期刊文献+

面向DSP的无人机航拍图像清晰化系统的设计与实现 被引量:3

Design and implementation of UAV aerial photography image clarity system based on DSP
下载PDF
导出
摘要 影响无人机航拍成像清晰化的原因有两点:一是无人机在飞行拍摄过程中因存在多种运动模式,导致航拍成像存在运动模糊问题;二是航拍观测由于成像光路长,即使在无雾天气下空气粒子对光线的衰减也较大。此外,现有图像清晰化处理的各类主流算法大都基于PC机设计实现,未考虑底层硬件资源配置问题,难以应用到资源有限的嵌入式设备当中。该文针对上述影响航拍观测清晰化的问题进行了原理性探索。首先,结合无人机成像特点对其飞行模式进行理论分析,估算航拍运动模糊核的像移长度和像移角度两个模拟参数;其次,使用DSP开发板,搭建了一种结构简单、运行简单方便、适于资源有限的嵌入式DSP平台的航拍图像清晰化(去模糊和去雾)系统;最后,对所提算法和所搭建系统展开实验测试,进一步验证了该文所提无人机航拍清晰化方法的有效性,以及所设计的DSP嵌入式图像清晰化系统的创新性和实用性。 The reasons that affect the clearness of UAV aerial images are relatively complicated.Firstly,due to multiple motion modes during UAV flight,the motion blurs in aerial images are common problems.Secondly,due to the long imaging light path of aerial observation,the light attenuation of air particles is also large even in haze-free weather.In addition,most of the existing mainstream algorithms for images clearness are designed and implemented based on PC,which does not consider whether the underlying hardware allocation is sufficient.Under the background,it is difficult to apply them to various embedded devices with limited resources.This paper makes a rational exploration of the prominent problems that affect the clearness of UAV aerial observation.Firstly,we theoretically analyze its flight mode combining the characteristics of UAV imaging,and estimate the two motion parameters of the motion blur kernel,which are image motion length and angle.Secondly,we use the DSP development board to build anaerial images clearness(deblurring and dehazing)system with a simple structure,convenient operation,which is suitable for embedded DSP platforms with limited resources.Finally,the experimental test of the proposed algorithm and the system is carried out,which further verifies the effectiveness of the proposed UAV aerial image clarity method,and the innovation and practicability of the DSP embedded image clarity system.
作者 宫玲瑞 姜博 王金帅 孟娜 汪霖 张嘉洋 陈晓璇 GONG Lingrui;JIANG Bo;WANG Jinshuai;MENG Na;WANG Lin;ZHANG Jiayang;CHEN Xiaoxuan(School of Information Science and Technology, Northwest University, Xi′an 710127, China)
出处 《西北大学学报(自然科学版)》 CAS CSCD 北大核心 2022年第2期270-278,共9页 Journal of Northwest University(Natural Science Edition)
基金 国家重点研发计划项目(2019YFC1510503) 国家自然科学基金(61801384) 西北大学古生物信息学创新团队项目(2019TD-012) 陕西省自然科学基础研究计划项目(2020JM-415) 陕西省国际科技合作计划项目(2021KW-05)。
关键词 航拍图像去模糊 航拍图像去雾 DSP aerial image deblurring aerial image dehazing DSP
  • 相关文献

参考文献5

二级参考文献26

共引文献21

同被引文献32

引证文献3

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部