.png){kind=icon}
返回
Automated subway touch button detection using image process
.png){kind=icon}
OA
AI
DOI：10.1186/s40537-024-00941-6.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Subway button detection is paramount for passenger safety, yet the occurrence of inadvertent touches poses operational threats. Camera-based detection is indispensable for identifying touch occurrences, ascertaining person identity, and implementing scientific measures. Existing methods suffer from inaccuracies due to the small size of buttons, complex environments, and challenges such as occlusion. We present YOLOv8-DETR-P2-DCNv2-Dynamic-NWD-DA, which enhances occlusion awareness, reduces redundant annotations, and improves contextual feature extraction. The model integrates the RTDETRDecoder, P2 small target detection layer, DCNv2-Dynamic algorithm, and the NWD loss function for multiscale feature extraction. Dataset augmentation and the GAN algorithm refine the model, aligning feature distributions and enhancing precision by 6.5%, 5%, and 5.8% in precision, recall, and mAP50, respectively. These advancements denote significant improvements in key performance indicators.
Keyword:
Touch button detection
Camera
YOLOV8
RTDETRDecoder
P2
DCNv2-Dynamic
Small target
NWD
GAN
Multiscale
Indicators
期刊
IF:
6.4
论文数:
1.5K
被引数:
1.3W
机构
引用论文
A new series of vitamin D analogs is highly active for clonal inhibition, differentiation, and induction of WAF1 in myeloid leukemia.png)
BloodOAAI
BloodIF0
OAAI分享收藏Sociocultural Context, Entrepreneur Types, Mindsets and Entrepreneurial Action in Tiradentes, Brazil
PREAI
分享收藏
An effective railway intrusion detection method using dynamic intrusion region and lightweight neural network基于动态入侵区域和轻量级神经网络的铁路入侵检测方法MEASUREMENT
MEASUREMENTIF5.6
PREAI
分享收藏
查看更多内容