• IEIE Transactions on Smart Processing and Computing
• /
• 제4권4호
• /
• pp.237-250
• /
• 2015

This paper describes a hardware-oriented algorithm and its conceptual implementation in a real-time speed limit traffic sign detection system on an automotive-oriented field-programmable gate array (FPGA). It solves the training and color dependence problems found in other research, which saw reduced recognition accuracy under unlearned conditions when color has changed. The algorithm is applicable to various platforms, such as color or grayscale cameras, high-resolution (4K) or low-resolution (VGA) cameras, and high-end or low-end FPGAs. It is also robust under various conditions, such as daytime, night time, and on rainy nights, and is adaptable to various countries' speed limit traffic sign systems. The speed limit traffic sign candidates on each grayscale video frame are detected through two simple computational stages using global luminosity and local pixel direction. Pipeline implementation using results-sharing on overlap, application of a RAM-based shift register, and optimization of scan window sizes results in a small but high-performance implementation. The proposed system matches the processing speed requirement for a 60 fps system. The speed limit traffic sign recognition system achieves better than 98% accuracy in detection and recognition, even under difficult conditions such as rainy nights, and is implementable on the low-end, low-cost Xilinx Zynq automotive Z7020 FPGA.

• 한국전자파학회논문지
• /
• 제24권5호
• /
• pp.502-507
• /
• 2013

본 논문에서는 개구면이 큰 기존 원형 이중 모드 혼의 E/H 평면 방사 패턴 대칭성 향상을 위한 변형된 개구면을 갖는 원형 이중 모드 혼을 제안한다. 제안된 안테나는 급전부, 모드 발생부, 위상정합부로 구성되며, 이 중 위상정합부의 원형 개구면은 E/H 평면의 방사 패턴 대칭성 향상을 위해 타원형 혹은 직사각형으로 변이된 형상을 갖는다. 제안된 안테나와 기존 원형 이중 모드 혼의 성능 비교를 위해 15 GHz에서 동작하는 기존 원형 이중모드 혼과 직사각형 개구면을 갖는 원형 이중 모드 혼을 제작하고 분석하였다. 측정 결과, f/d 비가 1인 반사판 안테나의 최대 개구 효율을 위한 급전 혼의 빔 폭인 -11 dB 빔 폭($53^{\circ}$) 내에서 기존 원형 이중 모드 혼의 경우 3.394 dB의 E/H 평면 방사 패턴 차이를 갖는 반면, 제안된 안테나는 오직 0.539 dB 차이를 가진다.

• Journal of Astronomy and Space Sciences
• /
• 제33권3호
• /
• pp.221-227
• /
• 2016

As described in the previous paper (Park et al. 2013), the detector subsystem of optical wide-field patrol (OWL) provides many observational data points of a single artificial satellite or space debris in the form of small streaks, using a chopper system and a time tagger. The position and the corresponding time data are matched assuming that the length of a streak on the CCD frame is proportional to the time duration of the exposure during which the chopper blades do not obscure the CCD window. In the previous study, however, the length was measured using the diagonal of the rectangle of the image area containing the streak; the results were quite ambiguous and inaccurate, allowing possible matching error of positions and time data. Furthermore, because only one (position, time) data point is created from one streak, the efficiency of the observation decreases. To define the length of a streak correctly, it is important to locate the endpoints of a streak. In this paper, a method using a differential convolution mask pattern is tested. This method can be used to obtain the positions where the pixel values are changed sharply. These endpoints can be regarded as directly detected positional data, and the number of data points is doubled by this result.