The using of deep learning method belongs to the application and research of artificial intelligence technology in medical field for assisting medical image information processing. Due to the larger difference between the image underlying features and high-level semantic concept described, the existing image annotation algorithms perform not very ideal, therefore, we propose an image annotation framework based on multimodule deep transfer learning method, aiming at the application of deep convolution neural network technology to optimize the neural network parameters, and improve the tagging accuracy in detail. The proposed multimodule deep learning annotation framework uses two stages to optimize neural network parameters: (1) using deep neural network to optimize the single-mode condition parameters; (2) using the phase correlation to realize the multimodule state, the experiment of the optimal combination of public data sets shows that the scheme can effectively improve the performance of image annotation.
Keywords: High-level Semantic, Deep Transfer Learning, Neural Network, Multimodule State, Image Annotation
REFERENCES
[1] Tylecek, R., and R. Fisher (2018) Consistent semantic annotation of outdoor datasets via 2D/3D label transfer, Sensors 8(7). doi: 10.3390/s18072249
[2] Xue, X., W. Zhang, J. Zhang, et al. (2011) Correlative multi-label multi-instance image annotation, International Conference on Computer Vision. IEEE Computer Society, 651658. doi: 10.1109/ICCV.2011.6126300
[3] Zhang, D., M. M. Islam, and G. Lu (2012) A review on automatic image annotation techniques, Pattern Recognition 45(1), 346362. doi: 10.1016/j.patcog.2011. 05.013
[4] Chelgani, S. C., S. S. Matin, and S. Makaremi (2016) Modeling of free swelling index based on variable importance measurements of parent coal properties by random forest method, Measurement 94, 416422. doi: 10.1016/j.measurement.2016.07.070
[5] Zhu, S., and X. Tan (2011) A novel automatic image annotation method based on multi-instance learning, Procedia Engineering 15(15), 34393444. doi: 10. 1016/j.proeng.2011.08.644
[6] Ravishankar, H., P. Sudhakar, R. Venkataramani, et al. (2016) Understanding the Mechanisms of Deep Transfer Learning for Medical Images, International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis. Springer International Publishing, 188196.
[7] Matin, S. S., and S. C. Chelgani (2016) Estimation of coal gross calorific value based on various analyses by random forest method, Fuel 177, 274278. doi: 10. 1016/j.fuel.2016.03.031
[8] Zhang, Y. N., and M. Q. An (2017) Deep learning- and transfer learning-based super resolution reconstruction from single medical image, Journal of Healthcare Engineering. doi: 10.1155/2017/5859727
[9] Zhu, S., Z. Shi, C. Sun, et al. (2015) Deep neural network based image annotation, Pattern Recognition Letters 65(4), 103108. doi: 10.1016/j.patrec.2015. 07.037
[10] Sun, C., S. Zhu and Z. Shi (2015) Image annotation via deep neural network, Iapr International Conference on Machine Vision Applications. IEEE. doi: 10.1109/ MVA.2015.7153244
[11] Huang, Y., W. Wang, L. Wang, et al. (2014) Multi-task deep neural network for multi-label learning, IEEE International Conference on Image Processing. IEEE, 28972900. doi: 10.1109/ICIP.2013.6738596
[12] Yu, Y., H. Lin, J. Meng, et al. (2017) Deep transfer learning for modality classification of medical images, Information 8(3), 91. doi: 10.3390/info8030091
[13] Gaur, U., M. Kourakis, E. Newman-Smith, et al. (2016) Membrane segmentation via active learning with deep networks, IEEE International Conference on Image Processing, IEEE, 19431947. doi: 10.1109/ICIP.2016. 7532697
[14] Liu, T., S. Xie, J. Yu, etal. (2017) Classification of thyroid nodules in ultrasound images using deep model based transfer learning and hybrid features, IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE. doi: 10.1109/ICASSP.2017. 7952290
We use cookies on this website to personalize content to improve your user experience and analyze our traffic. By using this site you agree to its use of cookies.
