The advancement of computer vision technologies has led to significant improvements in object detection systems, which are critical for various applications in the automotive industry. The implementation and evaluation of object detection based on convolutional neural networks (CNN) takes front stage in this research article. A CNN-based object detection model is presented and evaluated for accuracy, precision, recall and F-1 score. The proposed model has shown the object detection accuracy of 97.33%. The experiments establish that the CNN based object detection substantially increases the precision of multiple objects’ recognition in the vehicular scenario. The proposed strategy of optimized CNN model is efficient for real-time vehicular applications thus contributing to the improvement of security and automation in current automobiles.
Scalable Object Detection Solution for Vehicular Applications
04.10.2024
566805 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Scalable parallel simulator for vehicular collision detection
Kraftfahrwesen | 2013
|Scalable parallel simulator for vehicular collision detection
Online Contents | 2013
|