Many real-life machine and computer vision applications are focusing on object detection and recognition. In recent years, deep learning-based approaches gained increasing interest due to their high accuracy levels. License Plate (LP) detection and classification have been studied extensively over the last decades. However, more accurate and language-independent approaches are still required. This paper presents a new approach to detect LPs and recognize their country, language, and layout. Furthermore, a new LP dataset for both multi-national and multi-language detection, with either one-line or two-line layouts is presented. The YOLOv2 detector with ResNet feature extraction core was utilized for LP detection, and a new low complexity convolutional neural network architecture was proposed to classify LPs. Results show that the proposed approach achieves an average detection precision of 99.57%, whereas the country, language, and layout classification accuracy is 99.33%.
Character recognition is an important research field of interest for many applications. In recent years, deep learning has made breakthroughs in image classification, especially for character recognition. However, convolutional neural networks (CNN) still deliver state-of-the-art results in this area. Motivated by the success of CNNs, this paper proposes a simple novel full depth stacked CNN architecture for Latin and Arabic handwritten alphanumeric characters that is also utilized for license plate (LP) characters recognition. The proposed architecture is constructed by four convolutional layers, two max-pooling layers, and one fully connected layer. This architecture is low-complex, fast, reliable and achieves very promising classification accuracy that may move the field forward in terms of low complexity, high accuracy and full feature extraction. The proposed approach is tested on four benchmarks for handwritten character datasets, Fashion-MNIST dataset, public LP character datasets and a newly introduced real LP isolated character dataset. The proposed approach tests report an error of only 0.28% for MNIST, 0.34% for MAHDB, 1.45% for AHCD, 3.81% for AIA9K, 5.00% for Fashion-MNIST, 0.26% for Saudi license plate character and 0.97% for Latin license plate characters datasets. The license plate characters include license plates from Turkey (TR), Europe (EU), USA, United Arab Emirates (UAE) and Kingdom of Saudi Arabia (KSA).
Fading in a wireless channel has negative effects on the performance of communication systems. Bell Laboratories layered space-time (BLAST) has been used to get a high data rate while space-time block codes (STBC) have been used to get a low bit error rate (BER) performance. Under deep faded channels, hybrid BLAST-STBC systems are considered as a trade-off between BLAST and STBC systems. By exploiting the benefits of both systems, a new method to represent a 4 × 4 multiple-input and multiple-output (MIMO) system is proposed and studied, in which the transmission process is carried out adaptively between both 4 × 4 VBLAST, Quasi-Orthogonal STBC (QOSTBC) and Hybrid systems according to the transmit links state. The proposed adaptive switching hybrid system (ASHS) reduces the total transmitted power, achieves the maximum throughput by obtaining the best BER. An adaptive switching transmission scheme using the strategy of measuring the transmit links fading is investigated as well. The simulated results are obtained in an environment of a 4 × 4 MIMO system using MATLAB platform where the total transmitting power is normalized to unity. The detections are done using the maximum likelihood (ML) receiver. The proposed ASHS system shows a lot of advantages such as maximum throughput is obtained in bad channel states, no additional transmit power is needed and no additional bandwidth is needed. Finally, under deep fading condition, the proposed ASHS transmission scheme obtains the best BER, reduces wasting the total transmitted power, achieves the maximum throughput and obtains the best BER.
The hottest issue of next generation communication systems is data throughput improvement for any wireless channel conditions. Multi-Input Multi-Output (MIMO) systems are the key technology for the next generation communication systems. Bill Lab Layered Space Time (BLAST) system achieves high data rates with acceptable BER performance over a good channels state. However, when the wireless channel has a considerable fading then the performance will be decreased and at a deep-fading state the system may fail to transmit any signal. This paper studies and compares the conventional 4 × 4 Vertical BLAST system capacity and bit error rate (BER) performance at Maximum Likelihood (ML) receiver through a simulation. It also studies the effect of a transmit-link deep-fading on the effective signal to noise ratio, system BER and system capacity in 4 × 4 VBLAST system in a comparative way with the conventional 4× 4, 3× 4, 2× 4 and 1× 4 VBLAST system. Considering that every possible case of transmit link deep fade lower than -20dB fading gain is equivalent to switching off this transmit link and mostly all of its power will be wasted. MATLAB software has been used as the main platform for system simulation. تسعى هذه الدراسة للكشف عن اهمية تحسين معدل نقل البيانات في انظمة الاتصالات بغض النظر عن ظروف القناة اللاسلكية، لان انظمة الاتصالات ذات المداخل والمخارج المتعددة تعتبر التكنولوجيا الاساسية في انظمة الاتصال في الاجيال القادمة من هذه التكنولوجيا، ولما كان نظام بيل المتعدد الطبقات الزمكاني (البلاست) يحقق معدل بيانات عالياً – مع نسبة خطأ مقبولة – حال كون القناة اللاسلكية في حالة جيدة، فإن هذا لا يمنع تراجع الاداء إذا عانت القناة اللاسلكية من تخميد وإن كان التخميد عميقاً قد يتعرض النظام للفشل في ارسال البيانات بصورة صحيحة. ومن هنا فإن الدراسة تسعى لفحص ومقارنة سعة ومعدل الخطأ في نظام 4 × 4بلاست العمودي حال استخدام نظام الأفضلية المثلى في عملية الاستقبال. كما تسعى ايضا لدراسة تأثير وجود تخميد عميق في أحد قنوات الارسال الاربعة على معامل نسبة الاشارة الي الضوضاء، وكذلك معدل الخطأ ومتوسط سعة القناة، اخذةً بعين الاعتبار أن معامل التخميد إذا قل عن -20 ديسبل فانه يكافئ حالة قطع الارسال عبر القناة، وأن طاقة الارسال المستخدمة في هذه الحالة ستضيع هباءاً. وقد استخدم الباحث برنامج الماتلاب كاساس للمحاكاة في هذه الدراسة.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.