• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.343-350
• /
• 2002

A scene change detection which involves finding a cut between two consecutive shots is an important step for video indexing and retrieval. This paper proposes an algorithm for fast and accurate detection of abrupt scene changes in an MPEG compressed domain with minimal decoding requirements arid computational effort. The proposed method compares two successive DC images of I-frames for finding the GOP (group of picture) which contain a scene change and uses macroblock-coded type information contained in B-frames to detect the exact frame where the scene change occurred. The experiment results demonstrate that the proposed algorithm has better detection performance, such as precision and recall rate, than the existing method using all DC images. The algorithm has the advantage of speed, simplicity and accuracy. In addition, it requires less amount of storage.

• Journal of Korea Multimedia Society
• /
• v.1 no.2
• /
• pp.194-203
• /
• 1998

In this paper, a new DCT-VQ method is proposed which can solve the problems of VQ such as the degradation of edge and enormous calculations. VQ is carried in DCT domain but spatial domain in order to protect the degradation of edge. DCT makes high correlated image data decorrelated and the energy concentrated on a few coefficients. In DCT domain, the DC coefficient is quantized with 8 bits uniform scalar quantizer and the AC coefficients are divided to three regions and coded with vector qiantizer for considering edge components. For the decrease of the calculation and memory, the vectors for three region have small dimension of $1{\times}7$ and use the same codebook. Thus, the proposed method can fully express the edge components by considering AC coefficients in DCT domain and decrease the calculation and memory be reducing the dimension of vectors.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.145-148
• /
• 2007

The video shot detection based on video picture type is presented in this paper. The detection algorithm is used MPEG compressed video frame directly, not reconstructed the original image. For shot detection, I and P frame of MPEG video bit stream are classified. The detecting scene cuts at I pictures are detected by reconstructed DC image. While scene cuts at P picture frame by monitoring the percentage of Intra-macroblocks per P picture. Experimental results on the test video bit stream is shown the detection rate of $85\sim98%$ and searching time is 4 times faster than the previously known video shot detection algorithm on the decompressed video shot.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.23-32
• /
• 2011

In this paper, we develop a method for generating thumbnail images at hybrid domain combined the spatial domain and transform domain. The proposed method generates a pixel of a thumbnail image by adding a DC value of residual transform coefficients and an average value of an estimate block. For effectively calculating average values of estimate blocks, we propose a method for reconstructing the boundary pixels of a block. In comparison to the conventional method of decoding the bit stream then scaling down the decoded images, the developed method reduces the complexity by more than 60% while producing identical thumbnail images.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.3
• /
• pp.17-22
• /
• 2023

In this paper, we propose a black ice detection system using Convolutional Neural Networks (CNNs). Black ice poses a serious threat to road safety, particularly during winter conditions. To overcome this problem, we introduce a CNN-based architecture for real-time black ice detection with an encoder-decoder network, specifically designed for real-time black ice detection using thermal images. To train the network, we establish a specialized experimental platform to capture thermal images of various black ice formations on diverse road surfaces, including cement and asphalt. This enables us to curate a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Additionally, in order to enhance the accuracy of black ice detection, we propose a multi-scale dilation convolution feature fusion (MsDC-FF) technique. This proposed technique dynamically adjusts the dilation ratios based on the input image's resolution, improving the network's ability to capture fine-grained details. Experimental results demonstrate the superior performance of our proposed network model compared to conventional image segmentation models. Our model achieved an mIoU of 95.93%, while LinkNet achieved an mIoU of 95.39%. Therefore, it is concluded that the proposed model in this paper could offer a promising solution for real-time black ice detection, thereby enhancing road safety during winter conditions.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.77-84
• /
• 2021

Today, the restoration of cultural properties is done by applying the latest IT technology from relying on existing data and experts. However, there are cases where new data are released and the original restoration is incorrect. Also, sometimes it takes too long to restore. And there is a possibility that the results will be different than expected. Therefore, we aim to quickly restore cultural properties using DeepLearning. Recently, so the algorithm DcGAN made in GANs algorithm, and image creation, restoring sectors are constantly evolving. We try to find the optimal GAN algorithm for the restoration of cultural properties among various GAN algorithms. Because the GAN algorithm is used in various fields. In the field of restoring cultural properties, it will show that it can be applied in practice by obtaining meaningful results. As a result of experimenting with the DCGAN and Style GAN algorithms among the GAN algorithms, it was confirmed that the DCGAN algorithm generates a top image with a low resolution.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.4
• /
• pp.408-417
• /
• 2002

Gabor cosine and sine functions have widely been used to describe the human visual filters. This paper presents a new method to locally represent image frequency components using these functions. The parameters of basis functions are determined based on dc ripple and the sidelobe strength of step response. The resultant transform consisting of Gabor cosine and sine functions is compared with existing transforms by computing the joint effective width and by applying to the image reconstruction with the limited number of transformed coefficients. The experimental results show that the proposed transform has better performance than DGT and DCT.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.437-440
• /
• 2000

This paper proposes a method to improve video quality on images that have blocking artifacts at block boundary. Block image transform coding suffers from blocking artifact that is a main cause of degrading video quality because of the quantization error of transform coefficients in quantization process. filtering and DPCM for DC components have been widely used to reduce blocking artifact. Recently, lots of works focus on the technique that minimizes block effects using discontinuity of block boundaries. In this paper, image blurring in decoding stage is improved by adding compensation factor to each transformed blocks so that discontinuity of block boundaries can be decreased. The compensation factor is applied on each block without much loss of edge components.

• Journal of Broadcast Engineering
• /
• v.13 no.4
• /
• pp.489-500
• /
• 2008

We propose a novel high-definition content watermarking algorithm that is highly feasible in UCC (User Created Contents) environment. We begin by addressing an association between broadcasting content and UCC in a view of copyrights, then present watermark requirements by analyzing various UCC editing-effect. Also, we provide a brief review of previous watermarking techniques that are supposed to satisfy the requirements. Our proposed scheme inserts a invisible watermark into both DC and AC components on $8{\times}8$ block DCT domain and extracts them after synchronization using DC image. Experimental results show that our technique satisfies the requirements of invisibility and robustness to a variety of attacks such as rotation, scaling, cropping, and JPEG compression, etc. Note that the proposed scheme is highly resilient to UCC edit attacks that are combined by many different types of watermark attacks.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.4-7
• /
• 2010

3D 콘텐츠에 대한 관심이 증가함에 따라 3D 시청 및 제작에 대한 가이드라인의 필요성도 함께 증가하고 있다. 3D 안정 시청 가이드 라인은 3D 시청으로 인한 시청자의 시각적 불쾌감이나 피로감을 방지하는데 목적을 두고 있으며, 최근 일본의 3DC에서는 과도한 수렴-조절 불일치를 방지하기 위해 양안 시차 $1^{\circ}$를 쾌적 시차 범위로 권고하고 있다. 하지만 이 쾌적 시차는 절대적인 수치가 아니며, 콘텐츠의 특성 및 시청 조건에 따라 변하는 것으로 추정된다. 본 논문에서는 쾌적 시차를 갖는 3D 영상 콘텐츠에서 객체의 모션으로 인해 양안시차가 시공간적으로 변할 때, 야기되는 시각적 불편함의 변화에 대하여 관찰한다. 특히, 깊이 방향 모션 및 수평 방향 모션에서 객체의 속도 변화에 대한 시각적 피로감의 정도를 주관적 평가를 통하여 측정한다.