Gas identification using electronic nose via gramian-angular-field-based image conversion and convolutional neural networks architecture search

Zhu, Yudi1,2; Wang, Tao1,2; Li, Zhuoheng1,2; Ni, Wangze1,2; Zhang, Kai1,2; He, Tong1,2; Fu, Michelle4; Zeng, Min1; Hu, Nantao1,2; Cai, Wei3; Yang, Zhi1

2024-10-15

10.1016/j.snb.2024.136198

SENSORS AND ACTUATORS B-CHEMICAL   影响因子和分区

0925-4005

Recent years have witnessed the splendid performance of deep learning methods used in gas recognition for electronic noses (E-nose). In addition to effective feature extraction, the architecture of the deep neural network plays a vital role. Currently, most applied network structures are hand-crafted by human experts, which is timeconsuming and problem-dependent, making it necessary to design the structures of neural networks according to specific demands. In this work, a genetic algorithm with particle swarm optimization (GA-PSO), which possesses promising optimization capabilities, is applied to search for effective deep convolutional neural networks (CNNs) for gas classification based on E-nose technology. A novel image transformation strategy using Gramian angular field and a hybrid cost-saving method is employed in the search process, enabling adaptive and efficient CNN search on gas datasets. With the proposed methods, we can achieve an average classification accuracy of over 90 % on two public gas datasets, while also significantly reducing the model size compared to state-of-the-art CNNs. By using these novel strategies, our approach surpasses random search and basic PSO algorithm in achieving the global optimal solution, higher and more stable accuracy, and faster convergence in pattern recognition using E-nose. Our work suggests that the proposed method can quickly identify excellent CNN structures for E-nose applications.

Neural architecture search Hybrid algorithm Convolutional neural network Gas identification Gramian angular field

SCI ; EI

英语

其他

National Key Research and Development Program of China[2022YFC3104700] ; National Natural Science Foundation of China["62371299","62301314","62101329"] ; China Postdoctoral Science Foundation[2023M732198] ; Natural Science Foundation of Shanghai[23ZR1430100]

Chemistry ; Electrochemistry ; Instruments & Instrumentation

Chemistry, Analytical ; Electrochemistry ; Instruments & Instrumentation

WOS:001264859000001

ELSEVIER SCIENCE SA

20242716616417

Classification (of information) ; Convolution ; Convolutional neural networks ; Deep neural networks ; Gas detectors ; Gases ; Genetic algorithms ; Image processing ; Learning systems ; Network architecture ; Particle swarm optimization (PSO)

Ergonomics and Human Factors Engineering:461.4 ; Information Theory and Signal Processing:716.1 ; Computer Software, Data Handling and Applications:723 ; Data Processing and Image Processing:723.2 ; Chemistry:801 ; Information Sources and Analysis:903.1 ; Accidents and Accident Prevention:914.1 ; Optimization Techniques:921.5 ; Electric and Electronic Instruments:942.1 ; Special Purpose Instruments:943.3

CHEMISTRY

Web of Science

1.Shanghai Jiao Tong Univ, Natl Key Lab Adv Micro & Nano Manufacture Technol, Shanghai 200240, Peoples R China
2.Shanghai Jiao Tong Univ, Sch Elect Informat & Elect Engn, Dept Micro Nano Elect, Shanghai 200240, Peoples R China
3.Southern Univ Sci & Technol, Dept Ocean Sci & Engn, Shenzhen 518055, Peoples R China
4.Shanghai Qibao Dwight High Sch, Shanghai 201101, Peoples R China

Zhu, Yudi,Wang, Tao,Li, Zhuoheng,et al. Gas identification using electronic nose via gramian-angular-field-based image conversion and convolutional neural networks architecture search[J]. SENSORS AND ACTUATORS B-CHEMICAL,2024,417.

Zhu, Yudi.,Wang, Tao.,Li, Zhuoheng.,Ni, Wangze.,Zhang, Kai.,...&Yang, Zhi.(2024).Gas identification using electronic nose via gramian-angular-field-based image conversion and convolutional neural networks architecture search.SENSORS AND ACTUATORS B-CHEMICAL,417.

Zhu, Yudi,et al."Gas identification using electronic nose via gramian-angular-field-based image conversion and convolutional neural networks architecture search".SENSORS AND ACTUATORS B-CHEMICAL 417(2024).