• 전자공학회논문지C
• /
• 제36C권1호
• /
• pp.49-58
• /
• 1999

본 논문에서는 통과 대역폭이 5KHZ, ripple이 0.1dB이하인 스위치드 캐패시터(Switched Capacitor) 필터를 $0.8{\mu}m$ single poly CMOS ASIC 표준 공정을 이용하여 집적화된 단일 칩으로 제작하였다. 제안된 5차 타원 저역 통과 필털의 구성은 MOS 스위치와 poly 캐패시터, 5개의 2단 CMOS op-amp로 구성하였다. 필터구현은 LC 수동형으로부터 연속전달함수 H(s)를 구하고 쌍선형 z변환을 통하여 이산전달함수 H(z)으로 바꾸어 2차 바이쿼드(biquad)를 종속 연결하는 빌딩블록 방식을 택하였다. 또한 op-amp의 구동범위를 고려하면서 캐패시터 면적을 감소시킨 스케일링 실시한 동일 특성의 필터를 제작하여 그 특성 변화를 비교, 분석하였다. 측정결과 ${\pm}2.5V$ 전원, 50KHz의 표본 주파수에서 2종의 필터 모두 4.96~4.98KHz의 통과 대역폭에 0.7~0.81dB의 리플, 35~38dB정도의 저지대역 이득감쇠 특성을 보였다.

• 로봇학회논문지
• /
• 제18권3호
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• 마이크로전자및패키징학회지
• /
• 제22권2호
• /
• pp.11-19
• /
• 2015

Since CMOS device scaling has stalled, three-dimensional (3-D) integration allows extending Moore's law to ever high density, higher functionality, higher performance, and more diversed materials and devices to be integrated with lower cost. 3-D integration has many benefits such as increased multi-functionality, increased performance, increased data bandwidth, reduced power, small form factor, reduced packaging volume, because it vertically stacks multiple materials, technologies, and functional components such as processor, memory, sensors, logic, analog, and power ICs into one stacked chip. Anticipated applications start with memory, handheld devices, and high-performance computers and especially extend to multifunctional convengence systems such as cloud networking for internet of things, exascale computing for big data server, electrical vehicle system for future automotive, radioactivity safety system, energy harvesting system and, wireless implantable medical system by flexible heterogeneous integrations involving CMOS, MEMS, sensors and photonic circuits. However, heterogeneous integration of different functional devices has many technical challenges owing to various types of size, thickness, and substrate of different functional devices, because they were fabricated by different technologies. This paper describes new 3-D heterogeneous integration technologies of chip self-assembling stacking and 3-D heterogeneous opto-electronics integration, backside TSV fabrication developed by Tohoku University for multifunctional convergence systems. The paper introduce a high speed sensing, highly parallel processing image sensor system comprising a 3-D stacked image sensor with extremely fast signal sensing and processing speed and a 3-D stacked microprocessor with a self-test and self-repair function for autonomous driving assist fabricated by 3-D heterogeneous integration technologies.

• Journal of Periodontal and Implant Science
• /
• 제43권2호
• /
• pp.101-105
• /
• 2013

Purpose: The present study aimed to measure root surface roughness in teeth with periodontitis by a profilometer following root planning with ultrasonic and hand instruments with and without erbium-doped yttrium aluminium garnet (Er:YAG) laser irradiation. Methods: Sixty single-rooted maxillary and mandibular teeth, extracted because of periodontal disease, were collected. The crowns and apices of the roots were cut off using a diamond bur and water coolant. The specimens were mounted in an acrylic resin block such that a plain root surface was accessible. After primary evaluation and setting a baseline, the samples were divided into 4 groups. In group 1, the samples were root planned using a manual curette. The group 2 samples were prepared with an ultrasonic scaler. In group 3, after scaling with hand instrumentation, the roots were treated with a Smart 1240D plus Er:YAG laser and in group 4, the roots were prepared with ultrasonic scaler and subsequently treated with an Er:YAG laser. Root surface roughness was then measured by a profilometer (MahrSurf M300+RD18C system) under controlled laboratory conditions at a temperature of $25^{\circ}C$ and 41% humidity. The data were analyzed statistically using analysis of variance and a t-test (P<0.05). Results: Significant differences were detected in terms of surface roughness and surface distortion before and after treatment. The average reduction of the surface roughness after treatment in groups 1, 2, 3, and 4 was 1.89, 1.88, 1.40, and 1.52, respectively. These findings revealed no significant differences among the four groups. Conclusions: An Er:YAG laser as an adjunct to traditional scaling and root planning reduces root surface roughness. However, the surface ultrastructure is more irregular than when using conventional methods.

• 방송공학회논문지
• /
• 제16권1호
• /
• pp.33-43
• /
• 2011

3D 디스플레이의 급격한 공급으로 다양한 3D 입체 콘텐츠의 제작이 요구되고 있다. 3D 콘텐츠의 제작 방법으로는 일반적으로 스테레오 카메라로 입체물을 제작할 수 있지만, 기존 2D 콘텐츠에서 입체영상물을 제작하는 기법도 중요한 기술이다. 이러한 2D 콘텐츠의 변환작업을 통한 3D 영상물 제작은 3D 변환 분야에서 활발히 진행되고 있다. 그러나 단순히 한 장의 영상을 3D로 변환하는 것은 콘텐츠의 실감을 전달하지 못한다. 본 논문에서는 기존의 2D 영상에 존재하는 객체에 생명력을 불어넣어, 그림속이 객체가 살아나 움직이는 새로운 애니메이션 콘텐츠 제작 방법을 제안한다. 또한 3D 디스플레이에서 시청이 가능하도록 입체영상으로 제작된다. 입력영상이 주어지면 배경영상, 전경객체 마스크, 배경 깊이맵, 객체 깊이맵을 생성한다. 각 전경객체의 이동, 회전, 및 줌을 통해 생동감있는 객체를 구현하며, 변환된 객체의 깊이맵 및 배경영상을 이용하여 실감있는 입체영상을 제작한다. 실험영상은 신윤복의 단오풍정 및 픽사의 애니메이션 UP의 클립영상을 이용하여 3D 입체영상으로 구현하였다.

• International Journal of Industrial Entomology and Biomaterials
• /
• 제8권1호
• /
• pp.113-115
• /
• 2004

Six generations (P$_1$, P$_2$, F$_1$, BC$_1$, BC$_2$ and F$_2$) of four bivoltine silkworm crosses (SKAU-R-1 ${\times}$ Yakwei, SKAU-R-6 ${\times}$ Yakwei, CSR$_2$ ${\times}$ CSR$_4$ and SH$_{6}$${\times}$ NB$_4$D$_2$) were evaluated in a completely randomized block design with 5 replications for each treatment. The generation mean 2 in respect of 3 metric traits (single cocoon weight, single shell weight, and shell ratio ％), were subjected to Cavallis joint scaling test. Additive dominance model was found to be adequate in CSR$_2$${\times}$CSR$_4$ and SH$_{6}$${\times}$ NB$_4$D$_2$ for single cocoon weight and shell weight and SKAU-R-6 ${\times}$ Yakwei for shell ratio(％). Whereas, in rest of the crosses epistasis was evident in the traits under investigation. Magnitude of additive gene effect (d) was greater than dominance(h) in SH$_{6}$${\times}$NB$_4$D$_2$ and SKUA-R-6${\times}$Yakwei for shell ratio (％) and in CSR$_2$${\times}$ CSR$_4$ for shell weight. Thus selection for these traits in early segregating generations of these crosses would be effective for obtaining considerable genetic gain. gain.

• Agricultural and Biosystems Engineering
• /
• 제1권2호
• /
• pp.95-99
• /
• 2000

Nondestructive methods such as ultrasonic and magnetic resonance imaging systems have many advantages but still much expensive. And they do not give exact color information and may miss some details. If it is allowed to destruct some biological objects to get interior and exterior informations, constructing 3D image form a series of slices sectional images gives more useful information with relatively low cost. In this paper, a PC based automatic 3D model generator was developed. The system was composed of three modules. The first module was the object handling and image acquisition module, which fed and sliced the object sequentially and maintains the paraffine cool to be in solid state and captures the sectional image consecutively. The second one was the system control and interface module, which controls actuators for feeding, slicing, and image capturing. And the last was the image processing and visualization module, which processed a series of acquired sectional images and generated 3D volumetric model. Handling module was composed of the gripper, which grasped and fed the object and the cutting device, which cuts the object by moving cutting edge forward and backward. sliced sectional images were acquired and saved in a form of bitmap file. 2D sectional image files were segmented from the background paraffine and utilized to generate the 3D model. Once 3-D model was constructed on the computer, user could manipulated it with various transformation methods such as translation, rotation, scaling including arbitrary sectional view.

• 한국통신학회논문지
• /
• 제30권10C호
• /
• pp.1007-1016
• /
• 2005

본 논문은 3차원 모델의 모양 기반 검색을 하기 위한 모델의 특징을 추출하는 방법을 제시한다. 3차원 모델의 특징 기술자는 모델에 대한 위치, 회전, 크기 변환에서 그 특징이 불변해야 하기 때문에, 모델을 정규 좌표계로 표시하기 위한 선(先)처리 작업이 필요하다. 우리는 선처리 작업을 위해서 주성분 분석 방법을 사용하였으며, 이 방법은 최소 경계 상자와 외접구의 생성을 위해서도 이용되었다. 제안한 알고리즘은 다음과 같다. 반지름 1인 외접구를 만들고, 구의 중심에 3차원 모델을 위치시킨 후, 반지름이 다른 동심구($r_i=i/n,\;i=1,2,{\ldots},n$)를 생성하고, 이 동심구들과 모델이 접하는 면을 구한 다음 그 면에 대한 곡률을 계산한다. 여기서 구한 곡률을 3차원 모델의 특징 기술자로 사용하게 된다. 실험 결과는 타 알고리즘에 비해 제안하는 방법이 상대적으로 적은 빈(bin) 수를 가졌음에도 불구하고 ANMRR 평가 함수에 의해 최소 0.1에서 0.6 이상의 성능 개선 효과가 나타나고 있음을 보여 준다. 본 논문은 색인 기법으로 $R{^*}-tree$를 사용하였다.

• 화약ㆍ발파
• /
• 제16권3호
• /
• pp.25-34
• /
• 1998

This area is covered in Andesite of high compression strength and located in PUSAN SEO-KU. There are many old houses around shaft site. So, we must have a cautious blasting operation. A total of 40 blasts were test at DAE-SHIN Shaft site to study the magnitude and frequency characterization of blast-induced vibration. The effect of viblating frequency on structual damage and site-specific scaling to define th empirical equations were also discussed. The result can be summarized as follows: 1. Some empirical equations were obtained. $V=K\{{\frac{D}{W}}1/3\}^{-n}$ where the values for n and K are estimated to be -1.407 to -2.202 and 643.3489 to 7283.2104 respectively. 2. Dominant frequencies at short distance are in the range of about 75.0 to 91.8 Hz, with some exceptions of about 50Hz, Frequencies observed at long distance are in the range of 10 to 2Hz. It is apparent the shift of dominant frequency down to lower levels at long distance.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제9권2호
• /
• pp.110-116
• /
• 2009

The prominent advantages of Dual Material Surrounding Gate (DMSG) MOSFETs are higher speed, higher current drive, lower power consumption, enhanced short channel immunity and increased packing density, thus promising new opportunities for scaling and advanced design. In this Paper, we present Transconductance-to-drain current ratio and electric field distribution model for dual material surrounding gate (DMSGTs) MOSFETs. Transconductance-to-drain current ratio is a better criterion to access the performance of a device than the transconductance. This proposed model offers the basic designing guidance for dual material surrounding gate MOSFETs.