Abstract
As countries around the world improve their garbage recycling and processing policies, the intelligently and efficiently garbage classification and identification has become a key point for implementing policies. However, traditional image recognition methods still have disadvantages, for instance, it needs a large amount of data annotation and a long time is required to train the model. In response to these drawbacks, this paper proposes a transfer learning method based on ResNet model, which aims to solve the problem of efficient classification of small-scale garbage image data sets. For the small sample image data set, after the data augmentation, the pre-training model ResNet50 is migrated to the data set through two migration learning methods of fine-tuning and pre-training model as the feature extractor, so as to realize the training of the target model. The experimental results show that the model classification effect after fine-tuning method and hyperparameter adjustment is better than the model without transfer learning, which can effectively improve the training speed and accuracy, and reduce the impact of over-fitting.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Sajid, M., Ali, N., Dar, S.H., et al.: Short search space and synthesized-reference re-ranking for face image retrieval. Appl. Soft Comput. 99, 106871 (2021)
Bi, Y., Xue, B., Zhang, M.: Multi-objective genetic programming for feature learning in face recognition. Appl. Soft Comput. 103, 107152 (2021)
Ma, Y., Li, Z., Sotelo, M.A.: Testing and evaluating driverless vehicles’ intelligence: the Tsinghua lion case study. IEEE Intell. Transp. Syst. Mag. 12(4), 10–22 (2020)
Ma, L., Zhang, Y.: Research on vehicle license plate recognition technology based on deep convolutional neural networks. Microprocess. Microsyst. 82, 103932 (2021)
Duan, M., Wang, G., Niu, C.: Small sample image recognition method based on convolutional neural network. Comput. Eng. Des. 39(1), 224–229 (2018)
Wang, J.: Research on image recognition of crop diseases and weeds based on convolutional neural network and transfer learning (2019)
Chang, L., Deng, X.M., Zhou, M.Q., et al.: Convolutional neural networks in image understanding. Acta Automatica Sinica 42(9), 1300–1312 (2016)
Zhuang, F., Qi, Z., Duan, K., et al.: A comprehensive survey on transfer learning. Proc. IEEE 1–34 (2020)
Long, M., Wang, J., Ding, G., et al.: Transfer learning with graph co-regularization. IEEE Trans. Knowl. Data Eng. 26(7), 1805–1818 (2014)
Wang, J., Chen, Y., Hao, S., et al.: Balanced distribution adaptation for transfer learning. In: IEEE International Conference on Data Mining (2017)
Acknowledgement
This work was supported by the Major science and technology project of Hainan Province (Grant No. ZDKJ2020012), Key Research and Development Program of Hainan Province (Grant No. ZDYF2020040),Hainan Provincial Natural Science Foundation of China (Grant Nos. 2019RC098) and National Natural Science Foundation of China (NSFC) (Grant No. 62162022, 62162024 and 61762033).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Liu, L., Cheng, J., Xie, L., Song, J., Zhou, K., Liu, J. (2021). A Transfer Learning Method Based on ResNet Model. In: Tan, Y., Shi, Y., Zomaya, A., Yan, H., Cai, J. (eds) Data Mining and Big Data. DMBD 2021. Communications in Computer and Information Science, vol 1453. Springer, Singapore. https://doi.org/10.1007/978-981-16-7476-1_23
Download citation
DOI: https://doi.org/10.1007/978-981-16-7476-1_23
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-7475-4
Online ISBN: 978-981-16-7476-1
eBook Packages: Computer ScienceComputer Science (R0)