• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.8C
• /
• pp.790-800
• /
• 2009

For the improvement of coding efficiency, the H.264/AYC standard uses new coding tools such as 1/4-pel-accurate motion estimation, multiple references, intra prediction, loop filter, variable block size etc. Using these coding tools, H.264/AYC has achieved significant improvements from rate-distortion point of view compared to existing standards. However, the encoder complexity was greatly increased due to these coding tools. We focus on the complexity reduction method of intra macroblock mode selection. The proposed algorithm for fast intra mode selection calculates the edge activity in transform domain, and performs fast encoding of intra frame in H.264/AYC through the fast prediction mode selection of intra4x4 and chrominance blocks. Simulation results show that the proposed method saves about 59.76% for QCIF sequences and 65.03% for CIF sequences of total encoding time, while bitrate increase and PSNR decrease are very small.

• Journal of Broadcast Engineering
• /
• v.13 no.1
• /
• pp.137-151
• /
• 2008

H.264/AVC is showed high coding efficiency more than previous video coding standard by using new coding tools. Specially, Variable block-based motion estimation and Rate-Distortion Optimization are very important coding tools in H.264/AVC. These coding tools have high coding efficiency, however the encoder complexity greatly increase due to these coding tools. In this paper, we propose early SKIP mode decision and selective inter/intra mode decision to reduce the computational complexity. Simulation results show that the proposed method could reduce encoding time of the overall sequences by 30% on average than JM 10.2 without noticeable degradation of coding efficiency. Besides, the proposed method runs over twice as fast as the previous proposed Fast Coding Mode Selection method (FCMS)[5].

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.561-568
• /
• 2021

In recent years, autonomous driving technologies have become a high-value-added technology that attracts attention in the fields of science and industry. For smooth Self-driving, it is necessary to accurately detect an object and estimate its movement speed in real time. CNN-based deep learning algorithms and conventional dense optical flows have a large consumption time, making it difficult to detect objects and estimate its movement speed in real time. In this paper, using a single camera image, fast object detection was performed using the YOLOv5 algorithm, a deep learning algorithm, and fast estimation of the speed of the object was performed by using a local dense optical flow modified from the existing dense optical flow based on the detected object. Based on this algorithm, we present a system that can predict the collision time and probability, and through this system, we intend to contribute to prevent tram accidents.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.735-739
• /
• 2005

This paper presents a new system to estimate the head pose of human in interactive indoor environment that has dynamic illumination change and large working space. The main idea of this system is to suggest a new morphological feature for estimating head angle from stereo disparity map. When a disparity map is obtained from stereo camera, the matching confidence value can be derived by measurements of correlation of the stereo images. Applying a threshold to the confidence value, we also obtain the specific morphology of the disparity map. Therefore, we can obtain the morphological shape of disparity map. Through the analysis of this morphological property, the head pose can be estimated. It is simple and fast algorithm in comparison with other algorithm which apply facial template, 2D, 3D models and optical flow method. Our system can automatically segment and estimate head pose in a wide range of head motion without manual initialization like other optical flow system. As the result of experiments, we obtained the reliable head orientation data under the real-time performance.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.12
• /
• pp.1233-1240
• /
• 2010

In the flat panel display and semiconductor industries, the visual alignment system is considered as a core technology which determines the productivity of a manufacturing line. It consists of the vision system to extract the centroids of alignment marks and the stage control system to compensate the alignment error. In this paper, we develop a Kalman filter algorithm to estimate the alignment mark postures and propose a coarse-fine alignment control method which utilizes both original fine images and reduced coarse ones in the visual feedback. The error compensation trajectory for the distributed joint servos of the alignment stage is generated in terms of the inverse kinematic solution for the misalignment in task space. In constructing the estimation algorithm, the equation of motion for the alignment marks is given by using the forward kinematics of alignment stage. Secondly, the measurements for the alignment mark centroids are obtained from the reduced images by applying the geometric template matching. As a result, the proposed Kalman filter based coarse-fine alignment control method enables a considerable reduction of alignment time.

• Current Optics and Photonics
• /
• v.7 no.4
• /
• pp.408-418
• /
• 2023

Due to the technological limitations of infrared thermography, infrared focal plane array (IFPA) imaging exhibits stripe non-uniformity, which is typically fixed pattern noise that changes over time and temperature on top of existing non-uniformities. This paper proposes a stripe non-uniformity correction algorithm based on scene-adaptive nonlinear filtering. The algorithm first uses a nonlinear filter to remove single-column non-uniformities and calculates the actual residual with respect to the original image. Then, the current residual is obtained by using the predicted residual from the previous frame and the actual residual. Finally, we adaptively calculate the gain and bias coefficients according to global motion parameters to reduce artifacts. Experimental results show that the proposed algorithm protects image edges to a certain extent, converges fast, has high quality, and effectively removes column stripes and non-uniform random noise compared to other adaptive correction algorithms.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.1
• /
• pp.43-53
• /
• 1990

The manufacturing errors such as pitch error and run-out error in planetary gear system bring about the irregular displacement of the center of each gear, which cause the torqe variation, vibration and noise. In this study, the relation between manufacturing errors and error motions of the center of gear was analyzed, and it can be applied to identyfy the errors of gears by investigating the measured locus of the center of each gear. Also, another identification method of power spectrum estimation using FFT algorithm was introduced, which analyze the frequency of the measured error motions. The results show that the error of each gear had a corresponding unique frequency, therefore, this method proved to be more effective.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.2
• /
• pp.119-124
• /
• 1998

Block Matching Motion Estimation(BMME) Algorithm is demands a very large amount of computing power and have been proposed many fast algorithms. These algorithms are many problem that larger size of VLSI scale due to non-localized search block data and problem of non-reuse of input data for each processing step. In this paper, we designed systolic arry of high processing capacity, constraints input output pin size and reuse of input data for small VLSI size. The proposed systolic array is optimized memory access time because of iterative reuse of input data on search block and become independent of problem size due to increase of algorithm's parallelism and total processing elements connection is localized spatial and temporal. The designed systolic array is reduced O(N6) time complexity to O(N3) on moving vector and has O(N) input/output pin size.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2271-2280
• /
• 1999

In this paper, a variable motion estimation scheme based on HBMA(Hierarchical Block Matching Algorithm) to improve the performance and to reduce heavy computational and transmission load, is presented. The proposed algorithm is composed of four steps. First, block activity for each block is defined using the edge information of differential image between two sequential images, and then average block activity of the present image is found by taking the mean of block activity. Secondly, camera pan compensation is carried out, according to the average activity of the image, in the hierarchical pyramid structure constructed by wavelet transform. Next, the LUT classifying each block into one among Moving, No Moving, Semi-Moving Block according to the block activity compensated camera pan is obtained. Finally, as varying the block size and adaptively selecting the initial search layer and the search range referring to LUT, the proposed variable HBMA can effectively carries out fast motion estimation in the hierarchical pyramid structure. The cost function needed above-mentioned each step is only the block activity defined by the edge information of the differential image in the sequential images.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.389-391
• /
• 1999

영상 압축 기법에는 여러 가지 알고리즘을 적용되고 있다. 이런 알고리즘들에는 주파수 영역 중복을 제거하기 위한 DCT, 시간 중복성 제거를 위한 움직임 측정, 압축기법에 의해서 만들어진 정보를 부호화하는 VLC들이 있다. 이런 부호화 알고리즘들은 부호화기를 구현하는데 많은 시간을 요구하며 특히 움직임 추정은 부호화기의 절반에 가까운 시간을 소비한다. 움직임 측정 기술의 복잡도는 search algorithm, cost function, search range parameter의 요인으로 나타낼 수 있다. 본 논문에서는 기존의 Full Search 알고리즘 대신에 three-step 알고리즘을 사용하여 움직임 측정 시간을 줄였다. Full Search 알고리즘은 search area에서 모든 지역에 대해 cost function을 사용하여 이전 블록과 얼마나 유사한지를 조사한다. 따라서 이전 블록과 가장 유사한 부분을 찾는 좋은 방법이지만 그만큼 시간이 많이 사용한다. Three-step 알고리즘은 search area의 일정 지역에 대해 cost function를 사용하여 이전 블록과의 유사성을 찾는 fast 알고리즘이다. Three-step 알고리즘을 사용한 경우 기존의 full search 알고리즘을 사용할 때 보다 60% 정도의 시간이 단축되었다. 그리고 생성되는 압축 데이터의 크기는 full search 알고리즘을 사용할 때 보다 많이 차지한다. 생성되는 H.263파일의 화질에서는 Three-step 알고리즘을 사용한 경우일지라도 full search 알고리즘을 사용한 경우와 거의 비슷한 화질을 보여준다.