MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition
Object detection and recognition of road scenes are crucial tasks of the autonomous driving environmental perception system. The low inference speed and accuracy in object detection models hinder the development of autonomous driving technology. Searching for improvement of detection accuracy and sp...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10058516/ |
_version_ | 1811155598870315008 |
---|---|
author | Yining Cao Chao Li Yakun Peng Huiying Ru |
author_facet | Yining Cao Chao Li Yakun Peng Huiying Ru |
author_sort | Yining Cao |
collection | DOAJ |
description | Object detection and recognition of road scenes are crucial tasks of the autonomous driving environmental perception system. The low inference speed and accuracy in object detection models hinder the development of autonomous driving technology. Searching for improvement of detection accuracy and speed is still a challenging task. For solving these problems, we proposed an MCS-YOLO algorithm. Firstly, a coordinate attention module is inserted into the backbone to aggregate the feature map’s spatial coordinate and cross-channel information. Then, we designed a multiscale small object detection structure to improve the recognition sensitivity of dense small object. Finally, we applied the Swin Transformer structure to the CNN to enable the network to focus on contextual spatial information. Conducting ablation study on the autonomous driving dataset BDD100K, MCS-YOLO algorithm achieves a mean average precision of 53.6% and a recall rate of 48.3%, which are 4.3% and 3.9% better than the YOLOv5s algorithm respectively. In addition, it can achieve real-time detection speed of 55 frames per second in a real scene. The results show that the MCS-YOLO algorithm is effective and superior in the task of automatic driving object detection. |
first_indexed | 2024-04-10T04:36:16Z |
format | Article |
id | doaj.art-3d613b3b151e4d20908c18b14829aa7f |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-10T04:36:16Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-3d613b3b151e4d20908c18b14829aa7f2023-03-10T00:00:18ZengIEEEIEEE Access2169-35362023-01-0111223422235410.1109/ACCESS.2023.325202110058516MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment RecognitionYining Cao0https://orcid.org/0000-0003-1444-1387Chao Li1https://orcid.org/0000-0003-3067-0630Yakun Peng2https://orcid.org/0000-0001-6440-7342Huiying Ru3https://orcid.org/0000-0002-6875-4525College of Information Engineering, Hebei University of Architecture, Zhangjiakou, ChinaCollege of Information Engineering, Hebei University of Architecture, Zhangjiakou, ChinaCollege of Information Engineering, Hebei University of Architecture, Zhangjiakou, ChinaCollege of Science, Hebei University of Architecture, Zhangjiakou, ChinaObject detection and recognition of road scenes are crucial tasks of the autonomous driving environmental perception system. The low inference speed and accuracy in object detection models hinder the development of autonomous driving technology. Searching for improvement of detection accuracy and speed is still a challenging task. For solving these problems, we proposed an MCS-YOLO algorithm. Firstly, a coordinate attention module is inserted into the backbone to aggregate the feature map’s spatial coordinate and cross-channel information. Then, we designed a multiscale small object detection structure to improve the recognition sensitivity of dense small object. Finally, we applied the Swin Transformer structure to the CNN to enable the network to focus on contextual spatial information. Conducting ablation study on the autonomous driving dataset BDD100K, MCS-YOLO algorithm achieves a mean average precision of 53.6% and a recall rate of 48.3%, which are 4.3% and 3.9% better than the YOLOv5s algorithm respectively. In addition, it can achieve real-time detection speed of 55 frames per second in a real scene. The results show that the MCS-YOLO algorithm is effective and superior in the task of automatic driving object detection.https://ieeexplore.ieee.org/document/10058516/Coordinate attention mechanismsautonomous drivingroad environmental object detectionswin transformerYOLOv5 |
spellingShingle | Yining Cao Chao Li Yakun Peng Huiying Ru MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition IEEE Access Coordinate attention mechanisms autonomous driving road environmental object detection swin transformer YOLOv5 |
title | MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition |
title_full | MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition |
title_fullStr | MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition |
title_full_unstemmed | MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition |
title_short | MCS-YOLO: A Multiscale Object Detection Method for Autonomous Driving Road Environment Recognition |
title_sort | mcs yolo a multiscale object detection method for autonomous driving road environment recognition |
topic | Coordinate attention mechanisms autonomous driving road environmental object detection swin transformer YOLOv5 |
url | https://ieeexplore.ieee.org/document/10058516/ |
work_keys_str_mv | AT yiningcao mcsyoloamultiscaleobjectdetectionmethodforautonomousdrivingroadenvironmentrecognition AT chaoli mcsyoloamultiscaleobjectdetectionmethodforautonomousdrivingroadenvironmentrecognition AT yakunpeng mcsyoloamultiscaleobjectdetectionmethodforautonomousdrivingroadenvironmentrecognition AT huiyingru mcsyoloamultiscaleobjectdetectionmethodforautonomousdrivingroadenvironmentrecognition |