• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11a
• /
• pp.275-278
• /
• 2003

자동차 번호판 인식 시스템은 영상획득, 번호판추출, 전처리(이진화), 문자영역 분할, 문자인식 등의 5가지 핵심 부분으로 구성된다. 따라서 자동차 번호판 인식 시스템의 최종 인식율은 각 단계의 성능에 따라 직접적인 영향을 받는다. 본 논문은 영상처리 기법을 이용하여 영상에서 번호판 영역을 추출을 위한 연구로 문자인식 단계에서 높은 인식율을 확보할 수 있도록 빠른 연산속도와 추출 정확성을 높일 수 있는 알고리즘을 제안한다.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2004.07a
• /
• pp.352-352
• /
• 2004

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.01a
• /
• pp.109-110
• /
• 2016

본 논문에서는 효율적으로 주차 공간을 확보와 주차장 성능을 향상을 위한 방법을 조사하였다. 이러한 방법으로는 차량 번호판 조사를 이용하여 주차 회전율을 구하는 방법이 있었다. 본 연구로 효율적으로 주차장을 사용하고 있는지를 판단할 수 있다. 또한 차량의 주차를 하여 잘 소통되는지를 알 수 있었다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.346-351
• /
• 2002

본 연구에서는 차량 번호판에서 추출된 문자영역의 DCT(Digital Cosine Transform) 계수와 LVQ (Learning Vector Quantization) 신경회로망을 이용하여 차량 번호판 인식 시스템을 구성하였다. 입력된 차량영상의 RGB 칼라정보를 이용하여 번호판 영역을 추출하고 추출된 번호판의 히스토그램과 문자의 상대적 위치정보를 병합하여 문자영역을 추출하였다. 이렇게 추출된 문자영역의 명암도 영상에 DCT를 적용하여 얻은 특징 벡터는 LVQ 신경회로망의 입력으로 사용되어 인식 과정을 수행한다. 제안된 시스템의 검증을 위하여 다양한 환경에서 촬영된 109대의 자가용 차량영상에 대하여 실험하여 상대적으로 높은 번호판 영역 추출율과 인식률을 보였다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.6
• /
• pp.453-459
• /
• 2015

In this paper, we propose a new digit recognition method based on opened enclosure. In the proposed method, each digit is divided into two parts, an upper part and a lower part, which are based on a cutting line that is modified depending on the number of intersection points. In the simulation, the performance evaluation through the data acquisition and application of the proposed algorithm was carried out and the result was presented.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.185-197
• /
• 2022

Recently, deep learning networks with high performance for object recognition are emerging. In the case of object recognition using deep learning, it is important to build a training data set to improve performance. To build a data set, we need to collect and label the images. This process requires a lot of time and manpower. For this reason, open data sets are used. However, there are objects that do not have large open data sets. One of them is data required for license plate detection and recognition. Therefore, in this paper, we propose an artificial license plate generator system that can create large data sets by minimizing images. In addition, the detection rate according to the artificial license plate arrangement structure was analyzed. As a result of the analysis, the best layout structure was FVC_III and B, and the most suitable network was D2Det. Although the artificial data set performance was 2-3% lower than that of the actual data set, the time to build the artificial data was about 11 times faster than the time to build the actual data set, proving that it is a time-efficient data set building system.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.5
• /
• pp.987-994
• /
• 2018

In this paper, an automatic transmission system of traffic information is presented. Here, a traffic event is defined as an obstacle to an emergency vehicle such as an ambulance or a fire truck. When a traffic event is detected from a video recorded by a black box installed in a vehicle, the implemented system automatically transmits a proof image and corresponding information to the control center through an e-mail. For this purpose, we realize an algorithm of identifying the numbers and a character from the license plate, and an algorithm for determining the occurrence of a traffic event. To report the event, a function for automatic transmission of the text and image files through e-mail and file transfer protocol (FTP) is also appended. Therefore, if the traffic event is extended and applied to the presented system, it will be possible to establish a convenient reporting system for the violation of various traffic regulations. In addition, it will contribute to significantly reduce the number of traffic violations against the regulations.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.238-240
• /
• 2004

경제 성장과 생활 수준의 향상으로 인한 자동차 수의 증가는 않은 문제를 발생시키고 있다 제한된 인력과 비용으로 효율적인 자동차 관리를 위한 연구 분야 중에서 자동차 번호판 인식 (Vehicle Plate Recognition) 기술은 법규위반의 식별, 통행료 징수, 납세, 도난.도주 차량 확인 및 주차 관리 등의 않은 분야에 응용되고 있다. 자동차 번호판 문자 인식 문제와 같이 훈련 예제 수집 비용이 많이 드는 경우에 제한된 수의 훈련 예제를 최대한 활용하여 분류 성능을 향상시키기 위한 방안 중 하나로, 수집된 훈련 예제들로부터 가상의 예제를 생성하고, 생성된 가상 예제를 훈련 예제로 추가하여 학습하는 절러 연구가 수행된 바 있다. 본 논문에서는 자동차 번호판 문자 인식의 성능 향상을 위친 수집된 예제들을 적절히 병합하여 가상의 예제를 생성하는 방안에 관해 기술하고, 문자 인식 분야에서 일반적으로 많이 사용되는 여러 알고리즘에 대하여 다양한 가상 예제 생성 방안 및 다양한 생성 비율 따른 실험을 통해 그 효용성을 확인하였다

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.115.5-115
• /
• 2002

In this paper, we introduce a system that detects all kinds of violations at a street intersection such as red light running, speed violation, stop line violation and lane violation by tracking individual vehicles. Two cameras are used for defecting violations. One is an analog camera for real-time tracking and the other is a digital camera for license plate reading. This system is connected to the traffic signal system controller and monitors the red, arrow, yellow and green phases of an approach. Two loops in the road are used to detect vehicle approach and speed. The system takes pictures of all vehicles passing a second loop and tracks the vehicles until they go out a street intersection...

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.1
• /
• pp.73-80
• /
• 2016

The Car License Plate Recognition(: CLPR) is required in searching the hit-and-run car, measuring the traffic density, investigating the traffic accidents as well as in pursuing vehicle crimes according to the increasing in number of vehicles. The captured images on the real environment of the CLPR is contaminated not only by snow and rain, illumination changes, but also by the geometrical distortion due to the pose changes between camera and car at the moment of image capturing. We propose homographic transformation and intensity histogram of vertical image projection so as to transform the distorted input to the original image and cluster the character and number, respectively. Especially, in this paper, the Multilayer Perceptron Algorithm(: MLP) in the CLPR is used to not only recognize the charcters and car license plate, but also determine the optimized ratio among the number of input, hidden and output layers by the real experimental result.