Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (11): 3228-3233.DOI: 10.11772/j.issn.1001-9081.2021010073

• Artificial intelligence • Previous Articles     Next Articles

Road abandoned object detection algorithm based on optimized instance segmentation model

Yue ZHANG1, Liang ZHANG1,2(), Fei XIE1,2, Jiale YANG1, Rui ZHANG1, Yijian LIU1,2   

  1. 1.School of Electrical and Automation Engineering,Nanjing Normal University,Nanjing Jiangsu,210023,China
    2.Nanjing Industry Institute for Advanced Intelligent Equipment,Nanjing Jiangsu,210042,China
  • Received:2021-01-14 Revised:2021-03-25 Accepted:2021-04-06 Online:2021-04-26 Published:2021-11-10
  • Contact: Liang ZHANG
  • About author:ZHANG Yue, born in 1995, M. S. candidate. Her research interests include deep learning, computer vision, target detection, instance segmentation
    ZHANG Liang,born in 1974,M. S.,lecturer. His research interests include image processing,robot control
    XIE Fei,born in 1983,Ph. D.,associate professor. His research interests include machine vision, deep learning, intelligent transportation,navigation and positioning
    YANG Jiale,born in 1996,M. S. candidate. Her research interests include deep learning,computer vision
    ZHANG Rui, born in 1997, M. S. candidate. His research interests include deep learning,computer vision
    LIU Yijian, born in 1977, Ph. D., associate professor. His research interests include robotics,embedded system.

基于实例分割模型优化的道路抛洒物检测算法

章悦1, 张亮1,2(), 谢非1,2, 杨嘉乐1, 张瑞1, 刘益剑1,2   

  1. 1.南京师范大学 电气与自动化工程学院,南京 210023
    2.南京智能高端装备产业研究院,南京 210042
  • 通讯作者: 张亮
  • 作者简介:章悦(1995—),女,江苏南京人,硕士研究生,CCF 会员,主要研究方向:深度学习、计算机视觉、目标检测、实例分割
    张亮(1974—),男,江苏扬州人,讲师,硕士,主要研究方向:图像处理、机器人控制
    谢非(1983—),男,江苏徐州人,副教授,博士,主要研究方 向:机器视觉、深度学习、智能交通、导航定位
    杨嘉乐(1996—),女,江苏泰州人,硕士研究生,主要研究方向:深度学习、计算机视觉
    张瑞(1997—),男,山东泰安人,硕士研究生,主要研究方向:深度学习、计算机视觉
    刘益剑(1977—),男,江苏淮安人,副教授,博士,主要研究 方向:机器人技术、嵌入式系统。

Abstract:

In the field of traffic safety, the road abandoned objects easily cause traffic accidents and become potential traffic safety hazards. Focusing on the problems of low recognition rate and poor detection effect for different abandoned objects of traditional road abandoned object detection methods, a road abandoned object detection algorithm based on the optimized instance segmentation model CenterMask was proposed. Firstly, the residual network ResNet50 optimized by dilated convolution was used as the backbone neural network to extract image features and carry out the multi-scale processing. Then, the Fully Convolutional One-Stage (FCOS) target detector optimized by Distance Intersection over Union (DIoU) function was used to realize the detection and classification of road abandoned objects. Finally, the spatial attention-guided mask was used as the mask segmentation branch to realize the object shape segmentation, and the model training was realized by the transfer learning method. Experimental results show that, the detection rate of the proposed algorithm for road abandoned objects is 94.82%, and compared with the common instance segmentation algorithm Mask Region-Convolutional Neural Network (Mask R-CNN), the proposed road abandoned object detection algorithm has the Average Precision (AP) increased by 8.10 percentage points in bounding box detection.

Key words: instance segmentation, road abandoned object, dilated convolution, Distance Intersection over Union (DIoU) function, deep learning

摘要:

在交通安全领域,道路抛洒物易引发交通事故,构成了交通安全隐患。针对传统抛洒物检测方式识别率低、对于多类抛洒物检测效果不佳等问题,提出了一种基于实例分割模型CenterMask优化的道路抛洒物检测算法。首先,使用空洞卷积优化的残差网络ResNet50作为主干神经网络来提取特征并进行多尺度处理;然后,通过距离交并比(DIoU)函数优化的全卷积单阶段(FCOS)目标检测器实现对抛洒物的检测和分类;最后,使用空间注意力引导掩膜作为掩膜分割分支来实现对于目标形态的分割,并采用迁移学习的方式实现模型的训练。实验结果表明,所提算法对于抛洒物目标的检测率为94.82%,相较常见实例分割算法Mask R-CNN,所提的道路抛洒物检测算法在边界框检测上的平均精度(AP)提高了8.10个百分点。

关键词: 实例分割, 道路抛洒物, 空洞卷积, 距离交并比函数, 深度学习

CLC Number: