Abstract
This paper presents a theoretical comparison of early and late fusion methods. An initial discussion on the conditions to apply early or late (soft or hard) fusion is introduced. The analysis show that, if large training sets are available, early fusion will be the best option. If training sets are limited we must do late fusion, either soft or hard. In this latter case, the complications inherent in optimally estimating the fusion function could be avoided in exchange for lower performance. The paper also includes a comparative review of the fusion state of the art methods with the following divisions: early sensor-level fusion; early feature-level fusion; late score-level fusion (late soft fusion); and late decision-level fusion (late hard fusion). The main strengths and weaknesses of the methods are discussed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Usa, H., Escamilla-Ambrosio, P.J., Mort, N.: Hybrid kalman filter-fuzzy logic adaptive multisensor data fusion architectures (2003)
Vergara, L., Soriano, A., Safont, G., Salazar, A.: On the fusion of non-independent detectors. Digit. Signal Process. 50, 24–33 (2016)
Salazar, A., Safont, G., Vergara, L., Letters, E.V.-P.R.: Pattern recognition techniques for provenance classification of archaeological ceramics using ultrasounds. Pattern Recogn. Lett. 135, 441–450 (2020)
Jolliffe, I.T., Cadima, J.: Principal component analysis: a review and recent developments. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 374(2065), 20150202 (2016)
Soriano, A., Vergara, L., Ahmed, B., Salazar, A.: Fusion of scores in a detection context based on Alpha integration. Neural Comput. 27(9), 1983–2010 (2015)
Safont, G., Salazar, A., Vergara, L.: Multiclass Alpha integration of scores from multiple classifiers. Neural Comput. 31(4), 806–825 (2019)
Safont, G., Salazar, A., Vergara, L.: Vector score alpha integration for classifier late fusion. Pattern Recogn. Lett. 136, 48–55 (2020)
Salazar, A., Vergara, L., Vidal, E.: A proxy learning curve for the Bayes Classifier. Pattern Recogn. 136(109240), 1–14 (2023)
Kalman, R.E.: A new approach to linear filtering and prediction problems. J. Basic Eng. 82(1), 35–45 (1960)
Kalman, R.E., Bucy, R.S.: New results in linear filtering and prediction theory. J. Basic Eng. 83(1), 95–108 (1961)
Moon, S., Park, Y., Ko, D.W., Suh, I.H.: Multiple kinect sensor fusion for human skeleton tracking using Kalman filtering (2016)
Yazdkhasti, S., Sasiadek, J.Z.: Multi sensor fusion based on adaptive Kalman filtering. In: Dołęga, B., Głębocki, R., Kordos, D., Żugaj, M. (eds.) Advances in Aerospace Guidance, Navigation and Control, pp. 317–333. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-65283-2_17
Dempster, A.P.: A generalization of Bayesian inference. J. Roy. Stat. Soc.: Ser. B (Methodol.) 30(2), 205–232 (1968)
Coninx, A., et al.: Bayesian sensor fusion with fast and low power stochastic circuits. In: 2016 IEEE International Conference on Rebooting Computing, ICRC 2016 - November (2016)
Coué, C., Fraichard, T., Bessiere, P., Mazer, E.: Multi-sensor data fusion using Bayesian programming: an automotive application. In:IEEE International Conference on Intelligent Robots and Systems, vol. 1, pp. 141–146 (2002)
Amin, M., Akhoundi, A., Valavi, E.: Multi-sensor fuzzy data fusion using sensors with different characteristics (2010)
Stover, J.A., Hall, D.L., Gibson, R.E.: A fuzzy-logic architecture for autonomous multisensor data fusion. IEEE Trans. Industr. Electron. 43(3), 403–410 (1996)
McCulloch, W.S., Pitts, W.: A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys. 5(4), 115–133 (1943)
Quan, Y., Zhou, M.C., Luo, Z.: On-line robust identification of tool-wear via multi-sensor neural-network fusion. Eng. Appl. Artif. Intell. 11(6), 717–722 (1998)
Lee, J., Steele, C.M., Chau, T.: Swallow segmentation with artificial neural networks and multi-sensor fusion. Med. Eng. Phys. 31(9), 1049–1055 (2009)
Kańtoch, E.: Human activity recognition for physical rehabilitation using wearable sensors fusion and artificial neural networks. Comput. Cardiol. 2010(44), 1–4 (2017)
Hotelling, H.: Analysis of a complex of statistical variables into principal components. J. Educ. Psychol. 24(6), 417–441 (1933)
Abdi, H., Williams, L.J.: Principal component analysis. Wiley Interdiscip. Rev. Comput. Stat. 2(4), 433–459 (2010)
Hasan, M.M., Islam, N., Rahman, M.M.: Gastrointestinal polyp detection through a fusion of contourlet transform and neural features. J. King Saud Univ. – Comput. Inf. Sci. 34(3), 526–533 (2022)
Nasir, H., Stanković, V., Marshall, S.: Singular value decomposition based fusion for super-resolution image reconstruction. Signal Process Image Commun. 27(2), 180–191 (2012)
Zhao, X., Ye, B.: Singular value decomposition packet and its application to extraction of weak fault feature. Mech. Syst. Signal Process 70–71, 73–86 (2016)
Zhu, H., He, Z., Wei, J., Wang, J., Zhou, H.: Bearing fault feature extraction and fault diagnosis method based on feature fusion, vol. 21, no. 7, p. 2524 (2021)
Ye, X., Gao, W., Yan, Y., Osadciw, L.A.: Multiple tests for wind turbine fault detection and score fusion using two- level multidimensional scaling (MDS), vol. 7704, pp. 70–77 (2010)
Tian, G.Y., Taylor, D.: Colour image retrieval using virtual reality. In: Proceedings of the International Conference on Information Visualisation, vol. 2000-July, pp. 221–225 (2000)
Choo, J., Bohn, S., Nakamura, G.C., White, A.M., Park, H.: Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. In: Proceedings West Mark Ed Association Conference, pp. 177–188 (2012)
He, Q.-H., et al: Deep learning and radiomic feature-based blending ensemble classifier for malignancy risk prediction in cystic renal lesions (2023)
Wang, K., Xu, C., Li, G., Zhang, Y., Zheng, Y., Sun, C.: Combining convolutional neural networks and self-attention for fundus diseases identification (2023)
Jing, J., Wu, H., Sun, J., Fang, X., Zhang, H.: Multimodal fake news detection via progressive fusion networks. Inf. Process Manag. 60(1) (2023)
Salazar, A., Safont, G., Soriano, A., Vergara, L.: Automatic credit card fraud detection based on non-linear signal processing. In: ICCST, pp 207–2012 (2012). Article no. 6393560
Salazar, A., Safont, G., Vergara, L.: Surrogate techniques for testing fraud detection algorithms in credit card operations. ICCST 2014(6986987), 124–129 (2014)
Vergara, L., Salazar, A., Belda, J., Safont, G., Moral, S., Iglesias, S.: Signal processing on graphs for improving automatic credit card fraud detection. In: ICCST, pp. 1–6, Spain (2017)
Nanni, L., Lumini, A., Brahnam, S.: Likelihood ratio-based features for a trained biometric score fusion. Expert Syst. Appl. 38(1), 58–63 (2011)
Zafar, R., et al.: Prediction of human brain activity using likelihood ratio based score fusion. IEEE Access 5, 13010–13019 (2017)
Salazar, A., Safont, G., Rodriguez, A., Vergara, L.: Combination of multiple detectors for credit card fraud detection. In: ISSPIT, pp. 138–143, Limassol, Cyprus (2016)
Hube, J.P.: Neyman-Pearson biometric score fusion as an extension of the sum rule, vol. 6539, pp. 200–208 (2007)
Hammouche, R., Attia, A., Akhrouf, S.: Score level fusion of major and minor finger knuckle patterns based symmetric sum-based rules for person authentication. Evolving Syst. 13(3), 469–483 (2022)
Kittler, J., Hatef, M., Duin, R.P.W., Matas, J.: On combining classifiers. IEEE Trans. Pattern Anal. Mach. Intell. 20(3), 226–239 (1998)
Eskandari, M., Toygar, Ö.: Fusion of face and iris biometrics using local and global feature extraction methods. Signal Image Video Process 8(6), 995–1006 (2014)
Fierrez-Aguilar, J., et al.: A comparative evaluation of fusion strategies for multimodal biometric verification. Lect. Notes Comput. Sci. 2688, 830–837 (2003)
Arigbabu, O.A., et al.: Integration of multiple soft biometrics for human identification. Pattern Recogn. Lett. 68, 278–287 (2015)
Ruta, D., Gabrys, B.: Classifier selection for majority voting. Inf. Fusion 6(1), 63–81 (2005)
Jimenez, L.O., Morales-Morell, A.: Classification of hyperdimensional data based on feature and decision fusion approaches using projection pursuit, majority voting, and neural networks. IEEE Geosci. Remote. Sens. 37(3 I), 1360–1366 (1999)
Breiman, L.: Bagging predictors. Mach. Learn. 24(2), 123–140 (1996)
Ćwiklińska-Jurkowska, M.: Boosting, bagging and fixed fusion methods performance for aiding diagnosis. Biocybern. Biomed. Eng. 32(2), 17–31 (2012)
Dietterich, T.G.: Experimental comparison of three methods for constructing ensembles of decision trees: bagging, boosting, and randomization. Mach. Learn. 40(2), 139–157 (2000)
Ferreira, A.J., Figueiredo, M.A.T.: Boosting algorithms: a review of methods, theory, and applications. Ensemble Mach. Learn. 35–85 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Pereira, L.M., Salazar, A., Vergara, L. (2023). On Comparing Early and Late Fusion Methods. In: Rojas, I., Joya, G., Catala, A. (eds) Advances in Computational Intelligence. IWANN 2023. Lecture Notes in Computer Science, vol 14134. Springer, Cham. https://doi.org/10.1007/978-3-031-43085-5_29
Download citation
DOI: https://doi.org/10.1007/978-3-031-43085-5_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43084-8
Online ISBN: 978-3-031-43085-5
eBook Packages: Computer ScienceComputer Science (R0)