• 전기전자학회논문지
• /
• 제21권4호
• /
• pp.412-415
• /
• 2017

본 논문에서는 고해상도 $360^{\circ}$ 전방위 IP 카메라를 이용한 다중 번호판 인식 시스템을 제안한다. 제안한 시스템은 $360^{\circ}$ 원형영상의 평면 분할 부와 다중 번호판 인식 부로 구성되었다. $360^{\circ}$ 원형영상의 평면 분할 부는 고해상도 $360^{\circ}$ 전방위 IP 카메라에서 원형영상 획득, 원형영상 분할, 평면영상으로 변환, 보간법을 사용한 픽셀 보정 및 컬러보정, 에지 보정 등의 과정을 거쳐 화질이 개선된 평면영상으로 출력한다. 다중 번호판 인식 부는 평면영상에서 다중 번호판 후보영역 추출, 다중 번호판 후보영역 정규화 및 복원, 신경망을 사용한 다중 번호판 숫자, 문자 인식 과정을 거쳐 다중 번호판을 인식하게 된다. 제안된 고해상도 $360^{\circ}$ 전방위 IP 카메라를 이용한 다중 번호판 인식 시스템을 평가하기 위하여 지능형 주차관제시스템 운영 전문 업체와 공동으로 실험한 결과, 97.8%의 높은 번호판 인식률이 확인되었다.

• Journal of Korean Neurosurgical Society
• /
• 제62권4호
• /
• pp.414-421
• /
• 2019

Objective : To evaluate 3-dimensional magnetic resonance imaging (MRI) of Kambin's safe zone to calculate maximum cannula diameter permissible for safe percutaneous endoscopic lumbar discectomy. Methods : Fifty 3D MRIs of 19 males and 31 females (mean, 47 years) were analysed. Oblique, axial and sagittal views were used for image analysis. Three authors calculated the inscribed circle (cannula diameter) individually, within the neural (original) and bony Kambin's triangle in oblique views, disc heights on sagittal views and root to facet distances at upper and lower end plate levels on axial views and their averages were taken. Results : The mean root to facet distances at upper end plate level measured on axial sections increased from $3.42{\pm}3.01mm$ at L12 level to $4.57{\pm}2.49mm$ at L5S1 level. The mean root to facet distances at lower end plate level measured on axial sections also increased from $6.07{\pm}1.13mm$ at L12 level to $12.9{\pm}2.83mm$ at L5S1 level. Mean maximum cannula diameter permissible through the neural Kambin's triangle increased from $5.67{\pm}1.38mm$ at L12 level to $9.7{\pm}3.82mm$ at L5S1 level. The mean maximum cannula diameter permissible through the bony Kambin's triangle also increased from $4.03{\pm}1.08mm$ at L12 level to $6.11{\pm}1mm$ at L5S1 level. Only 2% of the 427 bony Kambin's triangles could accommodate a cannula diameter of 8mm. The base of the bony Kambin's triangle taken in oblique view (3D MRI) was significantly higher than the root to facet distance at lower end plate level taken in axial view. Conclusion : The largest mean diameter of endoscopic cannula passable through "bony" Kambin's triangle was distinctively smaller than the largest mean diameter of endoscopic cannula passable through "neural" Kambin's triangle at all levels. Although proximity of exiting root to the facet joint is always taken into consideration before PELD procedure, our 3D MRI based anatomical study is the first to provide actual maximum cannula dimensions permissible in this region.

• 디지털융복합연구
• /
• 제11권8호
• /
• pp.273-279
• /
• 2013

본 연구에서는 변조전달함수(Modulation transfer function, MTF)를 이용한 컴퓨터 방사선영상의 감도 노출 분류에 따른 공간분해능을 평가하여 컴퓨터 방사선영상 획득의 기초자료로 제시하고자 하였다. 본 실험에서는 $100{\mu}mm$ 픽셀의 영상판을 이용하여 엣지법의 MTF를 측정하였다. 방사선 선질은 IEC 61267에서 권고하고 있는 RQA5를 이용하였다. X-선이 조사된 영상판은 감도 노출 분류를 50, 100, 200, 300, 400, 600, 800, 1200으로 각각 설정한 후 디지털 영상화하였다. 최종 획득된 영상들은 공인된 영상분석 프로그램인 image J와 Origin 8.0을 이용하여 MTF 50%와 10%를 구하여 평가하였다. 그 결과 감도 노출 분류 200에서 가장 우수한 MTF 50%($1.979{\pm}0.114lp/mm$)와 MTF 10%($3.932{\pm}0.041$)의 공간주파수를 구하였다. 따라서 골절 등과 같이 높은 공간분해능을 요구하는 질병진단에 유용할 것으로 사료된다.

• 대한조선학회논문집
• /
• 제47권3호
• /
• pp.342-349
• /
• 2010

This paper is a foundation work in developing a software for generating infrared images from a scene with various objects. The spectral radiance received by a remote sensor is consisted of the self-emitted, reflected and scattered components. In general, the self-emitted component is the most important part for generating Infrared signatures from the object. In this paper, the infrared image generation considering various surface temperature and optical surface property of a flat plate is demonstrated in MWIR($3{\sim}5{\mu}m$) and LWIR($8{\sim}12{\mu}m$) regions for different spatial resolutions of the images. Resulting spectral radiance values in the MWIR($3{\sim}5{\mu}m$) and LWIR($8{\sim}12{\mu}m$) regions arrived at the infrared sensor are compared numerically and graphically by recognizing that they are strongly dependent on the surface conditions such as the surface temperature and the surface emissivity. And these infrared images are also shown to be strongly dependent on the resolutions of the infrared imaging devices as well. This study reveals that the surface conditions are more dependent on the radiance level from the surface while the resolution of the imaging device is more responsible for identifying the shape of object.

• Journal of Korean Neurosurgical Society
• /
• 제41권5호
• /
• pp.295-300
• /
• 2007

Objective : Unilateral facet dislocation of the cervical spine occurs by flexion and rotation injuries and cannot be easily reduced by axial traction. We analyzed 14 consecutive patients with unilateral facet dislocation of the cervical spine to increase knowledge about anatomical reduction of locked facet and factors for successful reduction. Methods : Fourteen patients [10 men and 4 women] with unilateral facet dislocation of the cervical spine were retrospectively analyzed. Plain X-ray, computerized tomography scan, and magnetic resonance imaging were performed. All patients underwent manual reduction and surgery with anterior interbody fusion and plate fixation. The manual reduction was performed by neck flexion and rotation to the opposite side of dislocation, followed by rotation and flexion of the head toward the side of dislocation and extension with relaxation of traction. Mean follow-up period was 17 months. The level of spine, amount of subluxation, combined facet fracture, and time from injury to initial reduction were analyzed using the data obtained from medical records. Results : Thirteen [93%] patients were reduced successfully. Immediate reduction was achieved in 7 patients but failed in 7 patients. Seven patients underwent delayed closed reduction under general anesthesia, and successful reduction was achieved in 6 patients. Only one patient with bone chips between articular facets failed to achieve anatomical reduction. Conclusion : In order to reduce the locked facet more easily and safely, we recommend manipulative traction with anterior interbody fusion and plate fixation under general anesthesia after being aware of spinal cord injury with magnetic resonance imaging.

• 한국광학회지
• /
• 제5권3호
• /
• pp.349-355
• /
• 1994

Rayleigh의 분해능 한계 $\epsilon_R$보다 작은 선폭 $\epsilon_0$을 분해할 수 있는 광결상계사 이론적으로 취급되었고, 이어서 초분해능을 지니는 광학계가 제시되었다. 먼저 초분해능을 역변환 문제로 취급하여 초분해능 회절상의 진폭 임펄스 함수를 주는 적분방정식을 도입하고 매개변수 $\beta$(양정수)에 따라서 진폭 임펄스 함수의 반치폭이 조사되었다. 이 초분해능 광학계의 구경함수의 진폭 분포가 변조되었고 또 구경에서 나가는 Huygens 파동의 위상도 Holography로 변조되었다. 이들의 변조를 위하여 이용된 광변조 소자가 제시되었다. 초미세 광리소그라피에 유용한 결론을 얻었다.

• 전자공학회논문지
• /
• 제53권10호
• /
• pp.109-115
• /
• 2016

이 논문에서는 레이저를 통한 진동판 스캔과 열화상 카메라를 사용한 진동판 촬영, 두 가지 방법을 비교하며 슬림 스피커의 분할 진동을 검출하는 방법에 대해 살펴본다. 슬림 스피커는 평판형의 구조적인 특성상 분할진동이 두드러지게 나타나고, 설치되는 공간이 좁아 무빙 코일에서 발생하는 열의 냉각이 제한적이다. 이런 특성으로 인해 슬림 스피커에서 분할진동이 제품의 품질에 큰 영향을 미친다. 본 연구에서는 진동판에서 일어나는 분할진동의 영향과 무빙 코일에 의한 진동판의 열전달 관계를 비교 탐색한다. 비교를 위한 실험은 분할진동 모드의 측정과 진동판의 열 변화 측정을 진동판 스캔과 열화상 카메라 촬영의 2단계의 실험으로 진행한다. 동일 주파수에서 발생하는 분할진동 모드와 열전달 형태를 비교하여 서로 간에 어떤 영향을 미치고 있는 지 파악할 수 있다. 그리고 이를 통해 발견한 연관성을 통해 쉽게 촬영할 수 있는 열화상만으로도 슬림 스피커가 가지는 분할진동의 형태와 경향성을 빠르게 파악하여 최적 설계에 도움이 되는 자료로 사용할 수 있을 것이다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제29권4호
• /
• pp.237-240
• /
• 2006

CR 시스템의 잡음(Noise)의 주요한 원인은 X-선 양자 모틀이며 그 외에 Imaging plate의 구조 노이즈, 휘진발광체의 광양자 노이즈, 전기계의 노이즈, 양자화 노이즈 등이 있다. Regius 150시스템에서 선량이 $0.1\;mR{\sim}100\;mR$까지 변화할 때 SNR 변화는 0.1 mR에서 8.2부터 선형성을 가지고 증가하고 있다. 이때 최저 SNR 값이 5 이상 되므로 충분한 신뢰성을 확보했다고 할 수 있다. 의료영상시스템의 가장 일반적인 노이즈 평가방법인 Noise Power Spectrum(NPS)을 이산적 화상데이터로부터 얻어진 Pixel 값의 요동으로부터 직접 푸리에 변환하여 구하였다. 주파수 $0.5\;lp/mm{\sim}2.5\;lp/mm$ 구간의 NPS은 선량이 많아질수록 개선되며 1 mR 조사 시 $10^{-4}{\sim}10^{-5}$ 사이에 분포하여, 시판되고 있는 타 회사의 CR시스템과 유사한 노이즈 특성을 가지고 있다.

• Smart Structures and Systems
• /
• 제8권3호
• /
• pp.303-320
• /
• 2011

Comparing to active damage monitoring, impact localization on composite by using time reversal focusing method has several difficulties. First, the transfer function of the actuator-sensor path is difficult to be obtained because of the limitation that no impact experiment is permitted to perform on the real structure and the difficulty to model it because the performance of real aircraft composite is much more complicated comparing to metal structure. Second, the position of impact is unknown and can not be controlled as the excitation signal used in the active monitoring. This makes it not applicable to compare the difference between the excitation and the focused signal. Another difficulty is that impact signal is frequency broadband, giving rise to the difficulty to process virtual synthesis because of the highly dispersion nature of frequency broadband Lamb wave in plate-like structure. Aiming at developing a practical method for on-line localization of impact on aircraft composite structure which can take advantage of time reversal focusing and does not rely on the transfer function, a PZT sensor array based phase synthesis time reversal impact imaging method is proposed. The complex Shannon wavelet transform is presented to extract the frequency narrow-band signals from the impact responded signals of PZT sensors. A phase synthesis process of the frequency narrow-band signals is implemented to search the time reversal focusing position on the structure which represents the impact position. Evaluation experiments on a carbon fiber composite structure show that the proposed method realizes the impact imaging and localization with an error less than 1.5 cm. Discussion of the influence of velocity errors and measurement noise is also given in detail.

• 대한디지털의료영상학회논문지
• /
• 제10권2호
• /
• pp.23-31
• /
• 2008

Computed radiography(CR) has been widely used in the field of diagnostic radiography since digital X-ray image was introduced. The imaging performance of CR system was studied by analyzing the digital image data of the CR images which are the outcomes of the whole imaging system composed of image plate(IP), laser digitizer, analoge-digital convertor, and a given image processing unit. In this study, we used a conventional CR system made by Agfa. From the flat field image of 150$\times$150 image pixels, signal-to-noise ratio(SNR) was calculated. SNR of the CR image increases in proportion to logarithm value of the X-ray exposure irradiated on the IP. SNR is less than about 6 at the exposure below 0.2mR and is more than 10 at the exposure above 0.54mR. In our study, most of images obtained by the smaller exposures less than 2.0mR can not be readable. In general, the minimum value of the SNR ranges from 3 to 5. We obtained modulation transfer function(MTF) by analyzing the bar pattern image which was made under conditions as follows: X-ray tube potential was 55kVp, the IP exposure was 0.54 mR, and the distance between X-ray source to IP was 2m, where bar pattern was located on the IP. MTF is 23% at 2.5lp/mm spatial frequency. Provided that the MTF of noise equivalent modulation is 10%, the CR system has the limiting spatial resolution of 3.2lp/mm. If the image sharpness is evaluated by the spatial frequency where MTF is 50%. the corresponding spatial frequency is 0.5$\sim$0.75lp/mm. MTFA(Modulation Transfer Function Area) is 1.0lp/mm. Compared with the Fuji CR whose MTFA is 1.1lp/mm, Agfa CR in this study shows almost same MTFA performance.