• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2012년도 춘계학술발표대회논문집
• /
• pp.30-33
• /
• 2012

대사성 골 질환인 골다공증(Osteoporosis)의 조기 진단을 위해 X 선 영상에서 골 밀도를 측정하는 방법이 최근 연구되고 있다. 골 밀도는 X 선 영상에서 뼈가 분리되고, 분리된 영역에서의 픽셀에 의해 BMD가 측정되는데, 개선된 영상에서의 정밀한 뼈 추출이 주요한 요소이므로 X 선 영상의 개선은 골다공증의 조기 진단을 위해 필수적이다. 본 논문에서는 sparse 표현법을 도입하여 X 선 영상을 개선시키는 방법을 제안한다. 실험을 통해 제안한 방법의 결과가 기존의 방법인 웨이블릿 BayesShrink에 비해 개선됨을 CNR(Contrast to Noise Ratio)을 통해 확인하였다.

• Journal of Multimedia Information System
• /
• 제1권1호
• /
• pp.1-10
• /
• 2014

In this paper, we propose a new view synthesis technique for coding of multi-view color and depth data in arbitrary camera arrangements. We treat each camera position as a 3-D point in world coordinates and build clusters of those vertices. Color and depth data within a cluster are gathered into one camera position using a hierarchical representation based on the concept of layered depth image (LDI). Since one camera can cover only a limited viewing range, we set multiple reference cameras so that multiple LDIs are generated to cover the whole viewing range. Therefore, we can enhance the visual quality of the reconstructed views from multiple LDIs comparing with that from a single LDI. From experimental results, the proposed scheme shows better coding performance under arbitrary camera configurations in terms of PSNR and subjective visual quality.

• 한국산업정보학회논문지
• /
• 제5권1호
• /
• pp.1-15
• /
• 2000

In this paper, an efficient system for finding answers to a given path algebra expression in a directed acylic graph is discussed more particulary, in a multimedia presentration graph. Path algebra expressions are formulated using revised versions of operators next and until of temporal logic, and the connected operator. To evaluate queries with path algebra expressions, the node code system is proposed. In the node code system, the nodes of a presentation graph are assigned binary codes (node codes) that are used to represent nodes and paths in a presentation graph. Using node codes makes it easy to find parent-child predecessor-sucessor relationships between nodes. A pair of node codes for connected nodes uniquely identifies a path, and allows efficient set-at-a-time evaluations of path algebra expressions. In this paper, the node code representation of nodes and paths in multimedia presentation graphs are provided. The efficient algorithms for the evaluation of queries with path algebra expressions are also provided.

• 한국멀티미디어학회논문지
• /
• 제6권6호
• /
• pp.985-990
• /
• 2003

본 논문에서는 3차원 사람얼굴의 굴곡표면에 대하여 특징 값들을 추출하여 회전벡터를 이용하여 회전한 후 그들을 분석, 표현하는 방법을 제안한다. 또한 실험을 통하여 정확하게 추출된 굴곡표면의 특징 값들은 3차원 사람얼굴을 세그멘테이션 하는데 적용되었다 사람얼굴의 표면은 메쉬(mesh)모델을 사용하여 파라메타를 계산, 추출하였으며, 추출된 특징 값들은 얼굴표면을 Gaussian과 Mean 곡면 분류표(classification)를 사용하여 임계 값을 사용하지 않고 3D 얼굴표면을 세그멘테이션 하였다.

• 한국멀티미디어학회논문지
• /
• 제17권12호
• /
• pp.1494-1502
• /
• 2014

Since smartphones are utilized in the ranges from personal usages to governmental data exchanges, known but not patched vulnerabilities in smartphone operating systems are considered as major threats to the public. To minimize potential security breaches on smartphones, it is necessary to estimate possible security threats. So far, there have been numerous studies conducted to evaluate the security risks caused by mobile devices qualitatively, but there are few quantitative manners. For a large scale risk evaluation, a qualitative assessment is a never ending task. In this paper, we try to calculate relative risk levels triggered by software vulnerabilities from unsecured smartphone operating systems (Android and iOS) among 51 Asian countries. The proposed method combines widely accepted risk representation in both theory and industrial fields. When policy makers need to make a strategic decision on mobile security related agendas, they might find the presented approach useful.

• 한국멀티미디어학회논문지
• /
• 제17권4호
• /
• pp.466-477
• /
• 2014

Many image hiding schemes based on least significant bit (LSB) transformation have been proposed. One of the LSB-based image hiding schemes that employs diamond encoding was proposed in 2008. In this scheme, the binary secret data is converted into base n representation, and the converted secret data is concealed in the cover image. Here, we show that this scheme has two vulnerabilities: noticeable spots in the stego-image, i.e., a non-smooth embedding result, and inefficiency caused by rough re-adjustment of falling-off-boundary value and impractical base translation. Moreover, we propose a new scheme that is efficient and produces a smooth and high quality embedding result by restricting n to power of 2 and using a sophisticated re-adjustment procedure. Our experimental results show that our scheme yields high quality stego-images and is secure against RS detection attack.

• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.126-133
• /
• 2017

The conventional hand-crafted features used to track objects have limitations in object representation. Convolutional neural networks, which show good performance results in various areas of computer vision, are emerging as new ways to break through the limitations of feature extraction. CNN extracts the features of the image through layers of multiple layers, and learns the kernel used for feature extraction by itself. In this paper, we use the feature map extracted from the convolution layer of the convolution neural network to create an outline model of the object and use it for tracking. We propose a method to adaptively update the outline model to cope with various environment change factors affecting the tracking performance. The proposed algorithm evaluated the validity test based on the 11 environmental change attributes of the CVPR2013 tracking benchmark and showed excellent results in six attributes.

• 한국멀티미디어학회논문지
• /
• 제21권4호
• /
• pp.478-484
• /
• 2018

The choice of the motion features influences the result of the human action recognition method directly. Many factors often influence the single feature differently, such as appearance of the human body, environment and video camera. So the accuracy of action recognition is restricted. On the bases of studying the representation and recognition of human actions, and giving fully consideration to the advantages and disadvantages of different features, the Dense Trajectories(DT) algorithm is a very classic algorithm in the field of behavior recognition feature extraction, but there are some defects in the use of optical flow images. In this paper, we will use the improved Dense Trajectories(iDT) algorithm to optimize and extract the optical flow features in the movement of human action, then we will combined with Support Vector Machine methods to identify human behavior, and use the image in the KTH database for training and testing.

• 한국멀티미디어학회논문지
• /
• 제21권7호
• /
• pp.787-794
• /
• 2018

We propose a new framework which expresses the mist and scratches of cracked ice by an impact. We combine the grid projection technique, boundary particles method, and level-set method commonly used in fluid simulations to determine the region on the surface of an ice object which is affected by a collision. Mist is then generated in proportion to the impact, and immediately diffused, using a geodesic distance field to limit dissipation. The gradient of the mist is subsequently used to create realistic patterns of scratches and elongated air bubbles. Cracks of the ice object can also be considered, and the density of the mist made to vary realistically between fragments. As a result, our method not only represents high-quality ice effects, but also allows easy integration into existing rigid body simulation solvers.

• 한국멀티미디어학회논문지
• /
• 제23권2호
• /
• pp.216-226
• /
• 2020

In this paper, a stacked convolutional autoencoder model is proposed in order to classify Alzheimer's disease with high accuracy in PET/CT images. The proposed model makes use of the latent space representation - which is also called the bottleneck, of the encoder-decoder architecture: The input image is sent through the pipeline and the encoder part, using stacked convolutional filters, extracts the most useful information. This information is in the bottleneck, which then uses Softmax classification operation to classify between Alzheimer's disease, Mild Cognitive Impairment, and Normal Control. Using the data from Dong-A University, the model performs classification in detecting Alzheimer's disease up to 98.54% accuracy.