• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.1
• /
• pp.49-59
• /
• 2009

In this paper a watermarking algorithm is proposed to optimize the capacity of the digital watermark insertion in an experimental threshold using the characteristics of human visual system(HVS), adaptive scale factors, and weight functions based on discrete wavelet transform. After the original image is decomposed by a 3-level discrete wavelet transform, the watermarks for capacity optimization are inserted into all subbands except the baseband, by applying the important coefficients from the experimental threshold in the wavelet region. The adaptive scale factors and weight functions based on HVS are considered for the capacity optimization of the digital watermark insertion in order to enhance the robustness and invisibility. The watermarks are consisted of gaussian random sequences and detected by correlation. The experimental results showed that this algorithm can preserve a fine image quality against various attacks such as the JPEG lossy compression, noise addition, cropping, blurring, sharpening, linear and non-linear filtering, etc.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.9
• /
• pp.1165-1171
• /
• 2018

CNN requires a large amount of computation and memory in the process of extracting the feature of the object. Also, It is trained from the network that the user has configured, and because the structure of the network is fixed, it can not be modified during training and it is also difficult to use it in a mobile device with low computing power. To solve these problems, we apply a pruning method to the pre-trained weight file to reduce computation and memory requirements. This method consists of three steps. First, all the weights of the pre-trained network file are retrieved for each layer. Second, take an absolute value for the weight of each layer and obtain the average. After setting the average to a threshold, remove the weight below the threshold. Finally, the network file applied the pruning method is re-trained. We experimented with LeNet-5 and AlexNet, achieved 31x on LeNet-5 and 12x on AlexNet.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.167-170
• /
• 2007

동일한 집단에 속하는 개체를 다른 집단에 속하는 개체로부터 구별할 수 있는 염기의 특징을 해당 집단의 시그너쳐라고 한다. 학습 데이터는 두 집단에 속하는 염기서열들이고, 염기서열에 대한 시그너쳐는 개체를 다른 집단과 구별할 수 있는 위치의 염기들로 구성된 서열이다. 제안한 방법에서는 각 집단에 대해서 위치별로 염기의 발생빈도를 계산하고, 가장 발생빈도가 높은 염기를 결정한 다음, 다른 집단의 대응 위치에서 해당 염기의 빈도를 계산하여, 빈도차이가 지정한 분류임계값 이상이면, 해당 위치의 염기를 시그너쳐를 구성하는 특징으로 간주한다. 시그너쳐를 대한 임의의 염기서열에 대한 부합정도는 시그너쳐에 속하는 염기의 학습집단에서의 상대빈도값을 가중치로 하여 계산한다. 임의의 염기서열이 특정 집단에 속하는지 판단하기 위해서는 해당 집단의 시그너쳐에 대한 부합정도를 계산하게 되는데, 부합정도가 얼마이상이 되어야 해당 집단에 속하는 것으로 간주할지 기준이 되는 임계값을 엄밀도 임계값이라고 한다. 엄밀도 임계값은 학습 데이터 집합에 대해서 주어진 시그너쳐에 대한 엄밀도 임계값이 민감도와 특이도를 최대로 하는 것을 선택한다. 제안한 방법을 구현한 바이오인포매틱스 도구를 개발하여, 한국형 HIV-1 바이러스 시그너쳐 추출에 적용하여 분류특성이 우수한 시그너쳐를 추출할 수 있음을 확인하였다.

• Journal of Korea Multimedia Society
• /
• v.8 no.7
• /
• pp.961-973
• /
• 2005

In this paper, we proposed an adaptive steganography of color image using bit-planes and multichannel characteristics. Applying fixing threshold, if we insert information into all bit-planes of RGB channel, each channels showed different image quality. Therefore, we first defined the channel weight and the bit-plane weight to solve the fixing threshold problem of BPCS (bit-plane complexity steganography) method. We then proposed a new adaptive threshold method using the bit-plane weight of channels and the bit-plane complexity of cover image to increase insertion capacity adaptively In the experiment, we inserted information into the color images with the same image quality and same insertion capacity, and we analyzed the Insertion capacity and image quality. As a result, the proposed method increased the insertion capacity and improved the image quality than BPCS method.

• The Journal of Society for e-Business Studies
• /
• v.20 no.2
• /
• pp.155-166
• /
• 2015

RWR (Random Walk with Restart) is frequently used by many graph-based ranking algorithms, but it does not consider a signed graph where edges may have negative weight values. In this paper, we apply the Balance Theory by F. Heider to RWR over a signed graph and propose a novel RWR, Balanced Random Walk (BRW). We apply the proposed technique into the domain of recommendation system, and show by experiments its effectiveness to filter out the items that users may dislike. In order to provide the reasonable performance of BRW in the domain, we modify the existing Top-k algorithm, BCA, and propose a new algorithm, Bicolor-BCA. The proposed algorithm yet requires employing a threshold. In the experiment, we show how threshold values affect both precision and performance of the algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.9B
• /
• pp.823-833
• /
• 2003

To satisfy the requirements of the real-time Internet services, queue management and scheduling schemes should be enhanced to accommodate the delay and jitter characteristic of them. Although the existing queue management schemes can address the congestion problems of TCP flows, they have some problems in supporting real-time services. That is, they show performance degradation when burst traffics are continuously going into the system after the queue is occupied at a predefined threshold level. In addition, under the congestion state, they show large jitter, which is not a desirable phenomenon for real-time transmissions. To resolve these problems, we propose a SDRED (State Dependent Random Early Detection) and dynamic scheduling scheme that can improve delay and jitter performances by adjusting RED parameters such as ma $x_{th}$ and $w_{q}$ according to the queue status. The SDRED is designed to adapt to the current traffic situation by adjusting the max,$_{th}$ and $w_{q}$ to four different levels. From the simulation results, we show that the SDRED decreases packet delays in a queue and has more stable jitter characteristics than the existing RED, BLUE, ARED and DSRED schemes.mes.mes.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.4
• /
• pp.877-885
• /
• 2011

A collaborative filtering(CF) is the most widely used technique in recommender system. However, CF has sparsity and scalability problems. These problems reduce the accuracy of recommendation and extensive studies have been made to solve these problems, In this paper, we proposed a method that uses a weight so as to solve these problems. After creating a user-item matrix, the proposed method analyzes information about users who prefer the item only by using data with a rating over 4 for enhancing the accuracy in the recommendation. The proposed method uses information about the genre of the item as well as analyzed user information as a weight during the calculation of similarity, and it calculates prediction by using only data for which the similarity is over a threshold and uses the data as the rating value of unrated data. It is possible simultaneously to reduce sparsity and to improve accuracy by calculating prediction through an analysis of the characteristics of an item. Also, it is possible to conduct a quick classification based on the analyzed information once a new item and a user are registered. The experiment result indicated that the proposed method has been more enhanced the accuracy, compared to item based, genre based methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.23-26
• /
• 2019

사물인터넷(IoT)은 지금의 우리가 살고 일하는 모든 방식을 변화시키고 있다. IoT를 통해 데이터를 생성하고 저장하고 연결된 장치와 상호작용하여 비즈니스는 물론 우리의 일상 생활을 개선하고 있는 것이다. 무수히 많은 센서들이 연결된 세상은 센서들에 의해 그 어느 때보다 거대한 양의 데이터들을 생산하고 있다. JSON, XML 같은 트리 구조의 데이터 타입은 대량 데이터 저장 전송 교환 등에 주요하게 사용되는데 이는 트리 구조가 이형 데이터 간의 유연한 정보 전송과 교환을 가능하게 하기 때문이다. 반면에, 효용성 높은 정보나 감추어져 있는 정보들을 트리 구조의 대량 데이터들로부터 추출하는 것은 일반 데이터 구조에 비해 훨씬 어려우며 더 난해한 문제들을 발생시킨다. 본 논문에서는 트리 구조의 대량 스트리밍 데이터로부터 가중치가 부여된 주요한 부정 패턴들을 추출하기 위한 방법을 공식화한다.

• Journal of Broadcast Engineering
• /
• v.7 no.4
• /
• pp.368-375
• /
• 2002

In this paper, we apply multiresolution wavelet transform for the digital image watermarking. For better result, we insert watermark into the images and set a proper weighting factor for each subband based on HVS model. With proposed algorithm, we confirm that the image quality is satisfactory and this method outperforms the previous methods against image compression and cropping.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.125-127
• /
• 2000

협력적 여과는 사용자의 아이템에 대한 단계적 평가에 기초하여 그 평가 패턴이 유사한 사용자를 찾아 그 사용자들이 선호한 아이템을 상대방에게 교차 추천을 해주는 방법이다. 따라서, 유사한 사용자를 찾는 방법이 중요한 문제가 되며, 현재까지 여러 가지 방법들이 제안되어 왔다. 순수한 협력적 여과 방법은 n차원 공간에서 사용자를 모델링하여 가장 유사한 이웃을 찾는다. 이러한 모델링의 문제점은 사용자가 평가한 아이템의 집합은 전체 아이템의 집합에 비해서 극히 작으므로 유사한 사용자를 찾기 위해서는 충분한 수의 아이템에 대해서 평가해야 한다는 것이다. 따라서, 본 논문에서는 유사란 사용자를 찾기 위해서 충분한 수의 평가를 요구하는 명백하게 사용자의 평가를 비교하는 것 대신에 특징 가중치에 초하여 사용자를 비교하는 방법을 사용하고 사용하는 방법의 정확성을 높일 수 있는 임계값을 제안하고자 한다.