2022
DOI: 10.1109/tits.2021.3063521
|View full text |Cite
|
Sign up to set email alerts
|

Distracted Driver Detection Based on a CNN With Decreasing Filter Size

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
41
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 47 publications
(43 citation statements)
references
References 33 publications
2
41
0
Order By: Relevance
“…Some studies have attempted to reduce the computation time. [19] proposed a new CNN-based driver activity recognition model by decreasing the filter size to reduce the size of the model. [21] proposed a lightweight CNN model with an octave-like convolution mixed block that uses pointwise convolution to expand the feature maps into two sets of branches.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Some studies have attempted to reduce the computation time. [19] proposed a new CNN-based driver activity recognition model by decreasing the filter size to reduce the size of the model. [21] proposed a lightweight CNN model with an octave-like convolution mixed block that uses pointwise convolution to expand the feature maps into two sets of branches.…”
Section: Related Workmentioning
confidence: 99%
“…Driver anomaly quantification, as a typical task of a DMS, has been studied for a long time [9][10][11][12][13][14][15]. Recently, many researchers have revisited this topic by leveraging the powerful representation capabilities of deep learning, leading to impressive achievements [16][17][18][19][20][21][22][23][24]. However, these studies usually treat it as a classification task, in which driver activity is classified into several predefined classes.…”
Section: Introductionmentioning
confidence: 99%
“…For example, Baheti et al [20] propose the MobileVGG, which reduces the number of parameters by replacing the traditional convolution in the classical VGG structure with depth-wise convolution and point-wise convolution. D-HCNN [21] is another example, which uses an architecture containing four convolution blocks with the filters of rather large spatial sizes and achieves high performance with small number of filters. However, these networks were designed entirely by hand based on experience with networks used for general-purpose computer vision tasks, so the potential of the network structure could not be reached to the maximum extent possible.…”
Section: A Existing Distracted Driver Recognition Approachesmentioning
confidence: 99%
“…To carry out engineering applications, some researchers carried out lightweight designs on the CNN model. For example, Qin et al [77] constructed a CNN lightweight model for the identification of distracted driving, and the recognition accuracy reached 95.59%. At present, CNN tends to use a smaller convolutional kernel, deeper structure, and fewer aggregation layers and gradually develop into fully convolutional network.…”
Section: Convolutional Neural Networkmentioning
confidence: 99%