Vertical-Edge-Based Car-License-Plate Detection Method

Slides:



Advertisements
Similar presentations
Patient information extraction in digitized X-ray imagery Hsien-Huang P. Wu Department of Electrical Engineering, National Yunlin University of Science.
Advertisements

By: Mani Baghaei Fard.  During recent years number of moving vehicles in roads and highways has been considerably increased.
People Counting and Human Detection in a Challenging Situation Ya-Li Hou and Grantham K. H. Pang IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART.
A Graph based Geometric Approach to Contour Extraction from Noisy Binary Images Amal Dev Parakkat, Jiju Peethambaran, Philumon Joseph and Ramanathan Muthuganapathy.
QR Code Recognition Based On Image Processing
Road-Sign Detection and Recognition Based on Support Vector Machines Saturnino, Sergio et al. Yunjia Man ECG 782 Dr. Brendan.
IEEE Transactions on Consumer Electronics, Vol. 45, No. 1, AUGUST 1999 Muhammad Bilal Ahmad and Tae-Sun Choi, Senior Member,IEEE.
IntroductionIntroduction AbstractAbstract AUTOMATIC LICENSE PLATE LOCATION AND RECOGNITION ALGORITHM FOR COLOR IMAGES Kerem Ozkan, Mustafa C. Demir, Buket.
Discrete Finger and Palmar Feature Extraction for Personal Authentication Junta Doi, Member, IEEE,and Masaaki Yamanaka Advisor:Wen-Shiung Chen Student:
M.S. Student, Hee-Jong Hong
Compiled and Presented by KJ Tsiri Supervisor : Mr. Ismail.
Texture Segmentation Based on Voting of Blocks, Bayesian Flooding and Region Merging C. Panagiotakis (1), I. Grinias (2) and G. Tziritas (3)
Vehicle License Plate (VLP) Recognition System By German H. Flores and Gurpal Bhoot.
Real-time Embedded Face Recognition for Smart Home Fei Zuo, Student Member, IEEE, Peter H. N. de With, Senior Member, IEEE.
1 Learning to Detect Objects in Images via a Sparse, Part-Based Representation S. Agarwal, A. Awan and D. Roth IEEE Transactions on Pattern Analysis and.
Efficient Moving Object Segmentation Algorithm Using Background Registration Technique Shao-Yi Chien, Shyh-Yih Ma, and Liang-Gee Chen, Fellow, IEEE Hsin-Hua.
Text Detection in Video Min Cai Background  Video OCR: Text detection, extraction and recognition  Detection Target: Artificial text  Text.
An Approach to Korean License Plate Recognition Based on Vertical Edge Matching Mei Yu and Yong Deak Kim Ajou University Suwon, , Korea 指導教授 張元翔.
Smart Traveller with Visual Translator. What is Smart Traveller? Mobile Device which is convenience for a traveller to carry Mobile Device which is convenience.
California Car License Plate Recognition System ZhengHui Hu Advisor: Dr. Kang.
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 20, NO. 11, NOVEMBER 2011 Qian Zhang, King Ngi Ngan Department of Electronic Engineering, the Chinese university.
Liquan Shen Zhi Liu Xinpeng Zhang Wenqiang Zhao Zhaoyang Zhang An Effective CU Size Decision Method for HEVC Encoders IEEE TRANSACTIONS ON MULTIMEDIA,
Digital Image Processing - (monsoon 2003) FINAL PROJECT REPORT Project Members Sanyam Sharma Sunil Mohan Ranta Group No FINGERPRINT.
Joint Histogram Based Cost Aggregation For Stereo Matching Dongbo Min, Member, IEEE, Jiangbo Lu, Member, IEEE, Minh N. Do, Senior Member, IEEE IEEE TRANSACTION.
Case Studies Dr Lee Nung Kion Faculty of Cognitive Sciences and Human Development UNIVERSITI MALAYSIA SARAWAK.
Stereo Matching Information Permeability For Stereo Matching – Cevahir Cigla and A.Aydın Alatan – Signal Processing: Image Communication, 2013 Radiometric.
New Segmentation Methods Advisor : 丁建均 Jian-Jiun Ding Presenter : 蔡佳豪 Chia-Hao Tsai Date: Digital Image and Signal Processing Lab Graduate Institute.
An efficient method of license plate location Pattern Recognition Letters 26 (2005) Journal of Electronic Imaging 11(4), (October 2002)
Institute of Informatics and Telecommunications – NCSR “Demokritos” TEXT EXTRACTION FROM IMAGES AND VIDEOS Ινστιτούτο πληροφορικής και τηλεπικοινωνιών.
Vehicle License Plate Detection Algorithm Based on Statistical Characteristics in HSI Color Model Instructor : Dr. K. R. Rao Presented by: Prasanna Venkatesh.
Video Tracking Using Learned Hierarchical Features
1 Webcam Mouse Using Face and Eye Tracking in Various Illumination Environments Yuan-Pin Lin et al. Proceedings of the 2005 IEEE Y.S. Lee.
指導老師 : 蔡亮宙 報告者 : 黃柏愷 A new method of vehicle license plate location under complex scenes.
Color Image Segmentation Speaker: Deng Huipeng 25th Oct , 2007.
Digital Image Processing - (monsoon 2003) FINAL PROJECT REPORT Project Members Sanyam Sharma Sunil Mohan Ranta Group No FINGERPRINT.
Hierarchical Method for Foreground DetectionUsing Codebook Model Jing-Ming Guo, Yun-Fu Liu, Chih-Hsien Hsia, Min-Hsiung Shih, and Chih-Sheng Hsu IEEE TRANSACTIONS.
NTIT IMD 1 Speaker: Ching-Hao Lai( 賴璟皓 ) Author: Hongliang Bai, Junmin Zhu and Changping Liu Source: Proceedings of IEEE on Intelligent Transportation.
Automatic License Plate Location Using Template Matching University of Wisconsin - Madison ECE 533 Image Processing Fall 2004 Project Kerry Widder.
IEEE Transactions on Consumer Electronics, Vol. 58, No. 2, May 2012 Kyungmin Lim, Seongwan Kim, Jaeho Lee, Daehyun Pak and Sangyoun Lee, Member, IEEE 報告者:劉冠宇.
James C. Tilton Code Computational & Information Sciences and Technology Office NASA Goddard Space Flight Center January 17, 2014 update National.
Compiled and Presented by KJ Tsiri Supervisor : Mr. Ismail.
Zhongyan Liang, Sanyuan Zhang Under review for Journal of Zhejiang University Science C (Computers & Electronics) Publisher: Springer A Credible Tilt License.
Automated Fingertip Detection
Robust Nighttime Vehicle Detection by Tracking and Grouping Headlights Qi Zou, Haibin Ling, Siwei Luo, Yaping Huang, and Mei Tian.
October 1, 2013Computer Vision Lecture 9: From Edges to Contours 1 Canny Edge Detector However, usually there will still be noise in the array E[i, j],
Wonjun Kim and Changick Kim, Member, IEEE
Face Detection Final Presentation Mark Lee Nic Phillips Paul Sowden Andy Tait 9 th May 2006.
Scene Text Extraction Using Focus of Mobile Camera Egyul Kim, SeongHun Lee, JinHyung Kim Artificial Intelligence & Pattern Recognition Lab, KAIST, Korea.
TOPIC 12 IMAGE SEGMENTATION & MORPHOLOGY. Image segmentation is approached from three different perspectives :. Region detection: each pixel is assigned.
Preliminary Transformations Presented By: -Mona Saudagar Under Guidance of: - Prof. S. V. Jain Multi Oriented Text Recognition In Digital Images.
Instructor: Mircea Nicolescu Lecture 5 CS 485 / 685 Computer Vision.
Chapter 6 Skeleton & Morphological Operation. Image Processing for Pattern Recognition Feature Extraction Acquisition Preprocessing Classification Post.
TUMOR BURDEN ANALYSIS ON CT BY AUTOMATED LIVER AND TUMOR SEGMENTATION RAMSHEEJA.RR Roll : No 19 Guide SREERAJ.R ( Head Of Department, CSE)
Shen-Chuan Tai, Chien-Shiang Hong, Cheng-An Fu National Cheng Kung University, Tainan City,Taiwan (R.O.C.),DCMC Lab Pacific-Rim Symposium on Image and.
Portable Camera-Based Assistive Text and Product Label Reading From Hand-Held Objects for Blind Persons.
Shadow Detection in Remotely Sensed Images Based on Self-Adaptive Feature Selection Jiahang Liu, Tao Fang, and Deren Li IEEE TRANSACTIONS ON GEOSCIENCE.
An intelligent strategy for checking the annual inspection status of motorcycles based on license plate recognition Yo-Ping Huang a, Chien-Hung Chen b,
Evaluation of Gender Classification Methods with Automatically Detected and Aligned Faces Speaker: Po-Kai Shen Advisor: Tsai-Rong Chang Date: 2010/6/14.
License Plate Recognition of A Vehicle using MATLAB
Submitted by ANGELA LINCY.J( ) RENJU.K.S( ) ELCY GEORGE( ) GUIDE NAME: Mrs. J. SAHAYA JENIBA ASSISTANT PROFESSOR, COMPUTER.
Face Detection 蔡宇軒.
Optical Character Recognition
April 21, 2016Introduction to Artificial Intelligence Lecture 22: Computer Vision II 1 Canny Edge Detector The Canny edge detector is a good approximation.
Automatic License Plate Recognition for Electronic Payment system Chiu Wing Cheung d.
Digital Image Processing - (monsoon 2003) FINAL PROJECT REPORT
A new data transfer method via signal-rich-art code images captured by mobile devices Source: IEEE Transactions on Circuits and Systems for Video Technology,
A New Approach to Track Multiple Vehicles With the Combination of Robust Detection and Two Classifiers Weidong Min , Mengdan Fan, Xiaoguang Guo, and Qing.
Image Segmentation Techniques
Computer Vision Lecture 9: Edge Detection II
Fast Decision of Block size, Prediction Mode and Intra Block for H
Presentation transcript:

Vertical-Edge-Based Car-License-Plate Detection Method Abbas M. Al-Ghaili, Syamsiah Mashohor, Abdul Rahman Ramli, and Alyani Ismail IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 62, NO. 1, JANUARY 2013

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Introduction The car-license-plate (CLP) recognition system is an image processing technology used to identify vehicles by capturing their CLPs. Usually, a Car License Plate Detection and Recognition System consists of three parts: license-plate detection (LPD), character segmentation, and character recognition Among these, LPD is the most important part in the system because it affects the system’s accuracy

Introduction Reduce the computation time and algorithm complexity. Build the LP recognition system with lower cost.

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Related Work There are several LPD methods that have been used before, such as morphological operations [10]–[12], edge extraction [13]–[18]… [10] J.-W. Hsieh, S.-H. Yu, and Y. S. Chen, “Morphology-based license plate detection from complex scenes,” in Proc. 16th Int. Conf. Pattern Recognit., Quebec City, QC, Canada, 2002, pp. 176–179. [11] A. A. Lensky, J. Kang-Hyun, and V. V. Gubarev, “Vehicle license plate detection using local fractal dimension and morphological analysis,” in Proc. IEEE 1st Int. Forum Strategic Technol., Ulsan, Korea, 2006, pp. 47–50. [12] F. Martin, O. Martin, M. García, and J. L. Alba, “New methods for automatic reading of VLP’s (Vehicle License Plates),” in Proc. IASTED Int. Conf. Signal Process. Pattern Recognit. Appl., Heraklion, Greece, 2002, pp. 126–131.

[13] S. -H. Le, Y. -S. Seok, and E. -J [13] S.-H. Le, Y.-S. Seok, and E.-J. Lee, “Multi-national integrated car-license plate recognition system using geometrical feature and hybrid pattern vector,” in Proc. Int. Tech. Conf. Circuits Syst. Comput. Commun., Phuket, Thailand, 2002, pp. 1256–1259. [14] J. R. Parker and P. Federl, “An approach to license plate recognition,” in Proc. Visual Interface, Kelowna, BC, Canada, 1997, pp. 178–182. [15] M. Yu and Y. D. Kim, “An approach to Korean license plate recognition based on vertical edge matching,” in Proc. IEEE Int. Conf. Syst., Man, Cybern., 2000, pp. 2975–2980. [16] H. Zhang, W. Jia, X. He, and Q. Wu, “A fast algorithm for license plate detection in various conditions,” in Proc. IEEE Int. Conf. Syst., Man, Cybern., Taipei, Taiwan, 2006, pp. 2420–2425. [17] D. Zheng, Y. Zhao, and J. Wang, “An efficient method of license plate location,” Pattern Recognit. Lett., vol. 26, no. 15, pp. 2431–2438, Nov. 2005. [18] J.-M. Guo and Y.-F. Liu, “License plate localization and character segmentation with feedback self-learning and hybrid binarization techniques,” IEEE Trans. Veh. Technol., vol. 57, no. 3, pp. 1417–1424, May 2008.

Related Work In [28], Matas and Zimmermann proposed an algorithm to detect LPs under various conditions. However, this algorithm could hardly highlight characters overlapping from the true LPs. [28] J. Matas and K. Zimmermann, “Unconstrained license plate and text localization and recognition,” in Proc. IEEE Int. Conf. Intell. Transp. Syst., Vienna, Austria, 2005, pp. 572–577.

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - Overview The color input image is converted to a grayscale image, and then, adaptive thresholding (AT) is applied on the image to constitute the binarized image.

Proposed method - Overview The ULEA is applied to remove noise and to enhance the binarized image. Then, the vertical edges are extracted by using the VEDA.

Proposed method - Overview The plate details are highlighted based on the pixel value with the help of the VEDA output. Then, several candidate regions will be found. Finally we can find the true plate region.

Proposed method - Overview

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - AT Average=SumWindow /S 2 (i,j) S S = 1/8 width

Proposed method - AT If the value of the current pixel is T percent lower than the average, then it is set to black; otherwise, it is set to white. t(i , j) = (1-T) X Average O(i , j) = { In this method, 0.1 < T < 0.2. 0, g(i , j) < t(i , j) 1, otherwise

Proposed method - AT Input image After AT is applied

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - ULEA Unwanted Line Elimination Algorithm Thresholding process in general produces many thin lines. We have proposed an algorithm to eliminate them.

Proposed method - ULEA There are four cases in which unwanted lines can be formed. — (0◦), | (90◦), / (45◦), \ (135◦)

Proposed method - ULEA Only black pixel values in the thresholded image are tested. Only the lines whose widths equal to 1 pixel are checked.

Proposed method - ULEA A 3 × 3 mask is used throughout all image pixels. If two corresponding values are white together, then the current pixel is converted to a white

Proposed method - ULEA Thresholded Image After ULEA

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - VEDA The advantage of the VEDA is to distinguish the plate detail region, particularly the beginning and the end of each character. The idea of the VEDA concentrates on intersections of black–white and white–black.

Proposed method - VEDA ULEA Img After VEDA

Proposed method - VEDA The proposed mask has the size of 2 X 4 If the four pixels at locations (0, 1), (0, 2), (1, 1), and (1, 2) are black, then the other mask values are tested if whether they are black or not. If the whole values are black, then the two locations at (0, 1) and (1, 1) will be converted to white.

Proposed method - VEDA The first edge can have a black-pixel width of 2, and the second edge can have a black-pixel width of 1.

Proposed method - VEDA ULEA Img After VEDA

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

HDDs based on VEDA This process depends on the VEDA output in highlighting the plate region. Horizontal Distance(Hd) : Distance between two vertical edge.

HDDs based on VEDA hd : distance between 2 vertical edge LP backgroung Character White backgroung

HDDs based on VEDA HDD Output

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - CRE Find the possible LP Regions. Find how many lines each row. 10 row => 1 Group. Find how many lines in one group.

Proposed method - CRE HwMnyLines[a] , where a = 0,1,2……,height-1.

Proposed method - CRE

Proposed method - CRE

Proposed method - CRE Use a threshold to eliminate those unsatisfied groups The optimal threshold for the best detection rate has been found is ≥ 1/20 × image_height.

Proposed method - CRE Threshold = 15 Lines

Proposed method - CRE Select Boundaries of Candidate Regions

Outline Introduction Related Work Proposed method Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Proposed method - PRS This process aims to select and extract one correct LP.

Proposed method - PRS

Proposed method - PRS The columns whose top and bottom neighbors have high ratios of blackness details are given one vote.

Proposed method - PRS

Proposed method - PRS

Outline Introduction Proposed method Experiment Result and Discussion Overview AT ULEA VEDA HDDs based on VEDA CRE PRS Experiment Result and Discussion

Experiment Result and Discussion VEDA Versus Sobel Accuracy :

Experiment Result and Discussion VEDA Versus Sobel Computation Time :

Experiment Result and Discussion Method Evaluation The whole tested images are classified into three groups. The first group (G1) includes the ideal, blurry, and defected plates. The second group (G2) contains double-row LPs, trees, similarity of LP and car body, and more than one LP or a car in a single image. The third group (G3) contains excessive light, rainy, shady, and low-contrast images.

Experiment Result and Discussion Method Evaluation

Experiment Result and Discussion Method Evaluation

[6] R. Parisi, E. D. Di Claudio, G. Lucarelli, and G [6] R. Parisi, E. D. Di Claudio, G. Lucarelli, and G. Orlandi, “Car plate recognition by neural networks and image processing,” in Proc. IEEE Int. Symp. Circuits Syst., 1998, pp. 195–198. [18] J.-M. Guo and Y.-F. Liu, “License plate localization and character segmentation with feedback self-learning and hybrid binarization techniques,” IEEE Trans. Veh. Technol., vol. 57, no. 3, pp. 1417–1424, May 2008. [21] H. Caner, H. S. Gecim, and A. Z. Alkar, “Efficient embedded neuralnetwork- based license plate recognition system,” IEEE Trans. Veh. Technol., vol. 57, no. 5, pp. 2675–2683, Sep. 2008. [25] S. K. Kim, D. W. Kim, and H. J. Kim, “A recognition of vehicle license plate using a genetic algorithm based segmentation,” in Proc. Int. Conf. Image Process., Lausanne, Switzerland, 1996, pp. 661–664. [42] C. C. Ming et al., “ProjekCARPET, Car License Plate Extraction & Recognition Technology,” presented at the National Innovation Summit, Kuala Lumpur, Malaysia, Apr. 29, 2004. [Online]. Available: http://www. projekcarpet.com.my/press77.html

Conclusion A new and fast algorithm for CLPD. The computation time of the CLPD method is 47.7 ms, which meets the real-time requirements. In the experiment, the rate of correctly detected LPs is 91.4%.