2022
DOI: 10.3390/s22093349
|View full text |Cite
|
Sign up to set email alerts
|

MobileYOLO: Real-Time Object Detection Algorithm in Autonomous Driving Scenarios

Abstract: Object detection is one of the key tasks in an automatic driving system. Aiming to solve the problem of object detection, which cannot meet the detection speed and detection accuracy at the same time, a real-time object detection algorithm (MobileYOLO) is proposed based on YOLOv4. Firstly, the feature extraction network is replaced by introducing the MobileNetv2 network to reduce the number of model parameters; then, part of the standard convolution is replaced by depthwise separable convolution in PAnet and t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
10
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 20 publications
(10 citation statements)
references
References 26 publications
0
10
0
Order By: Relevance
“…First, such an architecture is lightweight and can be deployed in field applications using mobile devices. 49,50 Recent comparative reviews agree that YOLOv2 gives the best compromise between the processing time and the detection accuracy for real-time drone detection. 33 Successful applications of MobileNet-v2 and YOLOv2 have been presented in literature.…”
Section: The Mobilenet-v2 and Yolov2 Object Detection Networkmentioning
confidence: 99%
“…First, such an architecture is lightweight and can be deployed in field applications using mobile devices. 49,50 Recent comparative reviews agree that YOLOv2 gives the best compromise between the processing time and the detection accuracy for real-time drone detection. 33 Successful applications of MobileNet-v2 and YOLOv2 have been presented in literature.…”
Section: The Mobilenet-v2 and Yolov2 Object Detection Networkmentioning
confidence: 99%
“…A series of models have emerged, such as SqueezeNet, ShuffleNet [ 17 ], Xception [ 18 ], and MobileNet [ 19 , 20 ], which make it possible for embedded and other edge devices to run deep learning models directly. Zhou et al [ 5 ] proposed the MobileYOLO model, which used the mobilenet to effectively reduce model parameters. Zhao et al [ 21 ] proposed an improved YOLO model to adapt to real-time detection in a special natural geographical environment.…”
Section: Related Workmentioning
confidence: 99%
“…At present, the remarkable progress of computer vision and computation tools have provided theoretical and technical support for autonomous environmental perception [ 2 ]. Therefore, vision-based object detection is a key means of autonomous environmental perception [ 3 , 4 ], and its detection performances, such as accuracy, speed, efficiency, and robustness, are very important for the detection of vehicles, pedestrians, and obstacles in traffic scenes [ 5 , 6 ].…”
Section: Introductionmentioning
confidence: 99%
“…Chen et al [23] proposed a DW-YOLO algorithm that improves vehicle object detection performance by increasing the depth and width of the network. Zhou et al [24] proposed a lightweight MobileYOLO algorithm that reduces the number of parameters and improves detection speed. Wang et al [25] applied MobileNet to a YOLOv4 network for driving scenarios and achieved a detection speed of 35 FPS.…”
Section: Introductionmentioning
confidence: 99%