Abstract
In order to meet the needs of diversified image information retrieval in the real world, to solve the “semantic gap” problem of image and natural language conversion, and to optimize the accuracy and efficiency of multi-label classification method, this paper proposes a deep learning and multi-label BR algorithm. An image classification method uses a residual neural network with better overall performance to extract image depth learning features, and takes the extraction result as an input, and generates a result vector through spatial regularization of the image space and the label, and the result is obtained. The elements are added as the final prediction result by directly using the residual network prediction result. The whole network is trained by the softmax loss function. Compared with other traditional models, the spatial relationship results of the labels provide a good regularization effect for multi-label image classification, which improves the accuracy on the NUS-WIDE dataset and its recall rate, etc.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Li, S., Li, N., Li, Z.: Multi-tag data mining technology: a review. Comput. Sci. 40(04), 14–21 (2013)
Chen, Z.: Research on Graph Structure Description and Several Learning Algorithms for Multi-label Classification Problem. South China University of Technology (2015)
Feng, X.: Summarization of multi-label classification problems. Inf. Syst. Eng. (03), 137 (2016)
Krizhevsky, A., Sutskever, I., Hinton, G.: ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25(2), 84–90 (2012)
Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional networks. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8689, pp. 818–833. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10590-1_53
Simonyan, K., Zisserman, A.: Very Deep Convolutional Networks for Large-Scale Image Recognition. Computer Science. arXiv:1049.1556 (2014)
Szegedy, C., Liu, W., Jia, Y., et al.: Going deeper with convolutions. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, pp. 1–12 (2015)
Xu, K., et al.: Show, attend and tell: neural image caption generation with visual attention. In: ICML, vol. 2, no. 3 (2015)
Li, X., Zhao, F., Guo, Y.: Multi-label image classification with a probabilistic label enhancement model. In: Proceedings of the Uncertainty in Artificial Intelligence, vol. 1, no. 2 (2014)
Liu, G., Liu, S., Wu, J., Luo, W.: Machine vision target detection algorithm based on deep learning and its application in bill detection. China Test 45(05), 1–9 (2019)
Wang, Y., Zhang, H., Huang, H.: A survey of image semantic segmentation algorithms based on deep learning. Appl. Electron. Tech. 45(06), 23–27+36 (2019)
Li, Y.: Multispectral image and multi-label scene classification based on convolutional neural network. Electron. Des. Eng. 26(23), 25–29 (2018)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. arXiv preprint arXiv:1512.03385 (2015)
He, K., Zhang, X., Ren, S., Sun, J.: Identity mappings in deep residual networks. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 630–645. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_38
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. In: ICML (2015)
Montañes, E., Senge, R., Barranquero, J., et al.: Dependent binary relevance models for multi-label classification. Pattern Recogn. 47(3), 1494–1508 (2014)
Jia, Y., et al.: Convolutional architecture for fast feature embedding. arXiv pre-print arXiv:1408.5093 (2014)
Wang, L., Xiong, Y., Wang, Z., Qiao, Y.: Towards good practices for very deep two-stream ConvNets. CoRR, abs/1507.02159 (2015)
Chua, T.-S., Tang, J., Hong, R., Li, H., Luo, Z., Zheng, Y.: NUS-WIDE: a real-world web image database from National University of Singapore. In: Proceedings of the ACM International Conference on Image and Video Retrieval (2009)
Wang, J., Yang, Y., Mao, J., Huang, Z., Huang, C., Xu, W.: CNN-RNN: a unified framework for multi-label image classification. In: CVPR (2016)
Gong, Y., Jia, Y., Leung, T., Toshev, A., Ioffe, S.: Deep convolutional ranking for multilabel image annotation. In: ICLR (2014)
Funding
This research was funded by the National Natural Science Foundation of China under Grant No. 61762079, and No. 61662070, Innovation ability improvement project of colleges and universities in Gansu Province in 2019, Grant No: 2019B-024, the Fundamental Research Funds for the Central University of Northwest Minzu University, Grant No: 31920180050.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Wang, X., Xu, J., Hua, J., Hao, Z. (2020). Multi-label Image Classification Optimization Model Based on Deep Learning. In: Hao, Z., Dang, X., Chen, H., Li, F. (eds) Wireless Sensor Networks. CWSN 2020. Communications in Computer and Information Science, vol 1321. Springer, Singapore. https://doi.org/10.1007/978-981-33-4214-9_20
Download citation
DOI: https://doi.org/10.1007/978-981-33-4214-9_20
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-33-4213-2
Online ISBN: 978-981-33-4214-9
eBook Packages: Computer ScienceComputer Science (R0)