Deep Learning Target Vehicle Detection Method Based on YOLOv3-tiny
Lifu Li, Yi Liang · 2021
When the millimeter-wave radar is tracking the target, it can only track the movement state of the target, and cannot detect the category of the tracking target. In order to obtain the category of the tracking target, a vision camera is usually used to detect the category of the tracking target when detecting the tracking target. So as to solve the problems of high time-consuming and low-precision caused by traditional machine vision vehicle detection methods used by vision cameras, this paper uses the YOLOv3 deep learning algorithm based on convolutional neural networks to detect tracking targets. In order to speed up the detection speed of the algorithm, the YOLOv3 algorithm model is simplified by the method of pruning quantization, and the training of the YOLOv3-tiny algorithm model is completed, so that the real-time and accurate tracking target detection requirements in terms of detection time and detection accuracy can be achieved.