• The Journal of Korean Association of Computer Education
• /
• v.9 no.2
• /
• pp.101-109
• /
• 2006

We propose an adaptive digital watermarking algorithm using HVS(human visual system) and SOM(Self-Organizing Map) among neural networks. This method adjusts adaptively the strength of the watermark which is embedded in different blocks according to block classification in DCT(Discrete Cosine Transform) domain. All blocks in 3 classes out of 4 are selected to embed a watermark. Watermark sequences are embedded in 6 lowest frequency coefficients of each block except the DC component. The experimental results are excellent.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.493-496
• /
• 2005

In this paper, we propose a new approach to detect candidate area of vehicle license plate using compounded color and vertical edge information it's own. Also, we propose a verification course, to compressed image generated by Fast DCT, using SVM to increase accuracy of extracted vechicle license plate area. Proposed method is consider that vehicle's position, become a object of it's license plate recognition, has various angle, scale and include enough environment informations. As a experimental results, proposed method shows a superior performance compared with the case that not includes verification course using SVM.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.653-656
• /
• 2019

현대 사회에서의 생체인식 기술의 발전을 통해 홍채, 지분 등의 정보를 사용하여 사용자의 신원확인 등이 가능하게 되었으며, 생체정보의 유출/위조 방지에 대한 중요성이 높아지면서 관련 연구가 활발히 이루어 지고 있다. 본 논문에서는 홍체 전체의 이미지에 삽입하거나 동공의 중심에 가까운 DCT(Discrete Cosine Transform)영역에 삽입방법, 동공 중심과 눈매의 영역을 검출하여 거리 및 DC 계수를 통하여 삽입 위치를 선택하는 방법 등 홍채영상 속 워터마크를 삽입하여 유출/위조된 홍채 영상을 검출하는 방법에서의 워터마크 삽입위치에 관한 방법을 연구하고, CASIA Irisimage Database ver 4.0의 워터마크를 삽입하여 NC(Normalized Correlation)의 값을 비교하여 워터마크 삽입 실험을 검증하였다.

• Journal of Broadcast Engineering
• /
• v.15 no.3
• /
• pp.434-453
• /
• 2010

In this paper, we propose 3DTE(3-Dimensional Transform Environment) that is based on 3DT(Dimensional Transform) that performs 2-dimensional integer DCT(Discrete Cosine Transform) based on $4{\times}4$ block and 1-dimensional integer DCT based on $4{\times}1$ block after collecting same frequency coefficients in neighboring $4{\times}4$ block and supports it's additional coding tools for high performance. The transform of 3DT can keep prediction error by using $4{\times}4$ block and reduce spatial redundancy additionally. The proposed 3DTE can provide coding tools to improve the coding efficiency with using 3DT. The performance of 3DTE compared to JM11.0 is average 3.58% and 5.40% bit savings for all test sequences and HD sequences, respectively, with keeping subjective video quality in High profile.

• Journal of Internet Computing and Services
• /
• v.6 no.2
• /
• pp.59-68
• /
• 2005

In this paper, for compression and transmission of 3D image, we propose an algorithm which executes 3D discrete cosine transform(DCT) for 3D images, hierarchically segments 3D blocks of an image in comparison with the original image and executes finite-state vector quantization(FSVQ) for each 3D block. Using 3D DCT coefficient feature, a 3D image is segmented hierarchically into large smooth blocks and small edge blocks, then the block hierarchy informations are transmitted. The codebooks are constructed for each hierarchical blocks respectively, the encoder transmits codeword index using FSVQ for reducing encoded bit with hierarchical segmentation information. The new algorithm suggested in this paper shows that the quality of Small Lobster and Head image increased by 1,91 dB and 1.47 dB respectively compared with those of HFSVQ.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.1
• /
• pp.11-17
• /
• 2023

This paper proposes a lightweight deep learning network for detecting an image splicing forgery. The research on image forgery detection using CNN, a deep learning network, and research on detecting and localizing forgery in pixel units are in progress. Among them, CAT-Net, which learns the discrete cosine transform coefficients of images together with images, was released in 2022. The DCT coefficients presented by CAT-Net are combined with the JPEG artifact learning module and the backbone model as pre-learning, and the weights are fixed. The dataset used for pre-training is not included in the public dataset, and the backbone model has a relatively large number of network parameters, which causes overfitting in a small dataset, hindering generalization performance. In this paper, this learning module is designed to learn the characterization depending on the DCT domain in real-time during network training without pre-training. The DCT RRU-Net proposed in this paper is a network that combines RRU-Net which detects forgery by learning only images and JPEG artifact learning module. It is confirmed that the network parameters are less than those of CAT-Net, the detection performance of forgery is better than that of RRU-Net, and the generalization performance for various datasets improves through the network architecture and training method of DCT RRU-Net.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.2C
• /
• pp.132-140
• /
• 2007

In this paper, we proposed a new technique to encode and decode the digital hologram. Since the digital hologram (or fringe pattern) is generated by interference of light, it has much different property from natural 2D (2 dimensional) images. First, we acquisite optical-sensed or computer-generated hologram by digital type, and then extract a chrominance component. The extracted digital hologram for coding is separated into segments to use multi-view properties. The segmented hologram shows the similar characteristics with picturing an object with 2D cameras in various point of view. Since fringe pattern is visually observed like as noise, we expect that the fringe pattern has poor coding efficiency. To obtain high efficiency, the segment is transformed with DCT (Discrete Cosine Transform) which resembles hologram generation process with high performance. Each transformed segment passes the 3D scanning process according to time and spatial correlation, and is organized into a video stream. Since the segment which correspond to frame of a video stream consists of the transformed coefficients with wide range of value, it is classified and re-normalized. Finally it is compressed with coding tools. The proposed algorithm illustrated that it has better properties for reconstruction of 16 times higher compression rate than the previous researches.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.4
• /
• pp.494-499
• /
• 2012

In this paper, a new method for detecting various objects that can be risks to safety navigation in sea environment is proposed. By analysing Infrared(IR) images obtained from various sea environments, we could find out that object regions include both horizontal and vertical direction edges while background regions of sea surface mainly include vertical direction edges. Therefore, we present an approach to detecting object regions considering horizontal and vertical edges. To this end, in the first step, image enhancement is performed by suppressing noises such as sea glint and complex clutters using a statistical filter. In the second step, a horizontal edge map and a vertical edge map are generated by 1-D Discrete Cosine Transform technique. Then, a combined map integrating the horizontal and the vertical edge maps is generated. In the third step, candidate object regions are detected by a adaptive thresholding method. Finally, exact object regions are extracted by eliminating background and clutter regions based on morphological operation.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.72-78
• /
• 2008

Although the H.264 video coding scheme is popular, it is not efficient for high-definition (HD) video coding because the size of its macroblock is relatively small for the HD video resolution. In this paper, we propose a new intra coding scheme based on the enlarged macroblock size. For the luminance component, intra $4{\times}4$ prediction and intra $16{\times}16$ prediction in H.264 are scaled into intra $8{\times}8$ prediction and intra $32{\times}32$ prediction, respectively. For the chrominance components, intra $8{\times}8$ prediction is extended to intra $16{\times}16$ prediction. Along with the $8{\times}8$ basic coding block size, an $8{\times}8$ integer discrete cosine transform (DCT) is used. Experimental results show that the proposed algorithm improves coding efficiency of the intra coding for HD video: PSNR gain by 0.23dB and bit-rate reduction by 5.32% on average.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.5
• /
• pp.42-53
• /
• 2000

Efficient implementation of pre-filtering has been an important issue in video sequence coding, because it can remove camera noise and improve coding efficiency dramatically This paper introduces a novel pre-filtering scheme that is performed inside a video encoder The proposed pre-filtering is based on the approximated generalized Wiener filtering and two-dimensional discrete cosine transform (DCT) factorization. and is achieved by scaling the DCT coefficients of original image blocks for intra block coding and those of motion-compensated error blocks for inter block coding, respectively Even though the pre-filtering operation is embedded in a video encoder, its additional computational complexity is marginal compared to the encoding process, and the overall architecture of the conventional video encoder is maintained In spite of its simplicity, the proposed pre-filtering scheme provides good filtering and coding performance for noisy video sequences.