当前位置: X-MOL 学术Softw. Pract. Exp. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
An unmanned aerial vehicles navigation system on the basis of pattern recognition applications—Review of implementation options and prospects for development
Software: Practice and Experience ( IF 3.5 ) Pub Date : 2021-03-03 , DOI: 10.1002/spe.2964
Nidal Al Said 1 , Yuri Gorbachev 2 , Aleksei Avdeenko 3
Affiliation  

The popularity of unmanned aerial vehicles (UAVs) kept growing over the past few decades. UAVs are widely used in various fields for monitoring, mapping, aerial photography, rescue operations, and so forth. However, the UAV navigation may be difficult in areas where global positioning systems is not available. This article presents an analytical review of the promising methods of creating a navigation system based on a pattern recognition algorithm. Linear algebra and mathematical morphology to describe the operation of pattern recognition applications in UAV's navigation systems, classification of images through profound research of intelligent navigation systems, analysis and digital image processing to identify options for the implementation of artificial intelligent systems of visual navigation were used to identify existing pattern recognition applications in the process of constructive systematization. It was shown that the UAV navigation requires objects in the image to be recognized and the distance to a possible obstacle to be estimated using deep learning method for image segmentation and depth map estimation. An obligatory initial point in the image recognition algorithm is the image preprocessing procedures based on machine vision techniques. The results of this study may be useful in developing efficient UAV applications for military and civil purposes.

中文翻译:

一种基于模式识别应用的无人机导航系统——实现方案与发展前景回顾

在过去的几十年中,无人驾驶飞行器 (UAV) 的普及率不断提高。无人机广泛应用于监控、测绘、航拍、救援等各个领域。但是,在没有全球定位系统的地区,无人机导航可能会很困难。本文对基于模式识别算法创建导航系统的有前途的方法进行了分析回顾。线性代数和数学形态学来描述模式识别应用在无人机导航系统中的操作,通过对智能导航系统的深入研究对图像进行分类,使用分析和数字图像处理来识别视觉导航人工智能系统的实施选项,以识别在建设性系统化过程中现有的模式识别应用程序。结果表明,无人机导航需要识别图像中的物体,并使用深度学习方法进行图像分割和深度图估计,估计与可能障碍物的距离。图像识别算法的一个强制性起点是基于机器视觉技术的图像预处理程序。这项研究的结果可能有助于开发用于军事和民用目的的高效无人机应用。结果表明,无人机导航需要识别图像中的物体,并使用深度学习方法进行图像分割和深度图估计,估计与可能障碍物的距离。图像识别算法的一个强制性起点是基于机器视觉技术的图像预处理程序。这项研究的结果可能有助于开发用于军事和民用目的的高效无人机应用。结果表明,无人机导航需要识别图像中的物体,并使用深度学习方法进行图像分割和深度图估计,估计与可能障碍物的距离。图像识别算法的一个强制性起点是基于机器视觉技术的图像预处理程序。这项研究的结果可能有助于开发用于军事和民用目的的高效无人机应用。
更新日期:2021-03-03
down
wechat
bug