Remote-Sensing Scene Classification Based on Memristor Convolutional Neural Network

Yibo Zhao; Yi Zhang; Chengcheng Yu; Qing Yang

doi:10.3788/LOP240560

Laser & Optoelectronics Progress, Volume. 61, Issue 18, 1828006(2024)

Remote-Sensing Scene Classification Based on Memristor Convolutional Neural Network

Yibo Zhao^1,2、*, Yi Zhang^1,2, Chengcheng Yu^1,2, and Qing Yang^1,2

¹School of Electronic & Information Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, Jiangsu, China

²Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing University of Information Science & Technology, Nanjing, 210044, Jiangsu, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. Simulation results of Data-Driven Verilog-A ReRAM. (a) Bipolar resistance switching characteristics of Data-Driven Verilog-A ReRAM; (b) hysteresis loop of Data-Driven Verilog-A ReRAM

Download full size

Fig. 2. Simulation results of Updated Data-Driven Verilog-A ReRAM. (a) Resistance switch characteristics of Updated Data-Driven Verilog-A ReRAM; (b) hysteresis loop of Updated Data-Driven Verilog-A ReRAM

Download full size

Fig. 3. Improved convolutional neural network structure

Download full size

Fig. 4. Cloblock structure diagram

Download full size

Fig. 5. SK module structure diagram

Download full size

Fig. 6. Structure of convolutional neural network based on memristor

Download full size

Fig. 7. Processing flow chart of memristor convolutional neural network

Download full size

Fig. 8. Sample graphs of UCMercedLandUse dataset

Download full size

Fig. 9. Sample maps of NWPU-RESISC45 dataset

Download full size

Table 1. Comparison of model simplification test set accuracy under data sets UC-21 and NP-45
View table
Table 1. Comparison of model simplification test set accuracy under data sets UC-21 and NP-45
Model Cloblock SK Accuracy（UC-21）/% Accuracy（NP-45）/%
ResNet18 88.81 82.47
ResNet18-C √ 92.85 85.84
ResNet18-S √ 92.62 86.17
Proposed √ √ 94.76 87.54

Table 2. Comparison of test set accuracy and parameter number of different models under data sets UC-21 and NP-45

View table

Table 2. Comparison of test set accuracy and parameter number of different models under data sets UC-21 and NP-45

Model	Parameter /10⁶	Accuracy（UC-21）/%	Accuracy（NP-45）/%
ResNet18	11.18	88.81	82.47
SENet-ResNet18^［14］	11.27	89.76	85.71
SANet-ResNet50^［36］	25.56	90.62	85.93
Xception^［37］	20.85	89.78	82.27
GhostNetv2^［38］	6.16	91.19	83.11
Mobilenetv3-small^［39］	2.54	85.95	84.52
SGENet-Res^［40］	2.78	93.09	85.90
LSKNet-Res^［41］	2.80	93.80	87.06
Proposed	3.14	94.76	87.54

Tools

Get Citation

Copy Citation Text

Yibo Zhao, Yi Zhang, Chengcheng Yu, Qing Yang. Remote-Sensing Scene Classification Based on Memristor Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2024, 61(18): 1828006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Remote Sensing and Sensors

Received: Jan. 19, 2024

Accepted: Mar. 4, 2024

Published Online: Sep. 9, 2024

The Author Email: Yibo Zhao (yibozhaodn@163.com)

DOI:10.3788/LOP240560

CSTR:32186.14.LOP240560

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Comparison of model simplification test set accuracy under data sets UC-21 and NP-45

Table 1. Comparison of model simplification test set accuracy under data sets UC-21 and NP-45

Table 2. Comparison of test set accuracy and parameter number of different models under data sets UC-21 and NP-45

Table 2. Comparison of test set accuracy and parameter number of different models under data sets UC-21 and NP-45