• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.6
• /
• pp.619-623
• /
• 2008

A lane detection system is one of the major components of intelligent vehicle systems. Difficulties in lane detection mainly come from not only various weather conditions but also a variety of special environment. This paper describes a simple and stable method for the broken lane tracking in various environments. Probabilistic Hough Transform (PHT) and the Least-square method (LSM) are used to track and correct the lane orientation. For the efficiency of the proposed method, two regions of interest (ROIs) are placed in the lower part of each image, where lane marking areas usually appear with less intervention in our system view. By testing in both a set of static images and video sequences, the experiments showed that the proposed approach yielded robust and reliable results.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2004.11a
• /
• pp.248.2-248.2
• /
• 2004

• Transactions on Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.17-20
• /
• 2024

This paper presents a novel lightweight object detection model tailored for low-powered edge devices, addressing the limitations of traditional resource-intensive computer vision models. Our proposed detector, inspired by the Single Shot Detector (SSD), employs a compact yet robust network design. Crucially, it integrates an 'enhancer block' that significantly boosts its efficiency in detecting smaller objects. The model comprises two primary components: the Light_Block for efficient feature extraction using Depth-wise and Pointwise Convolution layers, and the Enhancer_Block for enhanced detection of tiny objects. Trained from scratch on the Udacity Annotated Dataset with image dimensions of 300x480, our model eschews the need for pre-trained classification weights. Weighing only 5.5MB with approximately 0.43M parameters, our detector achieved a mean average precision (mAP) of 27.7% and processed at 140 FPS, outperforming conventional models in both precision and efficiency. This research underscores the potential of lightweight designs in advancing object detection for edge devices without compromising accuracy.

• Journal of KIISE:Software and Applications
• /
• v.36 no.7
• /
• pp.548-559
• /
• 2009

The cascade face detector learned by Adaboost algorithm, which was proposed by Viola and Jones, is state of the art face detector due to its great speed and accuracy. In spite of its great performance, it still suffers from false alarms, and more computation is required to reduce them. In this paper, we want to reduce false alarms with less computation using facial color. Using facial color information, proposed face detection model scans sub-window efficiently and adapts a fast face/non-face classifier at the first stage of cascade face detector. This makes face detection faster and reduces false alarms. For facial color filtering, we define a facial color membership function, and facial color filtering image is obtained using that. An integral image is calculated from facial color filtering image. Using this integral image, its density of subwindow could be obtained very fast. The proposed scanning method skips over sub-windows that do not contain possible faces based on this density. And the face/non-face classifier at the first stage of cascade detector rejects a non-face quickly. By experiment, we show that the proposed face detection model reduces false alarms and is faster than the original cascade face detector.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.2 no.2
• /
• pp.97-104
• /
• 2004

Neutron induced prompt gamma-ray spectroscopy(NIPS) system measures the prompt gamma-ray emitting by the interaction of a neutron with various materials. This system will be of great benefit to scientists worldwide, since it provides the non-destructive measurement of many element in either solid or liquid wastes. In this study, the full-energy-peak (FEP) efficiency calibration for a HPGe detector was constructed in the ${\gamma}$-ray energy range from 80 keV to 8 MeV, using $^{l33}$Ba and >TEX>$^{152}Eu$ RI sources and $ ^{35}Cl(n, ${\gamma}$)^{36}Cl$ thermal neutron captured reaction. The FEP efficiency curve for the higher energies using the $^{35}Cl(n, ${\gamma}$)^{36}Cl$ reaction was normalized with the curve obtained from the RI sources, since the accurate activity of its prompt ${\gamma}$-ray is unknown. The average thermal neutron flux was theoretically calculated using the FEP efficiency curve for the KCl standard solutions. The NIPS system equipped with a ${\gamma}$-${\gamma}$ coincidence setup with two n-type coaxial HPGe detectors was considered in order to reduce the interfering ${\gamma}$-ray background. The FEP efficiency curve for the ${\gamma}$-${\gamma}$ coincidence system was also obtained for full energy range. The performance of the normal and coincidence NIPS system was tested by comparing signal-to-noise ratio in each mode using the reference sample.e.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.10
• /
• pp.2335-2340
• /
• 2012

The detector is an uncooled detector system that functions inside the thermoelectric cooler (TEC) equipped with features instead of the cooler. The function of the thermoelectric device to control the temperature of the detector based on a function of temperature to prevent degradation of image quality to perform the role, the latest technology trend by removing the thermoelectric device size, cost a lot of effort to reduce has been studied. In this paper, It would be proposed of the actual test result using real chamber environment of for the best TECless algorithm as to minimize the degradation of image quality and obtain the low price of the uncooled detector.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.9
• /
• pp.39-45
• /
• 2004

A design method is proposed for the sequence detection with fixed decision delay with less hardware complexity using the concept of the Voronoi diagram and its dual, the Delaunay tessellation. This detector design is based on the Fixed Delay Tree Search (FDTS) detection. The FDTS is a computationally efficient sequence detection algerian and has been shown to achieve near-optimal performance in the severe Intersymbol Interference (ISI) channels when combined with decision feedback equalization and the appropriate channel coding. In this approach, utilizing the information contained in the Voronoi diagram or equivalently the Delaunay tessellation, the relative location of the detector input sequence in the multi-dimensional Euclidean space is found without any computational redundancy, which leads to a reduced complexity implementation of the detector.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.02a
• /
• pp.432-432
• /
• 2013

현대에 이르러 직접방식 엑스선 검출기에서는 기존의 a-Se을 주로 이용하였지만, 고전압 인가에 따른 회로 손상과 짧은 수명, 그리고 누설전류에 따른 안전의 문제 등으로 낮은 에너지 밴드갭과 높은 흡수효율, 비저항 등에 의거한 다양한 대체 물질에 대한 연구가 활발하게 이루어져가고 있다. 본 논문에서는 직접방식 엑스선 검출물질로 전기이동도와 흡수효율이 뛰어나고, 밴드갭이 낮아 태양전지분야 뿐만 아니라 최근 엑스선 검출물질로 각광받고 있는 CdTe를 선정하였다. 연구의 목적은 PVD (Physical Vapor Deposition)방식의 CdTe 검출 물질의 제작과정에서 CdTe가 기화되어 하부전극 기판에 증착될 시, 하부전극 기판 온도에 따른 CdTe의 박막형성과 전기적 측정을 실시하여 그에 따른 최적의 증착조건을 선정하는 것이다. 하부전극 기판으로는 Au/glass를 사용하였으며 증착 시, $200^{\circ}C$, $300^{\circ}C$, $400^{\circ}C$로 나누어 특성을 평가하였다. 시료는 파우더형태의 다결정CdTe를 120 g를 사용하여 증착완료 시, 약 $100{\mu}m$의 박막두께를 구현하였다. PVD증착의 조건으로는 Mo재질의 보트를 사용하였으며, 증착 시 진공도는 $5{\times}10^{-6}$ Torr, 보트온도는 약 $350^{\circ}C$ 소요시간은 5시간이었다. 증착이 완료된 CdTe의 표면구조와 전기적 특성평가를 위해 SEM촬영을 실시하였고, 전기적 특성 평가를 위해 CdTe표면에 Au를 PVD방식으로 증착하였다. 실험 결과 SEM촬영을 이용한 표면특성에서는 하부전극 기판의 온도가 높아질수록 표면 결정입자가 증가하는 것을 확인할 수 있었으며, 전기적 특성에서도 하부전극 기판의 온도가 증가할수록 RQA-5 조건의 70 kVp, 100 mA, 0.03 sec 엑스선에 대한 우수한 민감도와 암전류 값을 확인하였다. 이러한 결과는 증착과정에서 온도에 따른 다결정 CdTe의 표면결정 크기 증가는 동일한 면적에서 표면결정 수의 감소를 뜻한다. 이는 결정간의 경계에서 트랩 되어지는 전자가 감소하고, 전자의 이동도 또한 높은 효율을 나타냄을 확인할 수 있었다. 따라서 본 연구를 통하여 CdTe기반의 직접방식 엑스선 검출기 제작과정에서 증착 시 하부전극기판 온도가 증가할수록 결정의 크기가 증가하여 최적의 전기적 특성을 나타냄을 검증할 수 있었다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2004.05a
• /
• pp.195-198
• /
• 2004

본 논문은 움직임 예측에서 얻어지는 매크로블록(MB: MacroBlock)의 계수를 사용하는 장면전환 검출 알고리즘을 제안한다. 제안한 방법은 영상 시퀸스에서 픽쳐 형태에 따라 앞에 오는 2개의 영상 혹은 앞뒤의 두 영상과의 상관관계를 가지고 장면전환을 검출하므로 정확도가 높고 영상내의 모든 장면전환을 검출하였으며 P, B-picture는 물론이고 I-Picture에서 발생하는 장면전환도 검출할 수 있었다.

• 한국IT서비스학회:학술대회논문집
• /
• 2008.11a
• /
• pp.289-292
• /
• 2008

본 논문에서는 사용자에게 보다 효율적이고 직관적인 비디오 브라우징 서비스를 제공하기 위하여 비디오 데이터의 종류와 특성에 제한받지 않고 강건하게 적용될 수 있는 장면전환 검출 알고리즘을 제안하고자 한다. 제한된 알고리즘은 명암 값의 급 변화나 객체의 빠른 움직임, 영상의 왜곡 등에 의한 장면전환 검출의 오류를 제거할 수 있으며, 특히 연속된 프레임사이의 강건한 차이 값 추출을 위한 개선된 식을 제안하고, 추출된 차이 값들로부터 변화패턴을 학습하고 특징을 추출함으로서 자동 임계치 결정에 활용하였다. 제안된 방법은 급진적인 장면변화가 많고 플래시라이트와 같은 조명의 변화가 많은 다양한 비디오 데이터를 가지고 실험되었으며, 실험결과 기존의 방법에 비교하여 효율적이고 신뢰할 수 있는 결과 값들을 보여주었다.