Salemdeeb Mohammed, Ertürk Sarp
Department of Electrical-Electronics Engineering, Bartin University, Bartin, Turkey.
Department of Electronics & Communication Eng., Kocaeli University, Izmit, Kocaeli, Turkey.
PeerJ Comput Sci. 2021 Jun 18;7:e576. doi: 10.7717/peerj-cs.576. eCollection 2021.
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).
字符识别是许多应用中一个重要的研究领域。近年来，深度学习在图像分类方面取得了突破，特别是在字符识别方面。然而，卷积神经网络（CNN）在这一领域仍然取得了最先进的成果。受CNN成功的启发，本文提出了一种简单新颖的全深度堆叠CNN架构，用于拉丁和阿拉伯手写字母数字字符，该架构也用于车牌（LP）字符识别。所提出的架构由四个卷积层、两个最大池化层和一个全连接层组成。这种架构复杂度低、速度快、可靠性高，并且实现了非常有前景的分类准确率，可能会在低复杂度、高精度和全特征提取方面推动该领域向前发展。所提出的方法在四个手写字符数据集基准、Fashion-MNIST数据集、公共LP字符数据集和一个新引入的真实LP孤立字符数据集上进行了测试。所提出的方法测试报告显示，MNIST的错误率仅为0.28%，MAHDB为0.34%，AHCD为1.45%，AIA9K为3.81%，Fashion-MNIST为5.00%，沙特车牌字符为0.26%，拉丁车牌字符数据集为0.97%。车牌字符包括来自土耳其（TR）、欧洲（EU）、美国、阿拉伯联合酋长国（UAE）和沙特阿拉伯王国（KSA）的车牌。
