Otwarty dostęp

A Baseline for Violence Behavior Detection in Complex Surveillance Scenarios

Yingying Long

Yingying Long

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Wyszukaj tego autora
Sciendo|Google Scholar
Long, Yingying
, 
Zongxin Wang

Zongxin Wang

School of Computer Science and Engineering Xi’an Technological UniversityXi’an, China
Wyszukaj tego autora
Sciendo|Google Scholar
Wang, Zongxin
, 
Hanzhu Wei

Hanzhu Wei

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Wyszukaj tego autora
Sciendo|Google Scholar
Wei, Hanzhu
 oraz 
Xiaojun Bai

Xiaojun Bai

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Wyszukaj tego autora
Sciendo|Google Scholar
Bai, Xiaojun
  
31 gru 2024
A Baseline for Violence Behavior Detection in Complex Surveillance Scenarios's Cover Image
O artykule
Poprzedni artykuł
Następny artykuł
Zacytuj
Pobierz okładkę
Data publikacji: 31 gru 2024
Zakres stron: 48 - 58
DOI: https://doi.org/10.2478/ijanmc-2024-0036
Słowa kluczowe
© 2024 Yingying Long et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Illustration of a sample of labeled acts of violence
Illustration of a sample of labeled acts of violence

Figure 2.

The violence detection algorithm's framework is displayed in Fig.2. The model for extracting spatio-temporal features and the spatio-temporal feature fusion module make up the majority of the framework. The spatio-temporal feature extraction model is composed of the temporal feature extraction model and the spatial feature extraction module, and the I3D network is the network structure of the temporal feature extraction model, as illustrated in (a); (b)(c) are the 3D-CBAM Attention Mechanism and 3D Inception (3D Inc) module, respectively. The Atrous Spatial Pyramid Pooling (ASPP) model is added at the end of the spatial feature extraction model, which has the CSPDarkNet-Tiny network as its network structure, which is shown in (d), where rate denotes the expansion rate of the null convolution. atrous Spatial Pyramid Pooling (ASPP) has five branches, including one ordinary convolutional branch, three null convolutional branches, and one global average pooling branch; (e) shows the overall structure of Channel Fusion and Attention Mechanism(CFAM); D is the final output feature map of CFAM, and C1 and C2 are the number of feature map output channels for the I3D network and the ASP module, respectively.
The violence detection algorithm's framework is displayed in Fig.2. The model for extracting spatio-temporal features and the spatio-temporal feature fusion module make up the majority of the framework. The spatio-temporal feature extraction model is composed of the temporal feature extraction model and the spatial feature extraction module, and the I3D network is the network structure of the temporal feature extraction model, as illustrated in (a); (b)(c) are the 3D-CBAM Attention Mechanism and 3D Inception (3D Inc) module, respectively. The Atrous Spatial Pyramid Pooling (ASPP) model is added at the end of the spatial feature extraction model, which has the CSPDarkNet-Tiny network as its network structure, which is shown in (d), where rate denotes the expansion rate of the null convolution. atrous Spatial Pyramid Pooling (ASPP) has five branches, including one ordinary convolutional branch, three null convolutional branches, and one global average pooling branch; (e) shows the overall structure of Channel Fusion and Attention Mechanism(CFAM); D is the final output feature map of CFAM, and C1 and C2 are the number of feature map output channels for the I3D network and the ASP module, respectively.

Figure 3.

CSPDarkNet-Tiny Network Overall Structure
CSPDarkNet-Tiny Network Overall Structure

Figure 4.

Violence detection results
Violence detection results

Detection results of 3D-CBAM attention model embedded at different locations

Network Embedding position UCF101-24 JHMDB VioData
mAP
- 84.4% 80.4% 86.5%
3D Inc_1 86.1% 83.7% 89.0%
3D Inc_2 86.7% 83.3% 88.3%
3D Inc_3 85.9% 84.2% 89.6%
3D Inc_1+3D Inc_2 88.2% 87.5% 90.7%
I3D 3D Inc_1+3D Inc_3 89.8% 88.6% 91.8%
3D Inc_2+3D Inc_3 88.0% 88.0% 91.4%
3D Inc_1+3D Inc_2+3D Inc_3 90.0% 88.7% 92.0%

Parameter settings in network training

Parameter Setting
Initial Learning Rate 0.001
Epoch 230
ReSize (416,416)
ReSize (416,416)
Weight Decay 0.0005
Optimizer Adam

Results of violence detection accuracy of different models

Method UCF101-24 JHMDB VioData
mAP
MPS 82.4% - 85.3
P3D-CTN - 84.0% 84.9%
STEP 83.1% - 86.4%
YOWO 82.5% 85.7% 88.0%
ours 89.8% 88.6% 91.8%

Detection results with embedded ASPP model and introduction of spatio-temporal depth separable convolution

Network UCF101-24 JHMDB VioData
mAP
Baseline 78.5% 75.3% 78.9%
CSPDarkNet-Tiny+ASPP 80.7% 76.6% 82.0%
CSPDarkNet-Tiny+ASPP++I3D(Impr oved 3D Inc) 84.8% 80.4% 86.5%
Język:
Angielski
Częstotliwość wydawania:
4 razy w roku
Dziedziny czasopisma:
Informatyka, Informatyka, inne
Kanał RSS czasopisma