Small target rapid detection method based on deep convolution neural network

Abstract

The invention discloses a small target rapid detection method based on a deep convolution neural network. The deep convolution neural network is improved by the following steps: selecting the sliding windows on the convolution feature map of the last shared convolution layer of a VGG16 network as candidate boxes, wherein the sliding windows adopted are half-pixel precision sliding window; deleting a fifth pooling layer, and retaining other convolution layers and pooling layers; adding a convolution layer with a 3*3 convolution kernel; and using two convolution layers with 1*1 convolution kernels to replace all full-connection layers in the network to get the network adopted in the invention, training the network using collected data to get a small target classification model, and using the model to detect small targets. By using the method, the computational complexity is reduced, and the detection rate of small targets is improved.

Description

technical field [0001] The invention belongs to the field of automatic vehicle driving and the field of advanced driving assistance systems, and in particular relates to a fast detection method for small targets based on a deep convolutional neural network. Background technique [0002] Autonomous vehicles driving (AVD), also known as unmanned vehicles, computer-driven vehicles or wheeled mobile robots, is an intelligent vehicle that realizes unmanned driving through a computer system. Self-driving cars rely on artificial intelligence, visual computing, radar, surveillance devices and global positioning systems to work together to allow computers to operate motor vehicles automatically and safely without any human intervention. The technical field of advanced driver assistance system is derived from and serves the field of automatic vehicle driving technology. auxiliary system. It is not difficult to find that each subsystem is inseparable from the detection of small targe...

AI Technical Summary

Problems solved by technology

Experiments have proved that the model-based method works better, but because the neural network has a large number of convolution operations, the amount of calculation and memory usage in the training, testing and use of the model are very large, and the requirements for hardware are very high, and the real-time effect Poor, not suitable for the development and application of embedded products in traffic scenarios, most experts are currently working on improving the speed and accuracy of this method

Images