Time filter

Source Type

Seoul, South Korea

Lee S.,Intelligent Imaging Systems | Kim N.,Intelligent Imaging Systems | Paek I.,Nextchip Co. | Hayes M.H.,Intelligent Imaging Systems | Paik J.,Intelligent Imaging Systems
Digest of Technical Papers - IEEE International Conference on Consumer Electronics

In this paper, we present a robust motion-based object detection system that corrects for the motion of an unstable camera. Assuming that the global camera motion may be modeled as an affine transform of the image between two successive frames, the proposed method is able to correct for camera motion using an elastic registration algorithm (ER). The local motion is then estimated from a current image and affine-Transformed previous image. Finally, object regions are detected using the estimated local motions. Experimental results show that the proposed system is able to robustly detect moving objects in unstable imaging environment for consumer surveillance systems. © 2013 IEEE. Source

This study presents a spatially varying image registration method based on regions of the same depth. The proposed registration method uses phase correlation matching to measure colour shifting vectors (CSVs) between colour channels in a prespecified region of the same distance to the camera, and aligns colour channels of the corresponding region according to the CSV. The authors also present the foreground region detection method by using binary edge labelling and analysis of histograms of channel-shifting features. The major contribution of this study is 2-fold: (i) the proposed method can be considered as a region-wise approximated version of fully non-rigid registration, which is widely used in the medical imaging area, and (ii) it can compensate misalignment between red (R), green (G) and blue (B) colour channels caused by refraction and chromatic aberration of a multiple colour-filtered aperture (MCA) camera, which has been proposed as a single camera-based multifocusing system. Among various applications of non-rigid image registration, the proposed region-based registration method is particularly suitable for multifocusing images acquired by an MCA camera. In depth analysis of each step of the proposed algorithm is provided with experimental results, and its application to the MCA camera is also provided to realise efficient depth estimation and highly accurate multifocusing functions using a single camera. Without using joint histogram or geometric transformation, the proposed region-adaptive approach successfully approximates the fully non-rigid registration with significantly reduced amount of computation. © The Institution of Engineering and Technology 2013. Source

Lee S.,Chung - Ang University | Kim N.,Chung - Ang University | Jeong K.,Chung - Ang University | Paek I.,Nextchip Co. | And 2 more authors.

We present an efficient, robust moving object segmentation method that fully utilizes block motion information for high-resolution video surveillance systems. A high-resolution video surveillance system should satisfy two conflicting goals: (i) higher computational efficiency to manage the increasing amount of data and (ii) enhanced functionality in analyzing moving objects. In pursuit of both efficiency and functionality, we first quantize the orientation of motion vectors and then segment moving objects using adaptive block partitioning algorithm. We also present motion orientation histogram-based moving direction estimation. Major contribution of this work is the fully utilization of block motion information provided by either an image signal processing (ISP) chip or a digital signal processor (DSP) built-in software and the optimal representation of moving objects by the block divide-and-merge algorithm. Comparative experiments with the conventional video analysis algorithms show that the proposed method provides better segmentation results with the regular, efficient computational structure that can be easily embedded in an ISP chip or DSP software. © 2015 Elsevier GmbH. Source

Jeon J.,Chung - Ang University | Lee J.,Nextchip Co. | Lee E.,Chung - Ang University | Hayes M.H.,Chung - Ang University | Paik J.,Chung - Ang University
Proceedings - International Conference on Image Processing, ICIP

This paper presents a spatially adaptive super-resolution (SR) algorithm using homogeneous region analysis for minimizing undesired interpolation artifacts such as aliasing and jagged edges. The proposed regularized SR algorithm incorporates two constraints enforcing (i) disconnected edges to be reconnected and (ii) orientation-adaptive regularization. By combining two constraints in the regularization framework, the proposed SR algorithm can significantly reduce aliasing artifacts and, as a result, produce edge-preserved high-resolution (HR) images. In addition to the formulation of the regularized SR algorithm with hybrid constraints, experimental results show that the proposed SR algorithm improves peak-to-peak signal-to-noise ratio (PSNR) of the reconstructed HR images by up to 5[dB]over existing state-of-the-art SR methods. © 2012 IEEE. Source

Lee K.Y.,Seokyeong University | Kyung G.,Nextchip Co. | Park T.R.,Seokyeong University | Kwak J.C.,Seokyeong University | Koo Y.S.,Dankook University
2015 38th International Conference on Telecommunications and Signal Processing, TSP 2015

Mobile devices provide a more realistic image processing and various high spec features to satisfy users. So, mobile devices have been developing in the direction of computing acceleration by using a strong parallelism of GP-GPU processing. In this paper, a GP-GPU architecture is proposed based on stream processing architecture which has the advantage of high parallelism to enhance the image processing capability. The proposed GP-GPU architecture consists of 8 stream processors and has multi-banked cache memory structure. The results of verification shows that the proposed stream processor improves the performance of the integral image generation : 24.5%, 3×3 Gaussian filer mask : 4.7%, 5×5 Gaussian filter mask : 1.3% in comparison with ARM Cortex-A15 quad core. © 2015 IEEE. Source

Discover hidden collaborations