• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권3호
• /
• pp.99-108
• /
• 2021

차량 내부에는 조명 변화, 부분적인 가림 및 운전자의 상태 변화와 같은 제한되지 않은 조건들이 존재한다. 본 논문에서는 비 제약적인 차량 환경에서의 운전자 얼굴 인증 시스템을 제안한다. 제안한 방법은 차량 내부 및 외부의 조명 변화에 따라 발생하는 얼굴 이미지의 변화를 최소화하기 위해서 근적외선(NIR) 카메라를 사용한다. 특히 정면에서의 강한 빛에 노출된 얼굴 이미지를 처리하기 위해서, 학습 이미지의 평균과 분산을 사용하여 정상적인 얼굴 이미지로부터 빛에 과다하게 노출된 이미지로 변환하여 사용한다. 따라서 정상적인 조명에서의 얼굴 분류기와 강한 정면광에서의 얼굴 분류기를 각각 동시에 만들어진다. 제안하는 얼굴 분류기는 얼굴 랜드마크를 추출하고 각 랜드마크의 신뢰도 점수를 합산하여 얼굴을 최종적으로 식별한다. 특히 각 랜드마크를 인식하여 부분적인 얼굴 가림에 강하기 때문에 안경이나 선글라스를 착용하는 상황에서도 높은 성능 향상이 가능하다. 즉 가려지지 않은 남은 랜드마크의 점수를 사용하여 운전자를 인식할 수 있다. 또한 등록 운전자와 미등록 운전자 간의 관계를 고려한 새로운 인식 거부 방법과 새로운 평가 방법을 논문에서 제안한다. 자체 취득한 데이터 셋, 공인된 PolyU 및 ORL 데이터 셋으로 실험한 결과 제안한 방법이 효과적임을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권8호
• /
• pp.2895-2912
• /
• 2014

Content-based image retrieval has been the most important technique for managing huge amount of images. The fundamental yet highly challenging problem in this field is how to measure the content-level similarity based on the low-level image features. The primary difficulties lie in the great variance within images, e.g. background, illumination, viewpoint and pose. Intuitively, an ideal similarity measure should be able to adapt the data distribution, discover and highlight the content-level information, and be robust to those variances. Motivated by these observations, we in this paper propose a probabilistic similarity learning approach. We first model the distribution of low-level image features and derive the free energy kernel (FEK), i.e., similarity measure, based on the distribution. Then, we propose a learning approach for the derived kernel, under the criterion that the kernel outputs high similarity for those images sharing the same class labels and output low similarity for those without the same label. The advantages of the proposed approach, in comparison with previous approaches, are threefold. (1) With the ability inherited from probabilistic models, the similarity measure can well adapt to data distribution. (2) Benefitting from the content-level hidden variables within the probabilistic models, the similarity measure is able to capture content-level cues. (3) It fully exploits class label in the supervised learning procedure. The proposed approach is extensively evaluated on two well-known databases. It achieves highly competitive performance on most experiments, which validates its advantages.

• Journal of Korean Neurosurgical Society
• /
• 제40권3호
• /
• pp.180-185
• /
• 2006

Objective : Middle cerebral artery occlusion[MCAO] has widely been used to produce ischemic brain lesions. The lesions induced by MCAO tend to be variable in size because of the variance in the collateral blood supply found in the mouse brain. To establish a less invasive and reproducible focal ischemia model in mice, we modified the technique used for rat photo thrombosis model. Methods : Male C57BL/6 mice were subjected to focal cerebral ischemia by photothrombosis of cortical microvessels. Cerebral infarction was produced by intraperitoneal injection of Rose Bengal, a photosensitive dye and by focal illumination through the skull. Motor impairment was assessed by the accelerating rotarod and staircase tests. The brain was perfusion-fixed for histological determination of infarct volume four weeks after stroke. Results : The lesion was located in the frontal and parietal cortex and the underlying white matter was partly affected. A relatively constant infarct volume was achieved one month after photothrombosis. The presence of the photothrombotic lesion was associated with severe impairment of the motor performance measured by the rotarod and staircase tests. Conclusion : Photothrombotic infarction in mice is highly reproducible in size and location. This procedure can provide a simple method to produce cerebral infarction in a unilateral motor cortex lesion. In addition, it can provide a suitable model for study of potential neuroprotective and therapeutic agents in human stroke.

• 한국조경학회지
• /
• 제25권3호
• /
• pp.210-221
• /
• 1997

This study was conducted to investigte the visual preference and response between introduced and native trees for utilization of the south native trees in interior space. 1. The intensity of illumination was below 500lux at 24 buildings of interior spaces surveyed. Generally the temperature was 24~$26^{\circ}C$, and humidity was 60~70%. 2. Total number of introduced tree species were 9 familiar, 16 genus, 18 species, especially palm genus were the largest tree among them at these surveyed interior spaces, but native trees were only 2 species. The introduced tree of 2 to 3m in height was 46 percent, and the case of 1.2 to 2m was 42 percent. 3. The mean of visual preference was the highest at Rhapis excelsa and howeia belmoreana, and followed with Neolitsea sericea, Daphniphllum macropodum, Listsea japonicum. In the mean of visual preference for selected tree groups, native tree was not different as compared with introduced tree. 4. Visual image factor of the native and introduced trees at interior spaces was classified by individual factor and emotional factor. These 2 factors were shown 66.5% total variance. The native tree was strongly recognized on individual factor, as , and introduced tree was also strongly recognized on emotional factor. 5. In the analysis of visual preference and image, a difference between the selected groups of the introduced and native trees were not shown significant, and the introduced trees were similar to the native trees on the visual image. This result was as follow There was an analogy between Cinnamomum camphora, Ligustrum japonicum and Ficus retusa, Ficus benjamina. There is an analogy between Castanopsis cuspidata var. sieboldii, Cinnamomum japonicum and Ficus retusa, Ficus benjamina. There was an analogy between Elaeocarpus sylvestris var. ellipticus, Fatsia japonica and Rhapis excelsa, Howeia belmoreana. There was an analogy between Neolitsea sericea, Daphniphyllum macropo여, Listsea japonica and Rhapis excelsa, Howeia belmoreana. There was an analogy between Elaeocarpus sylvestris var. ellipticus, Fatsia japonica and Dracaena fragranse, Ficus elastica, Monstera deliciosa.

• 천문학논총
• /
• 제23권1호
• /
• pp.1-12
• /
• 2008

Understanding of the basic characteristics of an astronomical instrument is a prerequisite to obtaining reliable data from the instrument. We have analyzed more than 1,000 calibration images from the Fairchild 486 CCD (hereafter the Maidanak 4k CCD system) attached to the AZT-22 1.5m telescope at Maidanak Astronomical Observatory in Uzbekistan. The Maidanak 4k CCD system supports three readout modes through 1, 2, or 4 amplifiers. In most cases observers use 4-amplifier readout mode to save time. We have tested the stability and seasonal variation of zero levels and confirm that two quadrants of the images (Amp 1 & 2) show no appreciable seasonal variation. but the other two quadrants (Amp 3 & Amp 4) show an evident seasonal variation in the bias level. The Cryo Tiger, the cooling system used at the Maidanak 4k CCD system, maintains the CCD temperature at -108'E, and effectively suppresses the dark electrons. The mean value versus the variance plot of the flat images does not show the expected relation for an ideal Poisson noise distribution and this is attributed to the large variation in quantum efficiency between different pixels. In addition, we confirm that there is no appreciable difference in gain between readout amplifiers, but there is a large variation in quantum efficiency across CCD chip especially in U. Due to the finite length of shutter opening and closing time, the effective exposure time varies across the science images. We introduce two parameters to quantify the effect of this uneven illumination and present a method to remove these effects. We also present a method to remove the interference patterns appearing in the images obtained with longer wavelength filters and investigate the spatial variation of the point spread function.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1224-1242
• /
• 2018

Aiming at motion analysis and compensation, it is essential to conduct motion detection with images. However, motion detection and tracking from low-altitude images obtained from an unmanned aerial system may pose many challenges due to degraded image quality caused by platform motion, image instability and illumination fluctuation. This research tackles these challenges by proposing a modified joint transform correlation algorithm which includes two preprocessing strategies. In spatial domain, a modified fuzzy edge detection method is proposed for preprocessing the input images. In frequency domain, to eliminate the disturbance of self-correlation items, the cross-correlation items are extracted from joint power spectrum output plane. The effectiveness and accuracy of the algorithm has been tested and evaluated by both simulation and real datasets in this research. The simulation experiments show that the proposed approach can derive satisfactory peaks of cross-correlation and achieve detection accuracy of displacement vectors with no more than 0.03pixel for image pairs with displacement smaller than 20pixels, when addition of image motion blurring in the range of 0~10pixel and 0.002variance of additive Gaussian noise. Moreover,this paper proposes quantitative analysis approach using tri-image pairs from real datasets and the experimental results show that detection accuracy can be achieved with sub-pixel level even if the sampling frequency can only attain 50 frames per second.

• Restorative Dentistry and Endodontics
• /
• 제46권4호
• /
• pp.60.1-60.8
• /
• 2021

Objectives: The aim of this study was to verify the match between 5 shades of composites from different manufacturers with a shade guide and among the systems using a portable spectrophotometer. Materials and Methods: Shade measurements were performed on specimens of Z350 XT (3M ESPE), Charisma Diamond (Heraeus Kulzer GmbH), Esthet X-HD (Dentsply Caulk), and Empress Direct (Ivoclar-Vivadent) for shades A1, A2, A3, B1, and C3 using a Vita Easyshade spectrophotometer (Vita Zahnfabrik) against a white background. Corresponding shades of Vitapan Classical (Vita Zahnfabrik) guide were measured likewise and shade variation (ΔE) was calculated based on International Commission on Illumination L^*a^*b^* parameters. The ΔE of the composites in each shade was compared by one-way analysis of variance and Tukey's post hoc test (α = 0.05). Results: All composites presented ΔE > 3.7 compared with the shade guide. Variation in shades A3, B1, and C3 was significantly different for all composites. ΔE of Z350 XT was significantly lower for A1 than for the other shades, whereas ΔE of Z350 XT and Charisma Diamond were significantly lower for A2 than for the other shades. Conclusions: No composite shade matched with the shade guide. Equivalent shades of the restorative composite from different manufacturers may show clinically noticeable ΔE.

• Restorative Dentistry and Endodontics
• /
• 제49권1호
• /
• pp.7.1-7.11
• /
• 2024

Objectives: This study aimed to evaluate the impact of substrate color and interface distance on the color adjustment of 2 single-shade composites, Vittra APS Unique and Charisma Diamond One. Materials and Methods: Dual disc-shaped specimens were created using Vittra APS Unique or Charisma Diamond One as the center composite, surrounded by shaded composites (A1 or A3). Color measurements were taken with a spectrophotometer against a gray background, recording the color coordinates in the CIELAB color space. Illumination with a light-correcting device and image acquisition using a polarizing filter-equipped cell phone were performed on specimens over the same background. Image processing software was used to measure the color coordinates in the center and periphery of the inner composite and in the outer composite. The color data were then converted to CIELAB coordinates and adjusted using data from the spectrophotometer. Color differences (ΔE₀₀) between the center/periphery of single-shade and outer composites were calculated, along with color changes in single-shade composites caused by different outer composites. Color differences for the inner composites surrounded by A1 and A3 were also calculated. Data were analyzed using repeated-measures analysis of variance (α = 0.05). Results: The results showed that color discrepancies were lowest near the interface and when the outer composite was whiter (A1). Additionally, Charisma Diamond One exhibited better color adjustment ability than Vittra APS Unique. Conclusions: Color discrepancies between the investigated single-shade composites diminished towards the interface with the surrounding composite, particularly when the latter exhibited a lighter shade.

• 한국컴퓨터그래픽스학회논문지
• /
• 제16권1호
• /
• pp.9-20
• /
• 2010

3차원 공간에 존재하는 반투명 물질을 물리 기반으로 렌더링하기 위해서 빛의 진행 경로를 작은 구간으로 나눈 후, 각 구간에 대하여 빛의 직접적인 영향, 산란으로 인한 영향, 반투명 물질안에서의 소멸 및 물질의 발광으로 인한 영향 등을 고려한 빛의 에너지를 계산하여 누적하는 방식을 사용한다. 이중 빛의 산란은 연속 공간에서 매우 복잡한 방식으로 작용하기 때문에 이의 시뮬레이션을 위해서 상당한 노력이 필요하다. 빛의 산란을 효과적으로 계산하기 위해 여러근사 방법들이 고안되었는데, 그중 볼륨 포톤매핑 기법은 빛이 간섭매체 내부에서 산란되는 효과를 단순화된 시뮬레이션으로 미리 계산하여두고 이를 검색해 효율적으로 렌더링 하는 방법을 사용한다. 이 방법은 주변에서 검색한 시뮬레이션 정보를 이용하기 위해서 밀도추정방법을 적용하는데, 시뮬레이션자료의분포에 따라서 검색에 따른 편차가 있을 수 있게된다. 또한 시뮬레이션된 자료만을 이용하여 산란효과를 반영하기 때문에 고밀도이지만 시뮬레이션이 충분하지 못한 위치에 대해서 방향성 있는 위상함수에 대한 특징을 잘 표현하지 못한다는 문제가있다. 이러한 문제를해결하고자 하는 노력의 일환으로, 본 논문에서는 입자형태로 시뮬레이션된 볼륨데이터에 대해 밀도추정방법의 하나인 커널스무딩을 이용하여 표현한 산란효과를 반투명물질 자료구조에 저장하고 이를 복원하는 방법을 제안하고, 실험결과 분석을 통하여 장단점을 분석한다.

• 전자공학회논문지CI
• /
• 제45권2호
• /
• pp.1-9
• /
• 2008

휴대폰에서 보안 필요성이 증가함에 따라 개인 인증을 위하여 홍채, 지문, 얼굴과 같은 단일 생체 정보를 이용한 많은 연구들이 진행되었으나 단일 생체 인식에서는 인식 정확도에 한계가 있었다. 따라서 본 논문에서는 휴대폰 환경에서 고 인식율을 위해 얼굴과 홍채를 결합하는 방법에 대해 제안한다. 본 논문에서는 근적외선 조명과 근적외선 통과 필터를 부착한 휴대폰의 메가 픽셀 카메라를 사용하여 근적외선 얼굴 및 홍채 영상을 동시에 취득한 후, SVM(Support Vector Machine)을 기반으로 스코어 레벨에서 결합하였다. 또한, 저 연산의 로가리듬(Logarithm) 알고리즘을 사용한 얼굴 데이터의 조명 변화에 대한 정규화와 극 좌표계 변환 및 홍채 코드의 비트 이동 매칭에 의한 홍채 영역의 이동, 회전, 확대 및 축소에 대한 정규화를 통해 SVM의 분류 복잡도와 얼굴, 홍채 데이터의 본인 변화도를 최소화함으로써 인식 정확도를 향상시켰으며, 저 연산의 휴대폰 환경에서 정수혈 기반의 얼굴 및 홍채 인식 알고리즘을 사용하여 처리시간을 향상시켰다. 실험 결과, SVM을 사용한 인식의 정확성이 단일 생체(얼굴 또는 홍채), SUM, MAX, MIN 그리고 Weighted SUM을 사용하는 것보다 우수한 것을 알 수 있었다.