Automatic license plate recognition for Iranianvehicles
Publish Year: 1393
نوع سند: مقاله کنفرانسی
زبان: English
View: 928
This Paper With 8 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
CSCCIT02_001
تاریخ نمایه سازی: 9 فروردین 1395
Abstract:
In recent years we have witnessed an increasing number of vehicles, resulting in the need to control thetransport and traffic clearly felt. The automatic recognition of vehicle license plates plays an important role in thisissue. In this paper, an automatic vehicle license plate recognition (ALPR) based on neural network is presented.Proposed system consists of three main steps: license plate locating, character extraction and recognition of charactersin image that extracted from input images using a SOM neural network. In plate locating section methods of edgedetection and morphology is used, on the other hand, by using the color image as input, special features that help theIranian plate is extracted. With combination results of these two parts, place of plate is located. To identify thecharacters, properties of prepared numbers and the numbers obtained from the extraction are used to identify numberswith SOM neural network. In this work we used 525 images that divided into two groups: low complexity (220 images)and high complexity (305 images). These images with 640×480 size have been considered as input of image processingalgorithm. In proposed method a very good success rate are obtained that plate locating is 99.24% and success rate ofcharacter recognition is 95.04%. Then after combining these success rates, final success rate is 96.76%.
Keywords:
Authors
Reza Bahrami
ICT Department, Malek Ashtar University of Technology, Tehran, Iran
Vahid Dolati
ICT Department, Malek Ashtar University of Technology, Tehran, Iran
Ali Jafari
ICT Department, Malek Ashtar University of Technology, Tehran, Iran
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :