Abstract
Nowadays, pedestrian detection is widely used in fields such as driving assistance and video surveillance with the progression of technology. However, although the research of single-modal visible pedestrian detection has been very mature, it is still not enough to meet the demand of pedestrian detection at all times. Thus, a multi-spectral pedestrian detection method via image fusion and convolutional neural networks is proposed in this paper. The infrared intensity distribution and visible appearance features are retained with a total variation model based on local structure transfer, and pedestrian detection is realized with the multi-spectral fusion results and the target detection network YOLOv3. The detection performance of the proposed method is evaluated and compared with the detection methods based on the other four pixel-level fusion algorithms and two fusion network architectures. The results attest that our method has superior detection performance, which can detect pedestrian targets robustly even in the case of harsh illumination conditions and cluttered backgrounds.
Highlights
As an important task in target detection, pedestrian detection is widely used in traffic safety, video surveillance, human–computer interaction, and other fields [1,2,3]
It is worth noting that, as shown in the fourth column in Figure 9, the noise of theto visible image at night would be enhanced to varying degrees in the several fusion deIn order explore a better multi-spectral pedestrian detection method, results
STF was the most affected by noise because it retained a large amount of visible tectors based on different fusion methods were studied and tested in this paper
Summary
As an important task in target detection, pedestrian detection is widely used in traffic safety, video surveillance, human–computer interaction, and other fields [1,2,3]. Region-Based Convolutional Neural Networks (Fast-RCNNs) [4], Single-Shot Detection (SSD) [5], and You Only Look Once (YOLO) [6,7,8] have been proposed one after another, and the technology of pedestrian detection has achieved unprecedented development. These single-modal visual methods do not perform well in complex scenes, such as in poor lighting conditions and chaotic backgrounds. Infrared images are obtained by capturing the thermal radiation emitted by objects, which are less influenced by external conditions such as illumination, making robust pedestrian detection in complex scenes possible. Detection based on visible images usually achieves better performance under good illumination conditions due to abundant appearance details
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
