Research On Reconstruction And Display Technology Of Panoramic Perceptional Task Environment For Multiple UAVs

The perception and reconstruction of high-quality mission environment is a key problem for Unmanned Aerial Vehicle system to achieve autonomous control.Therefore,UAVs equipping with vision sensors is an efficient method to acquire the needed environmental information.This paper chooses the reconstruction of the mission environment of the multiple UAVs as the research object.Firstly,the sensor image preprocessing,including segmenting the sky region and the non-sky region,detecting moving target,etc.Then,the adaptive algorithm is designed to reconstruct the task environment.Finally,the reconstruction results are integrated to render real 3D output based on VR equipment.The main work and innovations are as follows:(1)Due to the complexity and dynamics of the environment,the images acquired by UAVs usually have uncertainty,which makes reconstruct the environment become a very challenging task.We put forward a general idea: Firstly,we segment the sky region and the non-sky region,then process the sky image and the ground image respectively,and finally conduct a high-immersion display.(2)In image preprocessing,we propose segmenting an image into sky and non-sky regions method which is applicable and adaptive is designed.It solves the problem of the traditional methods that are not adaptive to the changes in color and texture characteristics.(3)Aiming at the problem of "ghost" caused by sensor image containing moving objects.Firstly,motion object extraction is identified based on deep learning method.Then the image after extraction is automatically repaired.Finally,sending to the images stitching process.(4)This paper presents a new method for stitching panoramas from UAV’s images which is mainly composed of three steps.Firstly,once the images have been obtained,global motion model for UAV aerial photography can be used to establish a predicted region.Second,features in the onboard images are matched by using SIFT algorithm in this region.Finally,the accuracy of image mosaic is predicted through the Grey Relational Analysis.Then,according to the result,the algorithm process can adjust itself automatically to meet the needs of different types of sensor image stitching.(5)The experimental prototype of task environment reconstruction is designed.Meanwhile,based on Oculus headset display,an immersive VR display system is constructed.The result of real-time reconstruction is used to realize the embedding display of high dynamic and high authenticity task environment.