Figure 3 - available via license: Creative Commons Attribution 3.0 Unported
Content may be subject to copyright.
Detection results of Cascade RCNN (left) and VDDNet (right)
Source publication
Vehicle and driver detection in the highway scene has been a research hotspot in the field of object detection in recent years, and it is still a challenging problem in the research of traffic order and road safety. In this paper, we propose a novel end-to-end vehicle and driver detection method named VDDNet which is based on Cascade R-CNN and SENe...
Contexts in source publication
Context 1
... randomly select 7,200 images to train the VDDNet model, and the remaining 1800 images to verify the performance of the model. Figure 3, figure 4 and figure 5 show the detection results of Cascade R-CNN and VDDNet models under difficult environment(for example poor light, partial occlusion, sample blur, etc.) As can be seen from the figure, compared with Cascade R-CNN, the model proposed in this paper can improve the confidence of the object to be detected and optimize the positioning quality of the bounding box regression (figure 3), reduce false detection and improve accuracy (figure 4), and reduce leakage Inspection to improve recall (figure 5). ...
Context 2
... a result, the AP is improved, false alarms are reduced, and the recall rate is improved. It is worth emphasizing that in difficult scenarios (e.g., dark light ( figure 4,5), occlusion problems (figure 4), and low image resolution ( figure 3)), the improvement of these indicators is more significant. It can be seen that the improved model has better accuracy and robustness in challenging scenarios. ...
Similar publications
![Accidents recorded in Nvidia AI-City Challenge dataset [53]. Some cases...](publication/363194049/figure/fig1/AS:11431281095402860@1667876690414/Accidents-recorded-in-Nvidia-AI-City-Challenge-dataset-53-Some-cases-are-far-away-from_Q320.jpg)
Occasions such as stalled vehicles or crashes led by abnormal trajectories should be instantly identified and then dealt with quickly by the city traffic management system for the sake of road safety. However, a fast and accurate automatic detection system based on machine learning in general meets with great challenges from the shortage of recorde...
Citations
Power line inspection is an important means to eliminate hidden dangers of power lines. It is a difficult research problem how to solve the low accuracy of power line inspection based on deep neural network (DNN) due to the problems of multi‐view‐shape, small‐size object. In this paper, an automatic detection model based on Feature visual clustering network (FVCNet) for power line inspection is established. First, an unsupervised clustering method for power line inspection is proposed, and applied to construct a detection model which can recognize multi‐view‐shape objects and enhanced object features. Then, the bilinear interpolation method is used to Feature enhancement method, and the enhanced high‐level semantics and low‐level semantics are fused to solve the problems of small object size and single sample. In this paper, FVCNet is applied to the MS‐COCO 2017 data set and self‐made power line inspection data set, and the test accuracy is increased to 61.2% and 82.0%, respectively. Compared with other models, especially for the categories that are greatly affected by multi‐view‐shape, the test accuracy has been improved significantly.
The recent development in the domestic economy has increased the number of private vehicles leading to the issue of road congestions and traffic accidents. The diversity and severity of traffic problems leads to the requirements of modern intelligent transportation systems. In case of vehicle license identification, plate positioning plays a very vital role and this is the key factor affecting the accuracy of the system. In order to alleviate traffic pressure, solve the problem of road congestion, this paper is based on Convolutional Neural Network (CNN) license plate character identification. This article adopts the license plate character identification employing the CNN model and uses the neural network optimization principles for the improved construction. By adopting neural network principles, it is improved, and the license plate identification model is constructed. The results show that based on CNN-based license plate character identification model, the identification of license plate characters is completed, and the license plate recognition has been significantly improved, accurate rate of 99%.
