Abstract
In order to improve the accuracy of CNN (convolutional neural network) in image classification, an enhanced Inception-ResNet-v2 model based on CNN is designed through the comparative study and analysis of the structure of classification model. This paper proposes to use multi-scale depthwise separable convolution to replace the convolution structure in Inception-ResNet-v2 model, which can reduce the amount of model parameters and extract features under different receptive fields. At the same time, this paper establishes channel filtering module based on global information comparison to filter and join channels, which realizes the effective extraction of features. Finally, through data enhancement, batch normalization and learning rate adjustment, the effect of the model used in this paper is better than most other models in each dataset, and the accuracy rate can reach 94.8%.










Similar content being viewed by others
References
Can S, Deng X, Yang L, Zhu J (2020) PCB defect detection based on improved inception-ResNet-v2 [J]. Inform Technol 44(09):33–36
Guo R, Xiongjie C, Luo W, Changhong F (2019) Inception ResNet a module and gradient boosting for crowd counting [J]. J Tongji Univ (Natural Science Edition) 47(08):1216–1224
Hao C, Xiaolei C, Aihua Z, Ce L, Dongmei L (2021) Continuous blood pressure prediction based on convolutional neural network embedded with improved SENET [J]. Comp Eng Appl 57(07):130–135
He X, Wu L, Zheng G, Wu J (2020) Inception- Resnetv2 for the diagnosis of breast cancer [J]. Auto Inform Eng 41(01):16–21
Jingzhi F Application of modified inception-ResNet and CondenseNet in lung nodule classification[a]. Wuhan Zhicheng times cultural development co., ltd..Proceedings of the 3rd international conference on computer engineering, Information Science & Application Technology(ICCIA 2019)[C].Wuhan Zhicheng times cultural development co., Ltd.
Kewen L, Xinyu L (2020) Faster R-CNN pedestrian detection model based on SENET [J]. Comput Syst Appl 29(04):266–271
Lin C, Gansen Z, Yang Z, Aihua Y, Wang X, Guo L, Hanbiao C, Ma Z, Lei Z, Luo H, Wang T, Bichao D, Xiongwen P, Qiren C (2020) CIR-net: automatic classification of human chromosome based on inception-ResNet architecture. [J] IEEE/ACM Trans Comput Biol Bioinform
Linlin L, Qiang Y, He L (2021) Breast cancer metastasis detection based on SENet multi-channel networks [J/OL]. Comp Eng Appl:1–9
McNeely-White D, Beveridge JR, Draper BA (2020) Inception and ResNet features are (almost) equivalent [J]. Cogn Syst Res 59:312–318
Ni L, Zou WJ (2020) Improved Xception for animal species identification based on SE module [J]. Navigation Control 19(02):106–111
Peisen Y, Shen C, Xu H (2021) A method for identification of fine-grained mushrooms based on transfer learning and bilineal inception- ResNET-v2 [J/OL]. J Agricult Machinery:1–10
Qiaohong C, Yi C, Wenwen L, Yubo J J Zhejiang Univ (Eng Sci), 2020, 54(09):1727–1735.
Rahimzadeh M, Attar A (2020) A modified deep convolutional neural network for detecting COVID-19 and pneumonia from chest X-ray images based on the concatenation of Xception and ResNet50V2[J]. Inform Med Unlocked 19:100360
Shen R, Yinglai H, Xin W, Lan Z (2020) Method of mushroom classification based on Xception and Resnet50 model [J]. J Heihe Univ 11(07):181–184
Wenqian D, Yu P, Haiyan L, Lu X (2021) Weakly supervised fine-grained image classification based on Xception network [J/OL]. Comp Eng Appl:1–10
Xia L, Shen X, Yongxia Z, Wang X, Tie-Qiang L (2020) Classification of breast cancer histopathological images using interleaved DenseNet with SENet (IDSNet). [J] PloS One 15(5)
Xu X, Zhang J, Liu W, Lu L, Zhao Y (2021) Fusion space and channel characteristics of high precision breast classification method [J/OL]. Comput Appl:1–9
Yan Z, Wang M, Wang J, Kaining J, Yunhan Z (2020) Eval lower limb motor ability based Xception and LSTM [J]. Chin J Rehab Theory Pract 26(06):643–647
Yang H (2020) Study on small sample MRI detection method of cervical spondylosis based on SENET [D]. Jilin University
Yu H, Jingyu H, Yizhuo Z (2020) Classification of ground objects in hyperspectral images based on multi-level feature SENet and multi-scale wide residuals [J]. Lab Res Explor 39(07):28–34+44
Acknowledgements
This work is supported by Natural Science Foundation of China (No. 61871432, No. 61771492), the Natural Science Foundation of Hunan Province (No.2020JJ4275, No.2019JJ6008, and No.2019JJ60054), National College Students’ research based learning and innovation experimental project(No.201811535012), and Research based learning and innovative experiment project for college students in Hunan Province(No.S201911535027).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Peng, C., Liu, Y., Yuan, X. et al. Research of image recognition method based on enhanced inception-ResNet-V2. Multimed Tools Appl 81, 34345–34365 (2022). https://doi.org/10.1007/s11042-022-12387-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-12387-0