execution information of an application and analyzes the power consumption of each function for the application by synchronizing the collected execution information with the data from the multichannel power measurement system.

System generation tool for OSEK/VDX based automotive software development

This paper introduces software development platform and system generation tool based on OSEK/VDX standard which is established for automotive electric/electronic software development. It explains components of the platform and the structure of system generation tool we developed as a part of the platform in further detail. The OSEK/VDX operating system configuration of an application is described by writing a text script file of which the syntax and semantics given by the OSEK/VDX OIL specification. And the configuration is converted into lines of C codes using the system generator rather than loaded on startup. Also we present the design aspects of the tool related to development cycle of the platform and the tool itself.

Design and Implementation of OSEK/VDX System Generation for Automotive Domain Software Development Platform

This paper introduces software development platform and system generation tool based on OSEK/VDX standard which is established for automotive electric/electronic software development. It explains components of the platform and the structure of system generation tool we developed as a part of the platform in further detail. The OSEK/VDX operating system configuration of an application is described by writing a text script file of which the syntax and semantics given by the OSEK/VDX OIL specification. And the configuration is converted into lines of C codes using the system generator rather than loaded on startup. Also we present the design aspects of the tool related to development cycle of the platform and the tool itself.

An Adaptive Timeout Aggregation for Periodic Data Gathering in Wireless Sensor Networks

This paper proposes a novel timeout scheduling scheme for data aggergation in wireless sensor networks, which adaptively configures both the length of timeout and the collecting period according to the potential level of event occurrence. At normal times, the proposed scheme decreases the length of timeout that is a waiting time for messages from child nodes, and it makes longer the collecting period. That reduces the dissipated energy of the sensor node. Conversely, the proposed scheme increases the length of timeout and makes shorter the collecting period under event occurrences in order to fulfill more accurate data aggregation and early event detection. Extensive simulation work under various workloads has revealed that the proposed scheme enhances data accuracy and energy consumption by 15% and 25% respectively, as compared with the previous TAG data aggregation. Also, the proposed scheme detects the event occurrence six times earlier than the previous TAG and Cascading timeout aggregation.