• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.789-792
• /
• 2001

본 논문은 자동차의 번호판 인식 시스템의 한 부분인 번호판 추출과 자모 분리를 통한 문자 인식까지의 과정을 실험한 것이다. 본 논문은 gray-level에서 영상을 실험하였고, 번호판을 추출하기 위해서 morphology를 반복 적용하고 크기 보정을 통해 번호판을 추출하며, hough transform을 이용한 크기 재보정을 통해 최종적으로 번호판을 추출한다. 그리고, 문자 인식 단계에서는 먼저 hough transform을 사용하여 한글의 모음의 시작점을 얻고, 문자 특징을 이용하여 자음과 모음을 분리하여 모음을 인식한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.773-776
• /
• 2004

본 논문에서는 에지 검출기에 의해 다양한 명암이 존재하는 차량 번호판 영역의 사각형 에지를 검출시 사용되는 소벨 및 Prewitt, Roberts, 가우시안의 라플라시안, 그리고 Canny 검출기를 사용하여 처리 속도와 에지 검출의 정확성을 실험하여 각 연산자의 성능을 평가하였다. 기존의 Sobel 에지 검출기는 적응적 임계값을 구하지 않으면 다양한 조명의 영향에 강인하지 못하다. 또한 Canny 에지 검출기는 조명의 영향에 강인하기는 하나, 계산량이 Sobel 보다는 많아 처리 속도가 느리다. 색상에 의해 번호판 후보 영역을 추출한 후 에지 검출기 번호판 내의 명암이 둘 이상으로 차량 번호판 영역에 대해서, 다양한 에지 검출기를 적용하여 속도와 에지 검출 성능을 비교 평가하고자 한다.

• International journal of advanced smart convergence
• /
• v.7 no.4
• /
• pp.27-39
• /
• 2018

Since the global positioning system (GPS) has been included in mobile devices (e.g., for car navigation, in smartphones, and in smart watches), the impact of personal GPS log data on daily life has been unprecedented. For example, such log data have been used to solve public problems, such as mass transit traffic patterns, finding optimum travelers' routes, and determining prospective business zones. However, a real-time analysis technique for GPS log data has been unattainable due to theoretical limitations. We introduced a machine learning model in order to resolve the limitation. In this paper presents a new, three-stage real-time prediction model for a person's daily route activity. In the first stage, a machine learning-based clustering algorithm is adopted for place detection. The training data set was a personal GPS tracking history. In the second stage, prediction of a new person's transient mode is studied. In the third stage, to represent the person's activity on those daily routes, inference rules are applied.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.713-725
• /
• 2019

Automatic License Plate Recognition (ALPR) is a technology required for many applications such as Intelligent Transportation Systems and Video Surveillance Systems. Most of the studies have studied were about the detection and recognition of license plates on cars, and there is very little about detecting and recognizing license plates on motorbikes. In the case of a car, the license plate is located at the front or rear center of the vehicle and is a straight or slightly sloped license plate. Also, the background of the license plate is mainly monochromatic, and license plate detection and recognition process is less complicated. However since the motorbike is parked by using a kickstand, it is inclined at various angles when parked, so the process of recognizing characters on the motorbike license plate is more complicated. In this paper, we have developed a 2-stage YOLOv2 algorithm to detect the area of a license plate after detection of a motorbike area in order to improve the recognition accuracy of license plate for motorbike data set parked at various angles. In order to increase the detection rate, the size and number of the anchor boxes were adjusted according to the characteristics of the motorbike and license plate. Image warping algorithms were applied after detecting tilted license plates. As a result of simulating the license plate character recognition process, the proposed method had the recognition rate of license plate of 80.23% compared to the recognition rate of the conventional method(YOLOv2 without image warping) of 47.74%. Therefore, the proposed method can increase the recognition of tilted motorbike license plate character by using the adjustment of anchor boxes and the image warping which fit the motorbike license plate.

• Journal of Digital Convergence
• /
• v.13 no.10
• /
• pp.313-318
• /
• 2015

A traditional speech enhancement methods distort the sound spectrum generated according to estimation of the remaining noise, or invalid noise is a problem of lowering the speech recognition performance. In this paper, we propose a speech detection method that convergence the sound energy distribution process and sound energy parameters. The proposed method was used to receive properties reduce the influence of noise to maximize voice energy. In addition, the smaller value from the feature parameters of the speech signal The log energy features of the interval having a more of the log energy value relative to the region having a large energy similar to the log energy feature of the size of the voice signal containing the noise which reducing the mismatch of the training and the recognition environment recognition experiments Results confirmed that the improved recognition performance are checked compared to the conventional method. Car noise environment of Pause Hit Rate is in the 0dB and 5dB lower SNR region showed an accuracy of 97.1% and 97.3% in the high SNR region 10dB and 15dB 98.3%, showed an accuracy of 98.6%.

• Journal of Internet Computing and Services
• /
• v.9 no.3
• /
• pp.51-58
• /
• 2008

Wireless sensor networks are used by various application areas to implement smart data processing and ubiquitous system. In the recent research of parking management system based on wireless sensor networks, adaptive sensing and efficient data processing are not considered. The effectiveness of implementing these distributed computing devices affects the performance of the applications in parking management. This paper proposes an adaptive sensing using fuzzy wireless sensor for the ubiquitous networks of parking management system. The fuzzy inference system is encoded in the sensor for efficient car presence detection. Moreover, a rule base adaptive module is proposed which wirelessly transmit the new values to each sensor for adapting the environment of car park area. The result of experiments shows that the fuzzy wireless sensor provides more throughputs and less time delays compared to a normal method of data gathering by wireless sensors.

• Journal of Korean Society on Water Environment
• /
• v.26 no.4
• /
• pp.622-628
• /
• 2010

In this study, Solid-phase microextraction (SPME) with GC/FID was studied as a possible alternative to liquid-liquid extraction for the analysis of BTEX and MTBE. Experimental parameters affecting the SPME process (such as kind of fibers, adsorption time, desorption time, volume ratio of sample to headspace, salt addition, and magnetic stirring) were optimized. Experimental parameters such as CAR/PDMS, adsorption time of 20 min, desorption time of 5 min at $250^{\circ}C$, headspace volume of 50 mL, sodium chloride (NaCl) concentration of 25% combined with magnetic stirring were selected in optimal experimental conditions for analysis of BTEX and MTBE. The general affinity of analytes to CAR/PDMS fiber was high in the order p-Xylene>Toluene>Ethylbenzene>MTBE>Benzene. The linearity of $R^2$ for BTEX and MTBE was from 0.970 to 0.999 when analyte concentration ranges from $30{\mu}g/L$ to $500{\mu}g/L$, respectively. The relative standard deviation (% RSD) were from 2.5% to 3.2% for concentration of $100{\mu}g/L$ (n=5), respectively. Finally, the limited of detection (LOD) observed in our study for BTEX and MTBE were from $7.5{\mu}g/L$ to $15{\mu}g/L$, respectively.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.8
• /
• pp.2555-2564
• /
• 2000

In this paper, a new unsupervised image segmentation algorithm is proposed. To model the contextual information presented in images, the characteristics of the Markov random fields (MRF) are utilized. Textured images are modeled as realizations of the stationary Gaussian MRF on a two-dimensional square lattice using the conditional autoregressive (CAR) equations with a second-order noncausal neighborhood. To detect boundaries, hypothesis tests over two masked areas are performed. Under the hypothesis, masked areas are assumed to belong to the same class of textures and CAR equation parameters are estimated in a minimum-mean-square-error (MMSE) sense. If the hypothesis is rejected, a measure of dissimilarity between two areas is accumulated on the rejected area. This approach produces potential edge maps. Using these maps, boundary detection can be performed, which resulting no micro edges. The performance of the proposed algorithm is evaluated by some experiments using real images as weB as synthetic ones. The experiments demonstrate that the proposed algorithm can produce satisfactorY segmentation without any a priori information.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.5C
• /
• pp.632-641
• /
• 2004

In this paper, a new intermediate view reconstruction method employing a stereo image rectification algorithm by which an uncalibrated input stereo image can be transformed into the calibrated one is suggested and its performance is analyzed. In the proposed method, feature point are extracted from the stereo image pair though detection of the corners and similarities between each pixel of the stereo image. And then, using these detected feature points, the moving vectors between stereo image and the epipolar line is extracted. Finally, the input stereo image is rectified by matching the extracted epipolar line between the stereo image in the horizontal direction and intermediate views are reconstructed by using these rectified stereo images. From some experiments on synthesis of the intermediate views by using three kinds of stereo image; a CCETT's stereo image of 'Man' and two stereo images of 'Face' ＆ 'Car' captured by real camera, it is analyzed that PSNRs of the intermediate views reconstructed from the calibrated image by using the proposed rectification algorithm are improved by 2.5㏈ for 'Man', 4.26㏈ for 'Pace' and 3.85㏈ for 'Car' than !hose of the uncalibrated ones. This good experimental result suggests a possibility of practical application of the unposed stereo image rectification algorithm-based intermediate view reconstruction view to the uncalibrated stereo images.

• The Bulletin of The Korean Astronomical Society
• /
• v.36 no.2
• /
• pp.152.1-152.1
• /
• 2011

We report preliminary results of long-slit near-infrared (NIR) spectroscopy of Luminous Blue Variables (LBVs) with moderate resolution of R ~ 2400. We obtained Jshort (1.04-1.26 micron) and Ks (2.02-2.31 micron) band spectra of 4 LBVs and 3 LBV candidates in Southern hemisphere using IRIS2, infrared imager and spectrograph, mounted on the 4-m Anglo-Australian Telescope. All targets are fairly bright in NIR so that we can obtain high signal-to-noise ratio for clear line detection and modeling. They are also widely distributed in the HR diagram so that we can compare the spectral properties of LBVs in different temperature and luminosity ranges. Among them, we present the results of two well-known LBVs AG Car and HR Car. Their spectra show similar properties with hydrogen, He I, and metallic lines such as Fe II and Mg II, most of them in emission. We discuss, in particular, the He I 1.083 micron lines formed in stellar wind because these two LBVs show large variation in their He I line intensities, compared to previous studies. Since the He I 1.083 line is known to be anticorrelated with the photometric variation of LBVs, strong line intensities with P-Cygni profiles in both stars indicate that they are now near the visual minimum phase. We model the obtained spectra using non-LTE atmosphere code CMFGEN of Hillier (1998) to derive stellar parameters such as wind velocity and mass loss rate, and discuss the long-term variability of stellar parameters of these LBVs. deduced from our otometric solution.