• Journal of information and communication convergence engineering
• /
• 제9권5호
• /
• pp.505-509
• /
• 2011

This paper attempts to classify known facial expressions and to establish the correlations between two regions (eye + eyebrows and mouth) in identifying the six prototypic expressions. Covariance is used to describe region texture that captures facial features for classification. The texture captured exhibit the pattern observed during the execution of particular expressions. Feature matching is done by simple distance measure between the probe and the modeled representations of eye and mouth components. We target JAFFE database in this experiment to validate our claim. A high classification rate is observed from the mouth component and the correlation between the two (eye and mouth) components. Eye component exhibits a lower classification rate if used independently.

• 전기학회논문지
• /
• 제66권8호
• /
• pp.1236-1241
• /
• 2017

We deposited Al-doped ZnO (ZnO:Al) thin films on glass substrates ($200mm{\times}200mm$) by using in-line magnetron sputtering system. Effects of various deposition parameters such as working pressure, deposition power and substrate temperature on optoelectronic characteristics including surface-texture etching profiles were carefully investigated in this study. We found that relatively low working pressure and high deposition power offered to obtain enhanced conductivity and optical transmittance. Haze properties showed similar trend with the transmittance. Furthermore, surface-texture etching study exhibited good morphologies when the films were deposited at $200-300^{\circ}C$. On the basis of these optimizations, we could find the deposition region that produces highly transparent and conductive properties including efficient light scattering capability.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.148-150
• /
• 2006

In this paper, we propose a new method for detection of mass-type breast cancer in dense mammogram. As the proposed method analyzes texture of the breast tissue using method by fusing Homogeneity and Ranklets, improve problem of traditional method. Homogeneity gives the measure of uniform density, and Ranklets determine orientation selective property at vertical, horizontal and diagonal in mass region. The proposed method is suitable to dense mammogram with tangled normal tissue and cancer tissue. SVM(Support Vector Machine) classifier is used for effective detection of mass-type breast cancer in dense mammogram.

• 전자공학회논문지B
• /
• 제31B권4호
• /
• pp.62-76
• /
• 1994

In this paper, a segmentation-based coding technique as applied to video sequences is proposed. A proposed method separates an image into contour and texture parts, then the visually-sensitive contour part is represented by chain codes and the visually-insensitive texture part is reconstructed by a representative motion vector of a region and mean of the segmented frame difference. It uses a change detector to find moving areas and adopts variable blocks to represent different motions correctly. For better quality of reconstructed images, the displaced frame difference between the original image and the motion compensated image reconstructed by the representative motion vector is segmented. Computer simulation with several video sequences shows that the proposed method gives better performance than the conventional ones in terms of the peak signal to noise ratio(PSNR) and compression ration.

• Current Optics and Photonics
• /
• 제4권3호
• /
• pp.200-209
• /
• 2020

We describe an approach that uses a bilateral filter to reduce the ringing artifact in motion-blurred restoration image. It takes into account the specific physical structure of the ringing artifact combined with the properties of the human visual system. To properly reduce the ringing artifact, each of the adjacent pixels is limited in a straight line which has a given direction. To protect the edges and the texture regions of an image, our algorithm divides the image into texture regions and flat regions, and the artifact reduction algorithm is only applied to the flat region. Finally, we use 8 typical images and 5 objective quality evaluation indices to evaluate our algorithm. Experimental results show that our algorithm can obtain better results in subjective visual effect and in objective image quality evaluation.

• 한국멀티미디어학회논문지
• /
• 제24권3호
• /
• pp.373-381
• /
• 2021

In this paper, we propose a classification model based on convolutional neural network(CNN) for predicting 2-year recurrence in non-small cell lung cancer(NSCLC) patients using preoperative chest CT images. Based on the region of interest(ROI) defined as the tumor internal and external area, the input images consist of an intratumoral patch, a peritumoral patch and a peritumoral texture patch focusing on the texture information of the peritumoral patch. Each patch is trained through AlexNet pretrained on ImageNet to explore the usefulness and performance of various patches. Additionally, ensemble learning of network trained with each patch analyzes the performance of different patch combination. Compared with all results, the ensemble model with intratumoral and peritumoral patches achieved the best performance (ACC=98.28%, Sensitivity=100%, NPV=100%).

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2007년도 하계학술대회 논문집 Vol.8
• /
• pp.517-518
• /
• 2007

A method to recognize unpaved road region using a 3D depth measurement system is proposed for mobile robots. For autonomous maneuvering of mobile robots, recognition of obstacles or recognition of road region is the essential task. In this paper, the 3D depth measurement system which is composed of a rotating mirror, a line laser and mono-camera is employed to detect depth, where the laser light is reflected by the mirror and projected to the scene objects whose locations are to be determined. The obtained depth information is converted into an image. Such depth images of the road region represent even and plane while that of off-road region is irregular or textured. Therefore, the problem falls into a texture identification problem. Road region is detected employing a simple spatial differentiation technique to detect the plain textured area. Identification results of the diverse situation of Nonlinear trail are included in this paper.

• 방송공학회논문지
• /
• 제20권4호
• /
• pp.521-532
• /
• 2015

본 논문은 효과적인 열화영상의 복원을 위해 쿼드트리 구조를 갖는 다중-스케일 블록 지역적 이진 패턴 기반의 영역검출기를 제시하고, 이를 통한 구간적 잡음 제거 기법을 제안한다. 구간적 잡음 제거 기법은 영상 내 전체 화소를 일정한 블록 단위의 영역으로 나누어 화소의 변화량에 따라 검출을 수행하는 다중-스케일 블록 영역 검출기를 쿼드트리 형태로 제시하고 검출된 영역 특성에 맞게 영상분석을 진행한다. 처리되는 영역들은 강한 변화량을 갖는 영역, 약한 변화량을 갖는 영역, 평탄한 영역의 세 가지로 분류되며 차례로 주성분분석, 양방향 필터, 구조-텍스쳐 영상 분해의 기법들이 잡음제거를 위해 적용된다. 객관적 실험결과를 통하여 기존 알고리즘들 보다 제안하는 구간적 잡음 제거 기법이 최대 신호-대-잡음비 측면에서 이득을 가지며, 주관적 화질 비교를 통해 세부정보들이 최대한 보존되어 있음과 동시에 평탄한 영역에 대해서도 왜곡이 거의 없는 향상된 복원영상이 얻어지는 것을 확인할 수 있었다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권1호
• /
• pp.22-33
• /
• 2005

본 논문에서는 얼굴 형태와 유사한 타원체형 모델의 가변적 블록크기를 기반으로 하는 영상정렬 알고리즘을 제안한다. 블록정합 수행 시 좌우 곡률만을 고려한 기존의 실린더형 원형모델 방법은 좌우 영상에 대하여 정확한 정렬을 수행하지만 상하 영상에 대해서는 사람의 두상모양이나 턱 구조의 특징을 반영하지 않았기 때문에 정렬오류가 발생한다. 본 논문에서 제안한 알고리즘은 타원체형 얼굴 모델의 상하 좌우의 곡률에 따라 가변적인 블록크기를 사용하는 블록정합 알고리즘으로서 영상의 상관관계를 이용하여 정확한 정렬을 할 수 있다. 그리고 정렬된 영상으로부터 얼굴 텍스쳐 영상을 생성하기 위하여 영상 모자익 기법을 사용한다. 이때 중첩된 영역에 따라 선형적인 가중치를 부여하여 영상을 스티칭하고 부분적으로 나타나는 고스트 효과를 제거함으로써 더 실감 있는 텍스쳐를 생성한다.

• 한국산업정보학회논문지
• /
• 제7권5호
• /
• pp.183-189
• /
• 2002

본 논문에서는 명함 영상에서 e-mail 영역을 검출하기 위한 방법을 제안한다. 먼저, 입력된 명함 영상으로부터 e-mail 주소의 특성 및 명함 영상 내에서의 텍스춰 특성을 분석하여 배경으로부터 텍스트 영역을 분할한다. 분할된 텍스트 영역으로부터 불필요한 정보들을 제거하고 e-mail 행으로 추정될 수 있는 몇 개의 후보 영역을 추출한다. 각 행에 대하여 연결화소들을 조사하여 개별 문자로 분리한 후, at-symbol(@)을 인식하여 e-mail 영역을 식별한다. 제안한 방식을 구현하여 여러 가지 폰트의 텍스트뿐만 아니라, 배경 영상, 캐리커쳐 등을 포함하고 있는 다양한 형태의 명함 영상에 대하여 적용하였으며, 98％ 이상의 성공률을 보여주고 있다.