摘要
提高无人机的精度和处理性能是目前研究的一个热点。提出一种新的无人机图像处理和数据分析方法,使无人机在不同环境下能探测障碍物并找到最佳的飞行方向。该方法利用立体相机拍摄两幅不同的图像,然后利用图形处理单元(GPU)进行数据分析。与传统的CPU数据分析相比,可减少计算时间。该方法侧重于探测障碍物,并为无人机确定正确的飞行方向。
A popular topic for researchers is increasing the accuracy and processing performance of unmanned aerial vehicles(UAVs). This paper presents a novel image processing and data analysis method for UAVs to detect obstacles and find the optimal correct flight direction while the UAV is indifferent environments. This paper's proposed approach uses a stereo camera to grab two different images, and it then uses a graphic processing unit(GPU) to do data analysis. This will decrease the calculation time compared to data analysis in CPU.The approach focuses on finding obstacles and detecting the correct route position for the UAV.
引文
[1]H.Bavle.Car Detection in UAV Images[C].2014 International Journal of Thesis Projects and Dissertations(IJTPD),2014.
[2]J.SHI,J.WANG,Y.XU.Object-based change detection using georeferenced UAVimages[C].ISPRS-Int.Arch.Photogramm.R e m o t e S e n s.S p a t i a l I n f o r m.S c i.XXXVIII-1/C22:177-182.
[3]孟浩,程康.基于SIFT特征点的双目视觉定位[J].哈尔滨工程大学学报,2009,30(06):649-652+675.
[4]王民,刘伟光.基于改进SIFT特征的双目图像匹配算法[J].计算机工程与应用,2013,49(02):203-206.
[5]张文明,刘彬,李海滨.基于双目视觉的三维重建中特征点提取及匹配算法的研究[J].光学技术,2008(02):181-185.
[6]Tony Lindeberg.Feature Detection with Automatic Scale Selection[J].International Journal of Computer Vision,1998,30(2):79-116.
[7]Naveen Appiah,Nitin Bandaru.Obstacle Detection using stereo vision for selfdriving cars[M].
[8]B.JIA,R.LIU,M.ZHU.Real-time obstacle detection with motion features using monocular vision[J].The Visual Computer,2014,31(3):281-293.
[9]NVIDIA Tegra K1 Performance and Power Consumption Revealed-Xiaomi MiPad To Ship With 32-Bit and 64-Bit Denver Powered Chips[EB/OL].(2014-05-14)[2018-10].https://wccftech.com/nvidia-tegrak1-performance-power-consumptionrevealed-xiaomi-mipad-ship-32bit-64bitdenver-powered-chips.
[10]Depth Estimation From Stereo Video[EB/O L].(2 0 1 6)[2 0 1 8-1 0].h t t p s://w w w.mathworks.com/help/vision/examples/depth-estimation-from-stereo-video.html.
[11]J.Mrovlje,D.Vrancic?.Distance measuring based on stereoscopic pictures[C].9th International PhD Workshop on Systems and Control:Young Generation Viewpoint,Izola,Slovenia,2008-10-01.