• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.792-794
• /
• 2014

Character recognition is an image processing technique for obtaining the character information such as documents and automobile license plate and for this edge detection methods are commonly used. The previous edge detection methods are mostly applying the weighted value mask on the image and because it applies the same mask to the entire areas of the image, the processing results are somewhat insufficient. Therefore, this paper has proposed an edge detection algorithm by applying the weighted value mask considering the distribution and location of pixels to be suitable for the character recognition.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.67-69
• /
• 2010

최근 자동차 업계와 IT 기술의 융합이 새로운 트렌드로 자리 잡으면서 전자제어 기술뿐만 아니라 영상처리 기술이 융합된 지능형 자동차 개발에 대한 연구가 활발히 진행되고 있다. 차선 또는 번호판을 대상으로 하는 인식 알고리즘은 이미 다양한 방법으로 연구가 진행되어 왔으며 이미 몇몇 기술은 상용화 단계에 있다. 본 논문에서는 Viola-Jones 알고리즘을 이용하여 차량의 사각 지대에 위치하는 차량을 감지하고 이의 대략적인 거리 정보를 추정하는 것을 목표로 하여 차량의 형태 정보를 바탕으로 차량을 감지하는 알고리즘을 제안한다. 기본적인 방법은 Adaboost와 Harr-like 특징을 사용하여 얼굴을 성공적으로 검출한 Viola-Jones 알고리즘[1]을 차량에 적용하였다.

• Proceedings of the KAIS Fall Conference
• /
• 2010.05a
• /
• pp.225-227
• /
• 2010

최근 증강현실(Augmented Reality, AR) 기술의 중요성이 인식되면서 다양한 분야의 서비스에 기술 도입사례가 등장하고 있다. AR에 있어 마커 검출 기술은 가장 기본이 되는 중요한 기술이다. 하지만 마커 기반의 증강현실 시스템은 그 오차가 매우 크고 이로 인해 센서정보와 같은 다양한 보조적 정보를 요구하게 된다. 따라서 본 연구에서는 마커사용의 한계를 극복하기 위한 마커리스 트래킹(Markerless Tracking Technology)기술을 연구 하여 실시간 영상에서 목표로 하는 사물을 추적하여 증강현실 서비스로 도입하기 위한 시스템은 설계하였다. 본 시스템은 기존에 도입된 명함 분석, 자동차 번호판 인식 등 제한된 서비스의 한계를 극복하고 보다 다양한 연구 분야에 활용될 것이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.5
• /
• pp.877-889
• /
• 2006

This paper introduces a simple but effective method for recognizing vehicle models corresponding to each maker by information and images for moving vehicles. The proposed approach is implemented by combination of the breadth detection mechanism using the vehicle's pressure, exact height detection by a laser scanning, and license plate recognition for classifying specific vehicles. The implemented system is therefore capable of robust classification with real-time vehicle's moving images and established sensors. Simulation results using the proposed method on synthetic data as well as real world images demonstrate that proposed method can maintain an excellent recognition rate for moving vehicle models because of image acquisition by 2-D CCD and various image processing algorithms.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.7
• /
• pp.655-661
• /
• 2017

It is a very important task to extract character regions contained in various input color images, because characters can provide significant information representing the content of an image. In this paper, we propose a new method for extracting character regions from various input images using MCT features and an AdaBoost algorithm. Using geometric features, the method extracts actual character regions by filtering out non-character regions from among candidate regions. Experimental results show that the suggested algorithm accurately extracts character regions from input images. We expect the suggested algorithm will be useful in multimedia and image processing-related applications, such as store signboard detection and car license plate recognition.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1166-1174
• /
• 2017

In general, the detection of the vehicle license plate is a previous step of license plate recognition and has been actively studied for several decades. In this paper, we propose an algorithm to detect a license plate area of a moving vehicle from a video captured by a fixed camera installed on the road using the Convolution Neural Network (CNN) technology. First, license plate images and non-license plate images are applied to a previously learned CNN model (AlexNet) to extract and classify features. Then, after detecting the moving vehicle in the video, CNN detects the license plate area by comparing the features of the license plate region with the features of the license plate area. Experimental result shows relatively good performance in various environments such as incomplete lighting, noise due to rain, and low resolution. In addition, to protect personal information this proposed system can also be used independently to detect the license plate area and hide that area to secure the public's personal information.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.18 no.4
• /
• pp.1-11
• /
• 2022

Recently, research on license plate recognition, which is a core technology of an intelligent transportation system(ITS), is being actively conducted. In this paper, we propose a method to extract numbers and characters from low-quality license plate images by applying the YOLOv4 algorithm. YOLOv4 is a one-stage object detection method using convolution neural network including BACKBONE, NECK, and HEAD parts. It is a method of detecting objects in real time rather than the previous two-stage object detection method such as the faster R-CNN. In this paper, we studied a method to directly extract number and character regions from low-quality license plate images without additional edge detection and image segmentation processes. In order to evaluate the performance of the proposed method we experimented with 500 license plate images. In this experiment, 350 images were used for training and the remaining 150 images were used for the testing process. Computer simulations show that the mean average precision of detecting number and character regions on vehicle license plates was about 93.8%.

• Journal of the Korea Convergence Society
• /
• v.9 no.10
• /
• pp.7-12
• /
• 2018

In today's rapidly growing contemporary society, there is a tendency for demand to automate production processes by utilizing the vision system. In general, image recognition is mainly concerned with embossed characters such as license plates, and research on recognition of intaglio characters is very limited. Especially, intaglio characters, which are marked on rubber related products such as tire surfaces, have difficulty in recognizing characters or numbers through image because the difference in brightness with surrounding is not so large. In this paper, we propose a system to improve the recognition rate of characters marked on intaglio rubber products such as tire surfaces. Also, it can be applied flexibly according to the lighting environment. Through the proposed system, production and inventory management and defect detection can be processed quickly by applying to the production process of tire and rubber products.