Abstract
Image noise has a serious impact on image transmission, image data collection and image processing. Image noise is mainly denoised by additive noise and multiplicative noise. Firstly, based on the analysis that random noise has a direct impact on image recognition accuracy and registration performance, aiming at capturing and controlling the interaction between reinforcement learning agents and noise environment, an image recognition model based on noise stimulation is proposed, which will help capture and analyze random noise. Then, in order to construct a complete image data set and its linear combination of transfer process, a denoising algorithm based on reinforcement learning is proposed, which uses the sparse vector based on noise obtained by reinforcement learning to represent a given noise signal. Finally, based on image denoising based on reinforcement learning, an image registration optimization mechanism is proposed by maximizing the similarity measure between two images or minimizing the distance measure to find the coordinate correspondence between images. The simulation results show that the proposed algorithm is credible, effective and efficient in terms of consistency of noise analysis, accuracy of correspondence between feature points, registration error and computational utility of the algorithm.
Similar content being viewed by others
References
Chen K, Grapiglia GN, Yuan J et al (2019) Improved optimization methods for image registration problems. 80(2):305–336
Cun X, Pun CM, Gao H (2018) Applying stochastic second-order entropy images to multi-modal image registration. Signal Process Image Commun 65:201–209
El-Gamal EZA, Elmogy M, Atwan A (2016) Current trends in medical image registration and fusion. Egyptian Informatics Journal 17(1):99–124
Gou S, Chen L, Yu G et al (2018) Large-Deformation Image Registration of CT-TEE for Surgical Navigation of Congenital Heart Disease. Computational and Mathematical Methods in Medicine 2018:1–11
Heinrich MP, Simpson IJA, Papie BW et al (2016) Deformable image registration by combining uncertainty estimates from super voxel belief propagation. Med Image Anal 27:57–71
Jin Y, Xia K (2018) Intelligent Location and Recognition Mechanism of Abnormal Point of Medical Image Based on Reliable Transmission in Medial CT Scanner Local Big Data Networks. Journal of Medical Imaging & Health Informatics 8(3):609–617
Johnson B, Thomas N, Rani J S (2017) An FPGA Based High throughput Discrete Kalman Filter Architecture for Real-Time Image Denoising. International Conference on VLSI Design & International Conference on Embedded Systems
Kilicarslan A, Grossman RG, Contrerasvidal JL (2016) A robust adaptive denoising framework for real-time artifact removal in scalp EEG measurements. J Neural Eng 13(2):026013
Li Z, Tang J et al (2018) Weakly Supervised Deep Metric Learning for Community-Contributed Image Retrieval. IEEE Transactions on Multimedia 17(11):1989–1999
Maintz JB, Viergever MAA (2016) Survey of Medical Image Registration. Computer & Digital Engineering 33(1):140–144
Paquin D, Levy D, Schreibmann E et al (2017) Multiscale image registration. Mathematical Biosciences & Engineering Mbe 3(2):389–418
Paquin D, Levy D, Xing L (2017) Hybrid multiscale landmark and deformable image registration. Math Biosci Eng 4(4):711–737
Sun YJ, Yun JC, Kim PW (2017) Kernel design for real-time denoising implementation in low-resolution images. J Real-Time Image Proc (12):1–17
Yang X, Kwitt R, Styner M et al (2017) Quicksilver: Fast predictive image registration – A deep learning approach. Neuroimage 158:378
Yue W, Miao Q, Ma W et al (2018) PSOSAC: Particle Swarm Optimization Sample Consensus Algorithm for Remote Sensing Image Registration. IEEE Geoscience & Remote Sensing Letters 15(2):242–246
Acknowledgements
The authors would like to thank the support from Fujian Provincial Education Department Funds Science and Technology Projects (NO. 2017 JAT170321).
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
Ke, J., Zhang, Z. & Yingze, Y. Image registration optimization mechanism based on reinforcement learning and real time denoising. Multimed Tools Appl 79, 9489–9508 (2020). https://doi.org/10.1007/s11042-019-07914-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-019-07914-5