• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.3
• /
• pp.71-78
• /
• 2011

This paper propose image steganalysis scheme based on independence between parent and child subband on the multi-layer wavelet domain. The proposed method decompose cover and stego images into 12 subbands by applying 3-level Haar UWT(Undecimated Wavelet Transform), analyze statistical independency between parent and child subband. Because this independency is appeared more difference in stego image than in cover image, we can use it as feature to differenciate between cover and stego image. Therefore we extract 72D features by calculation first 3 order statistical moments from joint characteristic function between parent and child subband. Multi-layer perceptron(MLP) is applied as classifier to discriminate between cover and stego image. We test the performance of proposed scheme over various embedding rates by the LSB, SS, BSS embedding method. The proposed scheme outperforms the previous schemes in detection rate to existence of hidden message as well as exactness of discrimination.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.57-65
• /
• 2010

In this paper, we present a universal steganalysis scheme. The proposed method extract features of two types. First feature set is extracted from histogram characteristic of the wavelet subbands. Second feature set is determined by statistical moments of wavelet characteristic functions. 3-level wavelet decomposition is performed for stego image and cover image using the Haar wavelet basis. We extract one features from 9 high frequency subbands of 12 subbands. The number of second features is 39. We use total 48 features for steganalysis. Multi layer perceptron(MLP) is applied as classifier to distinguish between cover images and stego images. To evaluate the proposed steganalysis method, we use the CorelDraw image database. We test the performance of our proposed steganalysis method over LSB method, spread spectrum data hiding method, blind spread spectrum data hiding method and F5 data hiding method. The proposed method outperforms the previous methods in sensitivity, specificity, error rate and area under ROC curve, etc.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.219-222
• /
• 2008

3차원 인체 스캐너로부터 얻어진 인체형상데이터는 여러 인체에 대한 3차원 평균 모델을 만들어 내는 등의 통계적 분석이나 자세 변경을 위해 필요한 내부 골격 구조와 골격과 피부조직 사이의 관계 등을 계산해 내기 어렵다. 또, 이러한 통계적 분석을 위해서는 각 모델 간의 상응 관계가 확립되어야 하지만 스캐너로부터 얻어진 인체 형상 데이터들은 측정 환경이나 대상에 따라 각각이 서로 상이한 기하학적 구조로 이루어져 있다. 본 논문에서는 템플릿 모델을 3차원 인체데이터에 맞도록 변형함으로써 다수의 인체 형상에 대하여 토폴로지를 일치시키도록 한다. 3차원 인체 데이터에 대해 템플릿 모델이 가장 근사한 형상이 되도록 하는 변형을 자동으로 찾아내기 위해서 표면 위에 정의된 특징점들을 사용한다. 또한, 기존에 찾아둔 특징점군 및 변형정보 데이터가 충분히 많다면 새로운 변형을 계산하는 데 유용하게 사용될 수 있음을 보인다. 이렇게 상응 관계가 확립된 모델들은 삼차원 벡터 공간의 점들의 집합으로 표현 및 통계적 분석이 가능하게 된다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.97-105
• /
• 2015

Content-adaptive steganography methods embed secret messages in hard-to-model regions of covers such as complicated texture or noisy area. Content-adaptive steganalysis methods often need high dimensional features to capture more subtle relationships of local dependencies among adjacent pixels. However, these methods require many computational complexity and depend on the location of hidden message and the exploited distortion metrics. In this paper, we propose an improved steganalysis method for content-adaptive steganography to enhance detection rate with small number features. We first show that the features form the difference between DCT coefficients are useful for analyzing the content-adaptive steganography methods, and present feature extraction mehtod using first-order Markov probability for the the difference between DCT coefficients. The extracted features are used as input of ensemble classifier. Experimental results show that the proposed method outperforms previous schemes in terms of detection rates and accuracy in spite of a small number features in various content-adaptive stego images.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.8
• /
• pp.397-402
• /
• 2007

In this paper, we apply a support vector machine(SVM) that incorporates an optimized nonlinear decision rule over different sets of feature vectors to improve the performance of statistical model-based voice activity detection(VAD). Conventional method performs VAD through setting up statistical models for each case of speech absence and presence assumption and comparing the geometric mean of the likelihood ratio (LR) for the individual frequency band extracted from input signal with the given threshold. We propose a novel VAD technique based on SVM by treating the LRs computed in each frequency bin as the elements of feature vector to minimize classification error probability instead of the conventional decision rule using geometric mean. As a result of experiments, the performance of SVM-based VAD using the proposed feature has shown better results compared with those of reported VADs in various noise environments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.7A
• /
• pp.1044-1049
• /
• 2000

Projection that can be used as a feature for image representation, includes much available informations such as approximated shape and location. But when we retrieve image using it, there are some disadvantage such as requiring much index data and making different length of projected vector for differenr image size. In order to overcome these problems, we propose a method of using block variance for the projected vector. We use block variance of the projection vector to localize the characteristics of image and to reduce the number of index data in database. Proposed algorithm can make use of statistical advantage through database including various size of images and be executed with fast response time in implementation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.12B
• /
• pp.2085-2093
• /
• 2000

본 논문에서는 초음파 영상의 대한 내용기반 검색을 위한 초음파 영상의 특징 추출방법과 추출된 특징들을 이용한 검색 방법에 대해 제안한다. 내용기반 초음파 영상 검색을 위한 특징들로 공간영역에서 히스토그램과 웨이브릿 변환후 각 부대역에서 통계적 특성을 추출한다. 웨이브릿 변환 영역에서 추출된 특성은 질의 영상과 유사한 영상의 특성 벡터 거리가 평균 특성 벡터 거리보다 작다는 특성을 가진다. 이러한 특성을 이용하여 일차 검색을 수행하여 그 결과를 공간영역의 히스토그램을 이용한 이차 검색을 위한 후보로 사용함으로써 이차 검색의 대상이 줄어들게 된다. 히스토그램을 이용한 검색은 대상이 많을수록 오류를 범할 가능성이 높아짐으로 검색대상을 줄인다는 것은 매우 중요한 일이다. 또한 히스토그램을 사용함으로써 영상내 의학적 객체의 이동이나 회전에 무관하게 검색을 수행할 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.295-297
• /
• 2003

본 논문에서는 각 프레임으로부터 추출된 통계적 특성을 이용하여 동영상의 분할방법과 분할된 각 장면에 대한 대표프레임을 추출하는 방법을 제안한다. 제안된 방법은 동영상의 각 프레임에 대하여 다해상도 분석을 실시하여 저주파 부 밴드로부터 히스토그램 특징을 추출하여 급격한 장면전환을 분할하는데 이용하였으며 또한 점진적인 장면전환을 검출하기 위해서는 고주파 부 밴드로부터 계산되는 화소의 분산치를 계산하여 특징벡터로 사용하였다. 실험의 결과를 통하여 제안된 동영상의 분할방법과 대표프레임 추출에 대한 알고리즘들이 동영상 자료의 분석 및 색인을 위한 효율적인 동영상 분할을 가능하게 하며, 차후 내용기반 영상과 비디오의 색인 및 검색을 위한 전처리의 단계로 사용되어질 수 있는 매우 유용한 방법임을 보였다.

• The Journal of the Korea Contents Association
• /
• v.6 no.9
• /
• pp.50-59
• /
• 2006

This paper proposes an automatic video monitoring system and its application to emergency detection by analyzing human behavior using neural network. The object area is identified by subtracting the statistically constructed background image from the input image. The identified object area then is transformed to the feature vector. Neural network has been adapted for analyzing the human behavior using the feature vector, and is designed to classify the behavior in rather simple numerical calculation. The system proposed in this paper is able to classify the three human behavior: stand, faint, and squat. Experiment results shows that the proposed algorithm is very efficient and useful in detecting the emergency situation.

• The Journal of the Korea Contents Association
• /
• v.3 no.1
• /
• pp.9-13
• /
• 2003

We propose a block motion estimation algorithm based on a statistical image feature for video sequences. The statistical feature of the reference block is obtained, then applied to select the candidate starting points (SPs) in the regular starting points pattern (SPP) by comparing the statistical feature of reference block with that of blocks which are spread ower regular SPP. The final SPs are obtained by their Mean Absolute Difference(MAD) value among the candidate SPs. Finally, one of conventional fast search algorithms, such as BRGDS, DS, and three-step search (TSS), has been applied to generate the motion vector of reference block using the final SPs as its starting points. The experimental results showed that the starting points from fine SPs were as dose as to the global minimum as we expected.