| 引用本文: | 邹北骥,雷太航,刘姝,廖望旻,姜灵子.自然场景车标数据集的构建及其应用[J].国防科技大学学报,2021,43(1):95-102.[点击复制] |
| ZOU Beiji,LEI Taihang,LIU Shu,LIAO Wangmin,JIANG Lingzi.Construction of vehicle logo dataset in natural scenes and its application[J].Journal of National University of Defense Technology,2021,43(1):95-102[点击复制] |
|
| |
|
|
| 本文已被：浏览 242次 下载 212次 |
| 自然场景车标数据集的构建及其应用 |
| 邹北骥1,2,雷太航1,2,刘姝1,2,廖望旻1,2,姜灵子1,2 |
|
(1. 中南大学 计算机学院, 湖南 长沙 410083;2. 中南大学 湖南省机器视觉与智慧医疗工程技术研究中心, 湖南 长沙 410083)
|
| 摘要: |
| 车标作为车辆身份的关键特征之一,在车辆的监控与辨识中发挥着重要作用。由于自然场景复杂多变,对其中的车标进行准确识别仍具有很大的挑战性。目前公开数据库很少且存在诸多局限,导致研究缺乏可信度和实用性。本文建立了一个面向自然场景的全新数据集,包含多种采集环境下的10 324幅、67类车辆图像。基于此数据集开展应用研究,提出一个目标检测与深度学习相结合的车标识别方法,包括车标区域定位和车标种类预测两大步骤。实验表明,该方法对复杂背景有较强的适应性,在涉及30种车标的分类任务中达到89.0%的总体识别率。 |
| 关键词: 车标识别 自然场景 目标检测 深度学习 |
| DOI：10.11887/j.cn.202101013 |
| 投稿日期：2019-10-16 |
| 基金项目:国家自然科学基金资助项目(61902435)；湖南省科技计划资助项目(2017WK2074)；湖南省自然科学基金资助项目(2019JJ50808) |
|
| Construction of vehicle logo dataset in natural scenes and its application |
| ZOU Beiji1,2, LEI Taihang1,2, LIU Shu1,2, LIAO Wangmin1,2, JIANG Lingzi1,2 |
|
(1. School of Computer Science and Engineering, Central South University, Changsha 410083, China;2. Hunan Engineering Research Center of Machine Vision and Intelligent Medicine, Central South University, Changsha 410083, China)
|
| Abstract: |
| As one of the key characteristics of vehicle identity, vehicle logo plays an important role in vehicle monitoring and identification. Due to the complexity of the natural scene, it is still a great challenge to identifying the vehicle logo accurately. At present, there are few open databases and there are many limitations, which lead to the lack of credibility and practicability. In this paper, a new dataset for natural scenes which contains 10 324 images with 67 types of vehicle logos in various acquisition environments was established. Based on this dataset, a vehicle logo recognition method based on target detection and deep learning was proposed. The method includes two major steps:regional positioning of vehicle logo and prediction of vehicle logo type. Experiments show that the proposed method has strong adaptability to complex background, and the overall recognition rate reaches 89.0% in the classification task involving 30 kinds of vehicle logos. |
| Key words: vehicle logo recognition natural scene object detection deep learning |
|
|
|
|
|
