Double-light vehicle detection method and device based on uncertain sensing network

Abstract

The invention discloses a double-light vehicle detection method and device based on an uncertain sensing network. The method comprises the following steps: collecting and constructing a vehicle detection data set based on RGB-infrared rays of an unmanned aerial vehicle; constructing an uncertain sensing network, wherein the uncertain sensing network comprises an uncertain sensing module and a feature fusion framework; training the proposed uncertain sensing network through the constructed vehicle detection data set to obtain a classification and regression prediction result, and then calculating loss so as to update parameters in the network; and detecting the dual-light vehicle through the trained model. The device comprises a data set module, an uncertain sensing network module, a training module and a detection module. The uncertainty between the two modes is effectively measured, and the method can be flexibly applied to various multi-mode target detection algorithms; and no calculation consumption is increased, and various requirements in practical application are met.

Description

technical field [0001] The present invention relates to the field of multimodal target detection, in particular to the construction of a drone-based dual-light vehicle detection data set (DroneVehicle) and a method and device for improving multimodal target detection tasks through an uncertain perception network. Background technique [0002] In recent years, UAV-based computer vision technology has played an extremely important role in smart city construction and disaster relief. A drone equipped with a camera can collect images with a wider field of view, which is more conducive to capturing objects on the ground. However, the images collected above are all bird's-eye views, with highly complex backgrounds and extremely variable lighting. Therefore, object detection based on UAV images is an important but challenging task in the field of computer vision. [0003] There have been some influential works on object detection tasks based on UAV imagery. However, these studie...

AI Technical Summary

Problems solved by technology

These research works have a good role in promoting the development of this field, but these research works usually only collect data sets on urban streets, and there are still some new problems in aerial scenes

There are still no large-scale aerial scene datasets that can be used for multimodal object detection research

Images