• 전자공학회논문지S
• /
• 제36S권2호
• /
• pp.81-87
• /
• 1999

비월주사방식의 TV 방송은 전송대역폭을 효율적으로 사용할 수 있기 때문에 현재 방송표준에서 사용하고 있으며, MPEG-2와 같은 압축표준에서도 이를 지원하도록 되어있다. 본 논문에서는 이러한 비월주사방식을 사용하는 동영상의 화질개선을 위한 후처리기법으로 필드불일치 보정기법과 움직임열화를 제거하는 기법을 제안한다. 필드불일치 보정기법은 에지 분류를 기반으로하는 선형보간기법과 움직임을 추정하여 이를 보상하는 방법에 대해 설명한다. 선형보간을 위한 에지의 분류는 압축표준에서 가장 널리 사용되는 DCT의 계수를 이용함으로써 추가적인 계산과정 없이 단순히 계수의 비교만으로 이루어진다. 또한 움직임을 추정하여 이를 보상하는 방법은 움직임을 기반으로 하는 영상분할 기법을 적용하여 두 필드이 정보를 모두 이용하게 되므로 정보의 이용이라는 측면에서 매우 효율적이다. 동영상의 화질개선을 위한 움직임열화 제거기법은 제안되는 열화모델을 기반으로 공간적응적인 반복적 영상복원기법을 이용하여 화질개선 효과를 얻을 수 있다.

• 한국CDE학회논문집
• /
• 제17권3호
• /
• pp.156-163
• /
• 2012

Real-time three-dimensional shape measurement is becoming increasingly important in various fields, including medical sciences, high-technology industry, and microscale measurements. However, there are not so many 3D profile tools specially designed for specifically narrow space, for example, to scan the tooth shape of a human jaw. In this paper, a real-time 3D intraoral scanner is proposed for the measurement of tooth profile in the mouth cavity. The proposed system comprises a laser diode beam, a micro charge-coupled device, a graticule, a piezoelectric transducer, a set of optical lenses, and a polhemus device sensor. The phase-shifting technique is used along with an accurate calibration method for the measurement of the tooth profile. Experimental and theoretical inspection of the phase-to-coordinate relation is presented. In addition, a nonlinear system model is developed for collimating illumination that gives the more accurate mathematical representation of the system, thus improves the shape measurement accuracy. Experiment results are presented to verify the feasibility and performance of the developed system. The experimental results indicate that overall measurement error accuracy can be controlled within 0.4 mm with a variability of ${\pm}0.01$.

• 대한전자공학회논문지SD
• /
• 제41권10호
• /
• pp.9-16
• /
• 2004

유한상태기의 상태할당은 이로부터 구현되는 순차회로의 속도, 면적, 테스팅 및 소비전력에 큰 영향을 미친다. 본 논문에서는 상태변수 그룹들 사이에 상호 의존성(dependency)을 최소화하여 테스팅 및 전력소모를 개선하기 위한 m-블록 분할을 이용한 새로운 상태할당 기술을 소개한다. 제안된 알고리듬은 상태들을 그룹으로 나누어 상태변수의 상호의존성을 줄이고, 상태천이 확률에 의해 결정된 무게인자에 따라 상태 간 상태변수의 변화를 최소로하는 코드를 할당하여 상태 천이시 스위칭 횟수를 줄인다. 즉 피드백 순환의 길이와 수는 상태 변수들 간에 최소 전환 활동으로 감소됩니다 벤치마크 회로에 대한 실험결과는 테스팅 및 소비전력이 현저히 개선되었음을 확인하였다.

• 반도체디스플레이기술학회지
• /
• 제11권2호
• /
• pp.45-51
• /
• 2012

Multilayer co-extrusion blown film construction is a popular technique for producing plastic films for various packaging industries. Automated detection of defective films can improve the quality of film production process. In this paper, we propose a film inspection system that can detect and classify film defects robustly. In our system, first, film images are acquired through a high speed line-scan camera under an appropriate lighting system. In order to detect and classify film defects, an inspection algorithm is developed. The algorithm divides the typical film defects into two groups: intensity-based and texture-based. Intensity-based defects are classified based on geometric features. Whereas, to classify texture-based defects, a texture analysis technique based on local binary pattern (LBP) is adopted. Experimental results revealed that our film inspection system is effective in detecting and classifying defects for the multilayer co-extrusion blown film construction line.

• 반도체디스플레이기술학회지
• /
• 제7권1호
• /
• pp.11-16
• /
• 2008

Confocal scanning microscopy is a widely used technique for three dimensional measurements because it is characterized by high resolution, high SNR and depth discrimination. Generally an image is generated by moving one optical probe that satisfies the confocal condition on the specimen. Measurement speed is limited by movement speed of the optical probe; scanning speed. To improve measurement speed we increase the number of optical probes. Specimen region to scan is divided by optical probes. Multi-point information each optical probe points to can be obtained simultaneously. Therefore image acquisition speed is increased in proportion to the number of optical probes. And multiple optical probes from red, green and blue laser sources can be used for color imaging and image quality, i.e., contrast, is improved by adding color information by this way. To conclude, this technique contributes to the improvement of measurement speed and image quality.

• 대한원격탐사학회지
• /
• 제9권1호
• /
• pp.1-19
• /
• 1993

A simple bispectral threshold technique which reflects the temporal and spatial characteristics of the analysis area has been developed to classify the cloud type and estimate the cloud cover from GMS/S-VISSR(Stretched Visible and Infrared Spin Scan Radiometer) imagery. In this research, we divided the analysis area into land and sea to consider their different optical properties and used the same time observation data to exclude the solar zenith angle effects included in the raw data. Statistical clear sky radiance(CSRs) was constructed using maximum brightness temperature and minimum albedo from the S-VISSR imagery data during consecutive two weeks. The CSR used in the cloud anaysis was updated on the daily basis by using CSRs, the standard deviation of CSRs and present raw data to reflect the daily variation of temperature. Thresholds were applied to classify the cloud type and estimate the cloud cover from GMS/S-VISST imagery. We used a different thresholds according to the earth surface type and the thresholds were enough to resolve the spatial variation of brightness temperature and the noise in raw data. To classify the ambiguous pixels, we used the time series of 2-D histogram and local standard deviation, and the results showed a little improvements. Visual comparisons among the present research results, KMA's manual analysis and observed sea level charts showed a good agreement in quality.

• 전기전자학회논문지
• /
• 제23권1호
• /
• pp.44-52
• /
• 2019

본 논문은 라이다 센싱 정보와 RGB-D 카메라 영상 정보를 융합하여 이용하는 교육용 실내 자율주행 로봇 시스템을 구현한다. 이 시스템은 라이다 센싱 정보를 획득하기 위해 기존의 소 채널 라이다 센싱 방식을 이용한다. 또한 소 채널 라이다 센싱 방식의 약점을 보완하기 위해, RGB-D 카메라 깊이 영상과 딥러닝 기반 객체인식 알고리즘을 이용하는 3차원 구조물 인식 방법을 제안하고 이 시스템에 적용한다.

• 핵의학기술
• /
• 제28권1호
• /
• pp.35-40
• /
• 2024

Purpose: To improve the image quality in positron emission tomography (PET), the attenuation correction technique based on the computed tomography (CT) data is important process. However, the artifact is caused by metal material during PET/CT scan, and the image quality is degraded. Therefore, the purpose of this study was to evaluate image quality according to with and without iterative metal artifact reduction (iMAR) algorithm using customized 3D printing phantom. Materials and Methods: The Hoffman and Derenzo phantoms were designed. To protect the gamma ray transmission and express the metal portion, lead substance was located to the surface. The SiPM based PET/CT was used for acquisition of PET images according to application with and without iMAR algorithm. The quantitative methods were used by signal to noise ratio (SNR), coefficient of variation (COV), and contrast to noise ratio (CNR). Results and Discussion: The results shows that the image quality applying iMAR algorithm was higher 1.15, 1.19, and 1.11 times than image quality without iMAR algorithm for SNR, COV, and CNR. Conclusion: In conclusion, the iMAR algorithm was useful for improvement of image quality by reducing the metal artifact lesion.

• Journal of Korean Neurosurgical Society
• /
• 제56권2호
• /
• pp.114-120
• /
• 2014

Objective : At present, gold-standard technique of cervical cord decompression is surgical decompression and fusion. But, many complications related cervical fusion have been reported. We adopted an extended anterior cervical foraminotomy (EACF) technique to decompress the anterolateral portion of cervical cord and report clinical results and effectiveness of this procedure. Methods : Fifty-three patients were operated consecutively using EACF from 2008 to 2013. All of them were operated by a single surgeon via the unilateral approach. Twenty-two patients who exhibited radicular and/or myelopathic symptoms were enrolled in this study. All of them showed cervical cord compression in their preoperative magnetic resonance scan images. Results : In surgical outcomes, 14 patients (64%) were classified as excellent and six (27%), as good. The mean difference of cervical cord anterior-posterior diameter after surgery was 0.92 mm (p<0.01) and transverse area was $9.77mm^2$ (p<0.01). The dynamic radiological study showed that the average post-operative translation (retrolisthesis) was 0.36 mm and the disc height loss at the operated level was 0.81 mm. The change in the Cobb angle decreased to 3.46, and showed slight kyphosis. The average vertebral body resection rate was 11.47%. No procedure-related complications occurred. Only one patient who had two-level decompression needed anterior fusion at one level as a secondary surgery due to postoperative instability. Conclusions : Cervical cord decompression was successfully performed using EACF technique. This procedure will be an alternative surgical option for treating cord compressing lesions. Long-term follow-up and a further study in larger series will be needed.

• 한국전기전자재료학회논문지
• /
• 제17권8호
• /
• pp.875-881
• /
• 2004

Fundamental results obtained from an atomic force microscope (AFM) chemically-induced direct nano-lithography process are presented, which is regarded as a simple method for fabrication nm-scale devices such as superconducting flux flow transistors (SFFTs) and single electron tunneling transistors (SETs). Si cantilevers with Pt coating and with 30 nm thick TiO coating were used as conducting AFM tips in this study. We observed the surfaces of superconducting strip lines modified by AFM anodization' process. First, superconducting strip lines with scan size 2 ${\mu}{\textrm}{m}$${\times}$2 ${\mu}{\textrm}{m}$ have been anodized by AFM technology. The surface roughness was increased with the number of AFM scanning, The roughness variation was higher in case of the AFM tip with a positive voltage than with a negative voltage in respect of the strip surface. Second, we have patterned nm-scale oxide lines on ${YBa}-2{Cu}_3{O}_{7-x}$ superconducting microstrip surfaces by AFM conductive cantilever with a negative bias voltage. The ${YBa}-2{Cu}_3{O}_{7-x}$ oxide lines could be patterned by anodization technique. This research showed that the critical characteristics of superconducting thin films were be controlled by AFM anodization process technique. The AFM technique was expected to be used as a promising anodization technique for fabrication of an SFFT with nano-channel.