基于DSP的无人机遥感影像SIFT算法设计与实现

doi:10.11772/j.issn.1001-9081.2019091689

计算机应用 ›› 2020, Vol. 40 ›› Issue (4): 1237-1242.DOI: 10.11772/j.issn.1001-9081.2019091689

• 应用前沿、交叉与综合 • 上一篇

基于DSP的无人机遥感影像SIFT算法设计与实现

孙鹏^1,2, 肖经¹, 赵海盟^2,3, 刘帆², 晏磊^2,3, 赵红颖²

1. 桂林电子科技大学机电工程学院, 广西桂林 541004;
2. 空间信息集成与3S工程应用北京市重点实验室(北京大学), 北京 100871;
3. 广西高校无人机遥测重点实验室(桂林航天工业学院), 广西桂林 541004

收稿日期:2019-10-09 修回日期:2019-11-19 出版日期:2020-04-10 发布日期:2019-12-05
通讯作者: 赵海盟
作者简介:孙鹏(1992-),男,河南邓州人,硕士研究生,主要研究方向:嵌入式系统设计、数字图像处理;肖经(1978-),男,江西泰和人,副教授,博士,主要研究方向:光通信、光传输;赵海盟(1980-),男,陕西咸阳人,高级工程师,主要研究方向:航空航天遥感、智能仪器设备;刘帆(1995-),女,河北衡水人,硕士研究生,主要研究方向:遥感数字图像处理;晏磊(1956-),男,湖北武汉人,教授,博士,主要研究方向:高分辨率成像、遥感定标;赵红颖(1971-),女,吉林长春人,副教授,博士,主要研究方向:遥感成像。
基金资助:
国家重点研发计划项目（2017YFB0503003）；国家自然科学基金资助项目（61841101，61764002）。

Design and implementation of SIFT algorithm for UAV remote sensing image based on DSP platform

SUN Peng^1,2, XIAO Jing¹, ZHAO Haimeng^2,3, LIU Fan², YAN Lei^2,3, ZHAO Hongying²

1. School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin Guangxi 541004, China;
2. Beijing Key Laboratory of Spatial Information Integration and 3S Applications(Peking University), Beijing 100871, China;
3. Guangxi Colleges and Universities Key Laboratory of Unmanned Aerial Vehicle(UAV) Remote Sensing(Guilin University of Aerospace Technology), Guilin Guangxi 541004, China

Received:2019-10-09 Revised:2019-11-19 Online:2020-04-10 Published:2019-12-05
Supported by:
This work is partially supported by the National Key Research and Development Program of China(2017YFB0503003, the National Natural Science Foundation of China(61841101,61764002).

摘要/Abstract

摘要： 为了满足尺度不变特征变换（SIFT）算法临场处理大尺寸无人机（UAV）组网遥感观测影像的实时快速需求，提出一种基于数字信号处理器（DSP）内核的硬件乘法器来处理单精度浮点型像素数据乘法的算法实现方案。首先，根据DSP内核的硬件乘法器的数据输入、输出特性，重构SIFT算法的图像数据结构和图像函数，以实现硬件乘法器对SIFT算法单精度浮点型像素数据的乘法计算；其次，采用软件流水技术重新编排迭代计算，以增强算法的并行计算能力；最后，将在算法计算过程中产生的动态数据迁移至第三代双倍速率同步动态随机存储器（DDR3）中，以提升算法数据的存储空间。实验结果表明，DSP平台的SIFT算法可以实现对1 000×750的UAV遥感影像的高精度快速处理，所提方案满足无人机组网遥感影像临场处理对SIFT算法的实时快速要求。

关键词: 数字信号处理器, 乘法器, 图像数据结构, 迭代计算, 临场快速处理

Abstract: To satisfy the requirement of real-time and rapid processing of Scale-Invariant Feature Transform(SIFT) algorithm for remote sensing images of large-scale Unmanned Aerial Vehicle (UAV) network on the scene, an implementation scheme of the algorithm by using the hardware multiplier of Digital Signal Processor(DSP)kernel was proposed to process the multiplication of single-precision floating-point pixel data. Firstly,according to the characteristics of data input and output of the hardware multiplier with DSP kernel,the image data structure and the image function of SIFT algorithm were reconstructed in order to make the hardware multiplier perform the multiplication calculation of single-precision floating-point pixel data of SIFT algorithm. Secondly,the software pipelining technology was adopted to rearrange the iterative computation,so as to enhance the parallel computing ability of the algorithm. Finally,the dynamic data produced in the algorithm calculation process were transferred to the Double Data Rate 3 synchronous dynamic random access memory(DDR3)to enlarge the storage space of the algorithm data. Experimental results show that the SIFT algorithm on DSP platform is able to achieve high-precision and fast processing for 1 000×750 remote sensing images of UAV,and the scheme satisfies the requirement of real-time and rapid processing of SIFT algorithm for remote sensing images of UAV network on the scene.

Key words: Digital Signal Processor (DSP), multiplier, image data structure, iterative calculation, rapid processing on the scene

中图分类号:

TP391.75

孙鹏, 肖经, 赵海盟, 刘帆, 晏磊, 赵红颖. 基于DSP的无人机遥感影像SIFT算法设计与实现[J]. 计算机应用, 2020, 40(4): 1237-1242.

SUN Peng, XIAO Jing, ZHAO Haimeng, LIU Fan, YAN Lei, ZHAO Hongying. Design and implementation of SIFT algorithm for UAV remote sensing image based on DSP platform[J]. Journal of Computer Applications, 2020, 40(4): 1237-1242.

参考文献

[1] LOWE D G. Object recognition from local scale-invariant features[C]//Proceedings of the 7th IEEE International Conference on Computer Vision. Piscataway:IEEE,1999:1150-1157.
[2] 孙科林. 基于多核DSP的实时图像处理平台研究[D]. 成都:电子科技大学,2012. (SUN K L. Research on real-time image processing platform based on multi-core DSP[D]. Chengdu:University of Electronic Science and Technology of China,2012:1-2.
[3] 赵立双. 基于DSP的单目视觉定位方法关键技术研究[D]. 长沙:国防科学技术大学,2011. (ZHAO L S. Research on key technologies of monocular visual odometry based on DSP[D]. Changsha:National University of Defense Technology,2011:6-7.)
[4] 许飞, 刘威. 多核DSP下的SIFT算法优化[J]. 信息通信,2012(5):42-43. (XU F,LIU W. SIFT algorithm optimization based on multi-core DSP[J]. Information & Communications,2012(5):42-43.)
[5] 刘颜开, 陈远知. 基于C6678的SIFT算法并行计算研究[C]//第九届全国信号和智能信息处理与应用学术会议. 北京:计算机工程与应用,2015:12-16.(LIU Y K,CHEN Y Z. Research on parallel computing of SIFT algorithm based on C6678[C]//Proceedings of the 9th National Conference on Signal and Intelligent Information Processing and Application. Beijing:Computer Engineering and Applications,2015:12-16.)
[6] 高聪. 基于SIFT算法图像配准的FPGA实现[D]. 成都:电子科技大学, 2013. (GAO C. FPGA implementation of image registration based on SIFT algorithm[D]. Chengdu:University of Electronic Science and Technology of China,2013:3-4.)
[7] 贾云得. 机器视觉[M]. 北京:科学出版社,2000:26-32. (JIA Y D. Machine Vision[M]. Beijing:Science Press,2000:26-32.)
[8] 章毓晋. 图像工程(上册)-图像处理[M]. 北京:清华大学出版社,1999:10-20. (ZHANG Y J. Image Engineering(I)-Image Processing[M]. Beijing:Tsinghua University Press,1999:10-20.)
[9] QASAIMEH M,SALEH H,MOHAMMAD B,et al. A novel SIFT architecture and ASIC implementation for real time SOC application[J]. Analog Integrated Circuits and Signal Processing,2019,99(2):325-338.
[10] 宋海涛, 何文浩, 原魁. 一种基于SIFT特征的机器人环境感知双目立体视觉系统[J]. 控制与决策,2019,34(7):1545-1552. (SONG H T,HE W H,YUAN K. A stereo vision system based on SIFT feature for robot environment perception[J]. Control and Decision,2019,34(7):1545-1552.)
[11] 夏际金, 赵洪立, 李川. TI C66x多核DSP高级软件开发技术[M]. 北京:清华大学出版社,2017:6-8. (XIA J J,ZHAO H L, LI C. Advanced Software Development Technology for TI C66x Multi-core DSP[M]. Beijing:Tsinghua University Press,2017:6-8.)
[12] DAMODARAN R,ANDERSON T,TRAN J,et al. A 1.25 GHz 0.8W C66x DSP Core in 40 nm CMOS[C]//Proceedings of the 25th International Conference on VLSI Design. Piscataway:IEEE,2012:286-291.
[13] 王俊, 张玉玺, 杨彬. DSP/FPGA嵌入式实时处理技术及应用[M]. 北京:电子工业出版社,2015:67-68. (WANG J,ZHANG Y X, YANG B. The Real-time Processing Technology and Application on Embedded Platform with DSP or FPGA[M]. Beijing:Publishing House of Electronic Industry,2015:67-68.)
[14] 苏保禹. 基于TI-C6678的多核DSP图像处理系统研究[D]. 成都:中国科学院大学,2014. (SU B Y. Research on image processing system based on multi-core DSP TI-C6678[D]. Chengdu:University of Chinese Academy of Sciences,2014:45-46.)
[15] 朱慧. 基于C66AK多核处理器的H. 264到HEVC视频转码软件设计[D]. 杭州:浙江工业大学,2016. (ZHU H. Software design of H. 264 to HEVC video transcoder based on C66AK mult-icore processor[D]. Hangzhou:Zhejiang University of Technology,2016:17-18

基于DSP的无人机遥感影像SIFT算法设计与实现

Design and implementation of SIFT algorithm for UAV remote sensing image based on DSP platform

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

[1]	唐国斐, 周海芳, 谭庆平. 基于多核DSP的星载并行遥感图像压缩系统设计与实现[J]. 计算机应用, 2017, 37(5): 1246-1250.
[2]	黄胜兵, 郑启龙, 郭连伟. 分簇VLIW DSP上支持单双字模式选择的SIMD编译优化[J]. 计算机应用, 2015, 35(8): 2371-2374.
[3]	马慧芳, 师亚凯, 谢蒙, 庄福振. 融合用户内容与关系结构的用户影响力算法[J]. 计算机应用, 2015, 35(12): 3487-3490.
[4]	张毅刘钰然罗元. 基于视觉的手势识别方法及其在数字信号处理器上的实现[J]. 计算机应用, 2014, 34(3): 833-836.
[5]	黄一才郁滨. 基于蓝牙单芯片的复杂密码算法实现机制[J]. 计算机应用, 2012, 32(12): 3453-3455.
[6]	索维毅赵荣彩姚远刘鹏. 面向DSP的超字并行指令分析和冗余优化算法[J]. 计算机应用, 2012, 32(12): 3303-3307.
[7]	陶巍刘建平张一闻. 基于麦克风阵列的声源定位系统[J]. 计算机应用, 2012, 32(05): 1457-1459.
[8]	陈明宋宝唐小琦. 基于DSP和FPGA的网络转换网关[J]. 计算机应用, 2011, 31(10): 2617-2620.
[9]	陈海燕黄虹. 基于E-Bus的YHFT-DSP仿真测试平台的设计与实现[J]. 计算机应用, 2011, 31(04): 1129-1132.
[10]	鲍可进姚成. 基于双核数字信号处理器的人脸识别门禁系统[J]. 计算机应用, 2010, 30(9): 2553-2555.
[11]	彭元喜杨洪杰谢刚. X-DSP浮点乘法器的设计与实现[J]. 计算机应用, 2010, 30(11): 3121-3125.
[12]	母其勇任磊王俊肖卉. 高速USB2.0接口的音频多通道采集无线传输系统[J]. 计算机应用, 2010, 30(10): 2665-2668.
[13]	彭元喜邹佳骏. X-DSP ALU与移位部件的设计与实现[J]. 计算机应用, 2010, 30(07): 1978-1982.