Parallel design of convolutional neural networks for remote sensing images object recognition based on data-driven array processor

作     者：Shan Rui Jiang Lin Deng Junyong Cui Pengfei Zhang Yuting Wu Haoyue Xie Xiaoyan Shan Rui;Jiang Lin;Deng Junyong;Cui Pengfei;Zhang Yuting;Wu Haoyue;Xie Xiaoyan

作者机构：School of Electronic EngineeringXi'an University of Posts and TelecommunicationsXi'an 710121China Laboratory of Integrated Circuit DesignXi'an University of Science and TechnologyXi'an 710054China School of Computer Science and TechnologyXi'an University of Posts and TelecommunicationXi'an 710121China 

出 版 物：《The Journal of China Universities of Posts and Telecommunications》 (中国邮电高校学报（英文版）)

年 卷 期：2020年第27卷第6期

页      面：87-100页

核心收录：

学科分类：0810[工学-信息与通信工程] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 081002[工学-信号与信息处理] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported by the National Natural Science Foundation of China(61802304,61834005,61772417,61634004,61602377) the Shaanxi Provincial Co-ordination Innovation Project of Science and Technology(20I6KTZDGY02-04-02) the Shaanxi Provincial Key Research and Development Plan(2017GY-060) 

主　　题：convolutional neural networks remote sensing images object recognition array processor data-driven 

摘      要：Object recognition in very high-resolution remote sensing images is a basic problem in the field of aerial and satellite image *** the development of sensor technology and aerospace remote sensing technology,the quality and quantity of remote sensing images are *** recognition methods have a certain limitation in describing higher-level features,but object recognition method based on convolutional neural network(CNN)can not only deal with large scale images,but also train features automatically with high *** is mainly used on object recognition for remote sensing *** this paper,an AlexNet CNN model is trained using 2100 remote sensing images,and correction rate can reach 97.6%after 2000 *** based on trained model,a parallel design of CNN for remote sensing images object recognition based on data-driven array processor(DDAP)is *** consuming cycles are ***,the proposed architecture is realized on Xilinx V6 development board,and synthesized based on SMIC 130 nm complementary metal oxid semiconductor(CMOS)*** experimental results show that the proposed architecture has a certain degree of parallelism to achieve the purpose of accelerating calculations.