CNN-RNN based method for license plate recognition

CAAI Transactions on Intelligence Technology

article

Palaiahnakote Shivakumara¹  Dongqi Tang²  Maryam Asadzadehkaljahi¹  Tong Lu²  Umapada Pal³  Mohammad Hossein Anisi⁴ 

[1] Faculty of Computer Science and Information Technology, University of Malaya;National Key Laboratory for Novel Software Technology, Nanjing University;Computer Vision and Pattern Recognition Unit, Indian Statistical Institute;School of Computer Science and Electronic Engineering, University of Essex

关键词: image recognition;    learning (artificial intelligence);    image segmentation;    recurrent neural nets;    feature extraction;    image classification;    image colour analysis;    MIMOS;    recognition performance;    CNN-RNN based method;    license plate recognition;    multiple adverse factors;    foreground colour;    cursive handwriting;    vehicle movements;    Malaysia;    dark background;    public vehicles;    white background;    license plate images;    convolutional neural networks;    recurrent neural networks;    BLSTM;    bi-directional long-short term memory;    feature extraction;    high discriminative ability;    context information;    dense cluster-based voting;    classification methods;    Malaysian government;    tandard dataset UCSD;    (B6135E) Image recognition;    (C5260B) Computer vision and image processing techniques;    (C5290) Neural computing techniques;   

DOI  :  10.1049/trit.2018.1015

学科分类：数学（综合）

来源: Wiley

【 摘 要 】

Achieving good recognition results for License plates is challenging due to multiple adverse factors. For instance, in Malaysia, where private vehicle (e.g., cars) have numbers with dark background, while public vehicle (taxis/cabs) have numbers with white background. To reduce the complexity of the problem, we propose to classify the above two types of images such that one can choose an appropriate method to achieve better results. Therefore, in this work, we explore the combination of Convolutional Neural Networks (CNN) and Recurrent Neural Networks namely, BLSTM (Bi-Directional Long Short Term Memory), for recognition. The CNN has been used for feature extraction as it has high discriminative ability, at the same time, BLSTM has the ability to extract context information based on the past information. For classification, we propose Dense Cluster based Voting (DCV), which separates foreground and background for successful classification of private and public. Experimental results on live data given by MIMOS, which is funded by Malaysian Government and the standard dataset UCSD show that the proposed classification outperforms the existing methods. In addition, the recognition results show that the recognition performance improves significantly after classification compared to before classification.

【 授权许可】

CC BY|CC BY-ND|CC BY-NC|CC BY-NC-ND