计算机与现代化 ›› 2023, Vol. 0 ›› Issue (03): 23-28.

• 图像处理 • 上一篇    下一篇

基于图像增强和SKNet的交通标志识别

  

  1. (长安大学信息工程学院,陕西 西安 710064)
  • 出版日期:2023-04-17 发布日期:2023-04-17
  • 作者简介:廖聪(1998—),男,陕西渭南人,硕士研究生,研究方向:交通图像处理,深度学习,E-mail: congliao@chd.edu.cn; 郭凰(1998—),女,陕西宝鸡人,硕士研究生,研究方向:交通数据处理,机器学习,E-mail: hguo_mail@126.com; 赵茂军(1998—),男,山东济宁人,硕士研究生,研究方向:图像处理,深度学习,E-mail: mjzhao@chd.edu.cn; 王雨松(1998—),男,陕西西安人,硕士研究生,研究方向:交通流预测,E-mail: 2020124022@chd.edu.cn; 白俊峰(1997—),男,河南洛阳人,硕士研究生,研究方向:探地雷达图像处理,深度学习,E-mail: 2020224097@chd.edu.cn。
  • 基金资助:
    国家重点研发计划项目(2018YFC0808706); 国家自然科学基金资助项目(62001059); 陕西省重点研发计划项目(2021GY-019); 陕西省自然科学基金面上项目(2022JM-056)

Traffic Sign Recognition Based on Image Enhancement and SKNet

  1. (School of Information Engineering, Chang’an University, Xi’an 710064, China)
  • Online:2023-04-17 Published:2023-04-17

摘要: 针对现有交通标志识别系统对图像特征提取不充分和复杂情况下难以识别的问题,设计基于图像增强和SKNet的交通标志识别模型HE-SKNet。首先,采用直方图均衡化,对过亮或过暗的交通标志图像进行增强;然后使用自适应调节感受野大小的SKNet网络进行特征提取和分类。GTSRB数据集的实验结果表明,提出的HE-SKNet模型识别准确率达到了98.95%,相比ResNet、ResNeXt、SENet和SKNet准确率平均提高了2.77个百分点,验证了HE-SKNet模型自适应提取不同尺度特征的能力,更适用于过亮或过暗的复杂实际应用场景。

关键词: 图像增强, 直方图均衡化, SKNet, 深度学习, 交通标志识别

Abstract: Aiming at the problems that the existing traffic sign recognition systems extract image feature insufficiently and are difficult to recognize under complex situations, a traffic sign recognition model HE-SKNet is designed based on image enhancement and SKNet. Firstly, the histogram equalization is used to enhance the images of traffic signs with too bright or too dark. Then the SKNet network that adaptively adjusts the size of the receptive field is used for feature extraction and classification. The experimental results on the GTSRB dataset show that the recognition accuracy of the proposed HE-SKNet model reaches 98.95%, which enjoys 2.77 percentage points higher than that of ResNet, ResNeXt, SENet and SKNet on average. It verifies that the HE-SKNet model adaptively extracts different scales of feature and is more suitable for complex practical application scenarios with too bright or too dark.

Key words: image enhancement, histogram equalization, SKNet, deep learning, traffic sign recognition