| |
| For automatic driving, the perception of the surrounding environment of the car is one of the most basic prerequisites. Among them, visual-based environmental perception is an extremely important part. Traffic signal lights are very important information in real-time video captured by vehicle cameras, which can quickly and accurately detect traffic signal lights travel safety. Because the pixels of traffic lights are relatively small in the shooting pictures of vehicle-mounted cameras, the detection effect of general purpose target detection detectors is not ideal. On the basis of Faster R-CNN and from the point of view of feature fusion, this paper adjusts and improves the network structure by extracting two ROI features and subsequent fusion, which makes the detector have better detection effect on small targets such as traffic lights. Several sets of comparative experiments on TT100K data set are carried out and the experimental results are analyzed scientifically and fully, which proves that our new model can effectively improve the detection accuracy of traffic lights. |
|
| Keywords:Pattern Recognition; Object Detection; Traffic lights; Feature fusion |
|
