Disentangled Generation Network for Enlarged License Plate Recognition and A Unified Dataset

TL;DR

This paper introduces a new dataset and a disentangled generation network to improve enlarged license plate recognition under challenging real-world conditions, addressing diversity and noise issues.

Contribution

The work presents a novel dataset of enlarged license plates and proposes a task-level disentanglement framework for robust recognition, which is a new approach in this domain.

Findings

The proposed method improves recognition accuracy across different frameworks.

The dataset captures diverse real-world challenges for license plate recognition.

Disentangled generation enhances the robustness of recognition models.

Abstract

License plate recognition plays a critical role in many practical applications, but license plates of large vehicles are difficult to be recognized due to the factors of low resolution, contamination, low illumination, and occlusion, to name a few. To overcome the above factors, the transportation management department generally introduces the enlarged license plate behind the rear of a vehicle. However, enlarged license plates have high diversity as they are non-standard in position, size, and style. Furthermore, the background regions contain a variety of noisy information which greatly disturbs the recognition of license plate characters. Existing works have not studied this challenging problem. In this work, we first address the enlarged license plate recognition problem and contribute a dataset containing 9342 images, which cover most of the challenges of real scenes. However, the created data are still insufficient to train deep methods of enlarged license plate recognition, and building large-scale training data is very time-consuming and high labor cost. To handle this problem, we propose a novel task-level disentanglement generation framework based on the Disentangled Generation Network (DGNet), which disentangles the generation into the text generation and background generation in an end-to-end manner to effectively ensure diversity and integrity, for robust enlarged license plate recognition. Extensive experiments on the created dataset are conducted, and we demonstrate the effectiveness of the proposed approach in three representative text recognition frameworks.