• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.1795-1811
• /
• 2019

In order to achieve rapid and accurate detection of vehicle objects in complex traffic conditions, we propose a novel vehicle detection method. Firstly, more contextual and small-object vehicle information can be obtained by our Joint Feature Network (JFN). Secondly, our Evolved Region Proposal Network (EPRN) generates initial anchor boxes by adding an improved version of the region proposal network in this network, and at the same time filters out a large number of false vehicle boxes by soft-Non Maximum Suppression (NMS). Then, our Mask Network (MaskN) generates an example that includes the vehicle occlusion, the generator and discriminator can learn from each other in order to further improve the vehicle object detection capability. Finally, these candidate vehicle detection boxes are optimized to obtain the final vehicle detection boxes by the Fine-Tuning Network(FTN). Through the evaluation experiment on the DETRAC benchmark dataset, we find that in terms of mAP, our method exceeds Faster-RCNN by 11.15%, YOLO by 11.88%, and EB by 1.64%. Besides, our algorithm also has achieved top2 comaring with MS-CNN, YOLO-v3, RefineNet, RetinaNet, Faster-rcnn, DSSD and YOLO-v2 of vehicle category in KITTI dataset.

• 전자공학회논문지A
• /
• 제32A권12호
• /
• pp.115-121
• /
• 1995

A mask for x-ray lithography is fabricated with SiN membrane and WTi absorber. SiN membrane is deposited by plasma enhanced chemical vapor deposition, and the stress of SiN membrane is controlled to be less than 100 MPa by rapid thermal annealing. WTi absorber is reactively deposited by DC-magnetron type sputter, and the working gases are argon and nitrogen. Added nitrogen is contributed to the stress of WTi absorber. The stress of WTi absorber is controlled to be less than $\pm$ 100 MPa by controlling the deposition pressure. 10$\mu$m WTi pattern is delineated on SiN membrane by dry etching technique.

• 반도체디스플레이기술학회지
• /
• 제18권3호
• /
• pp.1-6
• /
• 2019

Foreground/background segmentation in video sequences is often one of the first tasks in machine vision applications, making it a critical part of the system. In this paper, we present an improved sample-based technique that provides robust background image as well as segmentation mask. The conventional multiple interval sampling (MIS) algorithm have suffer from the unbalance of computation time per frame and the rapid change of confidence factor of background pixel. To balance the computation amount, a random-based pixel update scheme is proposed and a spatial and temporal smoothing technique is adopted to increase reliability of the confidence factor. The proposed method allows the sampling queue to have more dispersed data in time and space, and provides more continuous and reliable confidence factor. Experimental results revealed that our method works well to estimate stable background image and the foreground mask.

• 대한치과교정학회지
• /
• 제53권1호
• /
• pp.54-64
• /
• 2023

This case report describes skeletal anchorage-supported maxillary protraction performed with the Alternate Rapid Maxillary Expansion and Constriction (AltRAMEC) protocol over a treatment duration of 14 months in a 16-year-old female patient who was in the late growth-development period. Miniplates were applied to the patient's aperture piriformis area to apply force from the protraction appliance. After 9 weeks of following the Alt-RAMEC protocol, miniplates were used to transfer a unilateral 500-g protraction force to a Petit-type face mask. A significant improvement was observed in the soft tissue profile in measurements made both cephalometrically and in three dimensional photographs. Subsequently, the second phase of fixed orthodontic treatment was started and the treatment was completed with the retention phase. Following treatment completion, occlusion, smile esthetics, and soft tissue profile improved significantly in response to orthopedic and orthodontic treatment.

• 한국컴퓨터정보학회논문지
• /
• 제15권6호
• /
• pp.19-27
• /
• 2010

JPEG2000 이미지에서 사용자의 관심영역에 따라 동적으로 이미지의 일부를 우선적으로 처리하는 것이 관심 영역 처리기법이다. 작은 이미지는 큰 의미가 없지만, 큰 이미지에서는 화면에 출력되는 속도가 느리기 때문에 사용자가 먼저 보고자하는 영역을 지정함으로써 지정된 부분을 우선처리하게 된다. 사용자는 대략의 이미지 중에서 관심영역을 지정하면 지정된 영역의 범위를 마스킹 하여 이미지를 전송한 곳으로 보내게 된다. 관심영역 마스킹 정보를 얻은 서버는 우선적으로 마스킹 되어 있는 코드 블록을 우선적으로 전송한다. 여기서, 빠르게 마스킹 정보를 생성하는 것이 중요한데, 본 연구에서는 미리 만들어 놓은 48개의 마스킹 패턴을 사용하여, ROI(Region-of-Interest)와 배경의 분포에 따라 마스킹 패턴 중에 하나를 선택함으로써 마스킹 영역을 계산하는 시간을 현저히 줄였다. 이 패턴이 적용되는 블록은 한 블록 내에 ROI 영역과 배경 영역이 섞여 있는 블록이다. 한 블록 전체가 ROI 이거나 배경이면 이 패턴은 사용되지 않는다. 실험한 결과, ROI와 배경을 정확하게 분리하여 처리하는 방법에 비하여 약간의 품질은 떨어지지만, 처리시간은 현저히 줄었음을 보였다.

• 대한원격탐사학회지
• /
• 제39권5_1호
• /
• pp.637-654
• /
• 2023

최근 지구관측 위성이 급격히 발전함에 따라 사용자의 수가 증가하고 있다. 이에 따라 지구관측위성위원회(Committee on Earth Observation Satellites, CEOS)에서는 분석준비자료(Analysis Ready Data, ARD)라는 개념을 제안하고 분석준비자료의 요구 조건을 CEOS ARD for Land (CARD4L)로 정의하여 사용자 친화적인 위성영상을 제공하기 위해 노력하고 있다. 분석준비자료에는 육상분석에 불필요한 픽셀이 식별된 마스크(Unusable Data Mask, UDM)가 영상과 함께 제공되어야 한다. UDM의 종류는 구름, 구름 그림자, 지형그림자 등이 있다. 지형그림자는 지형기복이 큰 산악지형에서 발생되며 지형그림자가 생긴 지역은 복사조도가 낮기 때문에 분석 결과에 오류를 야기시킨다. 기존 지형그림자 탐지연구는 지형그림자 보정을 위해 지형그림자 픽셀을 탐지하는데 목적을 두었지만, 이것은 지형보정 기법으로 대체 가능하다. 따라서 지형그림자 탐지 목적을 확장할 필요가 있다. 산림과 농업분석을 목적으로 한 차세대중형위성 4호(CAS500-4)의 활용을 위해 본 연구에서는 지형그림자 탐지 범위를 태양의 영향을 적게 받는 지역까지 확장하였다. 본 논문은 남북한을 대상으로 지형그림자 마스크 생성을 위해 지형그림자 탐지 가능성을 분석하는데 목적이 있다. 지형그림자 탐지를 위해서 태양의 위치, 지표면의 경사와 경사방향을 이용한 음영기복 알고리즘을 사용하였다. 한반도를 촬영한 5 m급 공간해상도의 RapidEye 영상과 10 m급 공간해상도의 Sentinel-2 영상들을 대상으로 참값과 비교하며 최적의 음영기복 임계값을 결정하였다. 결정된 임계값을 사용하여 지형 그림자 탐지를 수행하고 결과를 분석하였다. 정성적 결과로는 전체적으로 참값과의 형상이 유사함을 확인하였다. 정량적 실험결과는 F1 score가 대부분 0.8에서 0.94 사이인 것을 확인하였다. 본 연구 결과를 바탕으로 남북한을 대상으로 자동적인 지형그림자 탐지가 잘 수행됨을 확인하였다.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2002년도 International Meeting on Information Display
• /
• pp.1053-1055
• /
• 2002

Recently CRT is getting large-sized, Flatness and High Fine Pitched in the meantime the raw material for shadow mask is in rapid progress of thinness, Low Thermal Expansion and high strength.Until now we have used AK(Aluminum Killed) & Invar(Fe-Ni alloy) materials for main raw material of shadow mask component. However recently Nb and Co addition and Nb+Co addition, which has advantage of Low Thermal Expansion and High Strength. has been developed as well as applying in mass production as CRT's trend has become more flat and fine pitch. Among of them, Co addition has been mass production as forming type (Flat CRT) with the beneficial effect of low thermal expansion & high strength for the first time. Since then Nb+Co addition has been used in mass production by the request of much higher strength of shadow mask component. In case of Nb addition, It's thermal expansion coefficient is a little lower than normal Invar and a little higher than Co addition, meanwhile Its Mechanical property is almost similar to Co Addition. The used samples of this experiment are 36%Ni + Fe, 32%Ni + 5%Co + Fe, 32%Ni + 5%Co + 0.3%Nb + Fe, 32%Ni + 0.3%Nb + Fe with heat treatment temperature of 600$^{\circ}C$, 650$^{\circ}C$, 700$^{\circ}C$, 750$^{\circ}C$, 800$^{\circ}C$, 850$^{\circ}C$, 900$^{\circ}C$ respectively under the condition of 15min holding time. After heat treatment, we have observed the change of mechanical property with addition of small elements through mechanical property investigation and metal structure observation as well as transition of thermal expansion coefficient by measuring of thermal expansion coefficient at 850$^{\circ}C$. In conclusion, 5%Co addition indicates that its thermal expansion coefficient is very similar under the condition of at 850$^{\circ}C$ for 15min 's heat treatment. From the experimental result it is suggested that Co addition is mostly suitable for Doming property and Nb addition is mostly suitable for Drop property.

• 한국초전도저온공학회:학술대회논문집
• /
• 한국초전도저온공학회 2002년도 학술대회 논문집
• /
• pp.131-134
• /
• 2002

In this work, we have developed a systematic way of utilizing the basic design tools for superconductive electronics. This include WRSPICE, XIC, margin program, and L-meter. Since the high performance analog-to- digital converter can be built with Rapid Single Flux Quantum (RSFQ) logic circuits the development of superconductive analog-to-digital converter has attracted a lot of interests as one of the most prospective area of the application of Josephson Junction technology. One of the main advantages in using Rapid Single Flux Quantum logic in the analog-to-digital converter is the low voltage output from the Josephson junction switching, and hence the high resolution. To design an 1-bit analog-digital converter, first we have used XIC tool to compose a circuit schematic, and then studied the operational principle of the circuit with WRSPICE tool. Through this process, we obtained the proper circuit diagram of an 1-bit analog-digital converter circuit. Based on this circuit we performed margin calculations of the designed circuits and optimized circuit parameters. The optimized circuit was laid out as a mask drawing. Inductance values of the circuit layout were calculated with L-meter. Circuit inductors were adjusted according to these calculations and the final layout was obtained.

• 한국초전도ㆍ저온공학회논문지
• /
• 제4권1호
• /
• pp.8-12
• /
• 2002

Since the high performance analog-to-digital converter can be built with Rapid Single Flux Quantum (RSFQ) logic circuits the development of superconductive analog-to-digital converter has attracted a lot of interests as one of the most prospective area of the application of Josephson Junction technology. One of the main advantages in using Rapid Sng1e Flux Quantum logic in the analog-to-digital converter is the low voltage output from the Josephson junction switching, and hence the high resolution. To design an analog-digital converter, first we have used XIC tool to compose a circuit schematic, and then studied the operational principle of the circuit with WRSPICE tool. Through this process, we obtained the proper circuit diagram of an 1-bit analog-digital converter circuit. The optimized circuit was laid out as a mask drawing. Inductance values of the circuit layout were calculated with L-meter.

• 한국레이저가공학회지
• /
• 제9권1호
• /
• pp.17-23
• /
• 2006

The technologies of laser micromachining are changed toward more complex-micropatterning, from the micro circle-shaped hole drilling to the micro arbitrary-shaped hole drilling. In this paper, the fundamental experiments by using DPSS 3rd harmonic $Nd:YVO_4\;laser({\lambda}=355nm)$ were carried out in order to obtain the feasibility of flexible micropatterning by various fixed masks. Fixed masks and Galvano scanners were investigatde to make micro patterns. from these experimental results, micropatterns on PEN film were rapidly manufactured in large area.