VADA Lab, Department of Electronic Engineering, SKKU

Office: 21206 1st Engineering Building, Phone: 031.290.7200

SungKyunKwan's VADA (Mobile System on Chip Design & Automation Lab.) under the direction of Prof. Jun Dong Cho studies the efficient solution of HW and SW Codesign of Mobile-IP(Intellectual Property) and SoC(System on Chip) design problems for developing low-power and high-speed telecommunication and multimedia algorithms and architectures. As designs are entering feature sizes of nanometer scales with multiple GHz performance targets, new problems are emerging for CAD tools. Also, as the portable electronics market moves to computational-intensive products like cellular telephones (with 3D, Mobile Health, Robot, ITS, etc), the need to focus on low-power design becomes critical to extend battery life. To achieve this goal, the existing system on chip design methodologies can be brought to create new design automation solutions efficiently. Finally,with computer-aided design tools, the algorithm is programmed using System C and VHDl/Verilog to be simulated for verification.

Research Area

Student Members and Guidelines


VADA Lab's Publications and Patents


Recent Publications

  • Fast Motion Estimation with Considering Motion of Moving Objects Using Histogram Preprocessing This paper presents digital image stabilization for the case especially where sudden and unexpected moving objects appear on the screen during the video recording. Our proposed method based on histogram preprocessing excludes motion of the moving objects during image stabilization, while conventional method does not consider this motion of objects. Our experimental result showed that performance of our method is better than conventional method while moving objects appear in the video with about 92% reduction in computing time. In this manner, our method is suitable for real-time processing on hardware. pdf


  • Illumination Removal by Using Low-pass Filter and Trimmed Normalization This paper presents digital image stabilization for the case especially where sudden and unexpected moving objects appear on the screen during the video recording. Our proposed method based on histogram preprocessing excludes motion of the moving objects during image stabilization, while conventional method does not consider this motion of objects. Our experimental result showed that performance of our method is better than conventional method while moving objects appear in the video with about 92% reduction in computing time. In this manner, our method is suitable for real-time processing on hardware. pdf


  • A Real-time Rectification using an Adaptive Differential Encoding for High-resolution Video
    Previously, various rectification methods using compressed lookup table have been studied for real-time hardware stereo vision system. These loss compression methods may occur distortion that could corrupt disparity estimation process. Differentially encoded lookup table method which is lossless compression has no distortion with reasonable compression ratio. However, the method is limited to low-resolution and low warping movement. In this paper, we propose an adaptively differential encoding method. Our proposed algorithm has approximately 18∼26% compression ratio which is comparable to the previous method and tolerates high-resolution video. pdf


  • Projects

    Development of 3D Vision system for automated manufacturing robot for assembling mobile IT products

    Real-time Digital Image Stabilization using Motion Sensors

    • To generate sharp and clean picture images, we perform image stabilization. We first acquire the translational and rotational motion vector of the previous frame and the next one in the sequential moving picture. These motion values are used to correct the blurred images. However, in order to determine these values, we are required massive amount of search range and computations. Motivated by these facts, in this paper, we devise a new real-time stabilization technique with taking advantage of both digital image stabilization and motion sensors. The technique uses pitch and yaw values of sensor data for setting the search range, and calculates the translational motion value using SAD (i.e., block matching algorithm) within the search range. Also, roll value is used to compute the rotational motion value. Our experimental results showed that our method successfully performs image stabilization reducing the search range and computational time by 64% compared with the conventional one.
    • Related Publications
      • Jun Hoe Heo, Jong Hak Kim, Dong Hun Lee, Yong Han Kim and Jun Dong Cho, Real-time Digital Image Stabilization using Motion Sensors for Search Range Reduction, to be presented at ISOCC, nov. 2012, Jeju pdf


    3D Auto-Convergence and its applications using Real-time Depth-map

      pdf
    • This project aims to identify and develop an quality-enhanced 3D image processing, utilizing 3D depth information acquired by 3D camera of Mobile Devices. Throughout our research works, following techniques is being developed.
      • 1) Improvement of quality for real-time depth-map acquisition system to be applied for core ingredients of 3D camera system.
        • Rectification, real-time depth-map, Segmentation system realization using Stereo Camera : Algorithm and hardware design
        • Auto-convergence system utilizing real-time depth-map acquisition: Improvement of quality and Algorithm development aiming for HD 30 fps.
        • 3D Asymmetric Stereo Registration, Depth focus: VGA, 30fps
      • 2) Innovative technology of real-time high speed and cycle-accurate low power RTL structure will be exploited and verified using FPGA.
      • 3) Providing guideline for 3D effects and reduction of visual fatigue
    • Related Publications
      • Jae Gon Kim and Jun Dong Cho, "Simplified Relative Model to Measure Visual Fatigue in a Stereoscopy", IEICE TRANS. FUNDAMENTALS/COMMUN./ELECTRON./INF. & SYST., Vol.E94-A,No.12, to be published on Dec. 2011 pdf
      • Jae Gon Kim, Jong Hak Kim, Hun Ho Ham, Jueng Hun Kim, Chan Oh Park, Soon Suk Park, and Jun-Dong Cho, "A Real-Time Virtual Re-convergence Hardware Platform", ITC-CSCC 2011, pp.133-126, June 19-22, Gyeongju, Korea pdf
      • Chan Oh Park, Tahraoui Khaled, Jueng Hun Kim, and Jun Dong Cho, "Hardware Mass Object Analyser Implementation for Stereo Camera", SPA 2011 pdf
      • Jong Hak Kim, Chan Oh Park, Jueng Hun Kim, and Jun Dong Cho, "A Real-time 3D Disparity-map Acquisition hardware Architecture for a Multi-sliding-window-operation", SPA 2011 pdf
      • Jueng Hun Kim, Jong Hak Kim, Hun Ho Ham, and Jun Dong Cho,"Real-time Smoothing Filter for Three Dimensional Disparity Map Algorithm and Hardware Implementation", The 54th IEEE Midwest Symposium on Circuits and Systems, Yonsei Univ., Seoul Korea. pdf
      • Deuk Hyun Park, Hyoung Seok Ko, Jae Gon Kim, and Jun Dong Cho, "Real Time Rectification Using Differentially Encoded Lookup Table", ACM International Conference on Ubiquitous Information Management and Communication, Feb.2011, Sungkyunkwan Univ. Korea pdf
      • Jueng Hun Kim, Chan Oh Park, and Jun Dong Cho, "Hardware Implementation for Real-time Census 3D dispairty map Using dynamic Search Range", SPA 2011 pdf

    Multi-purpose 3D Multimedia SoC for Robot Applications

    • This project aims to develop a quality-enhanced 3D image processing, including 3D camera sensing and robot's cognitive recognition/modeling. Throughout our research works, the following techniques will be developed.
      • Improvement of quality for real-time 3D camera acquisition and recognition system and its real-time realization.
      • Real-time Realization of 3D graphic acceleration Engine
      • Development of Multi-Master System Bus for real-time System on Chip
      • Innovative high speed and cycle-accurate low power RTL structure will be exploited and verified using FPGA.
      • Related Publications
        • Byung Joo Hong, Chan Oh Park, and Jun-Dong Cho, "A Real-time Compact Structured-light based Range Sensing System", Journal of Semiconductor Technology and Science, June, 2012




    VADA Lab's Previous Projects

    Graduate Courses

    1. VLSI Design for Digital Signal Processing

    Undergraduate Courses

    Previous Industrial Liason Courses