Effective T2FPSO-Based T2FSVM Scene Classification Algorithm

Authors

Shuqiong Xu, Dongguan Polytechnic, Dongguan 523808, China;
Conggui Yuan, Dongguan Polytechnic, Dongguan 523808, China;

Abstract

Abstract: A systematic design methodology of Type-2 Fuzzy Particle Swarm Optimization (T2FPSO) based Type-2 Fuzzy Support Vector Machine (T2FSVM) classification system was proposed for scene image to improve selectivity and robustness in the machine vision. In the novel classification system, the T2FSVM model was presented to realize a comprehensive learning of the correct class and show the superiority of the generalization capability for classification problem. Furthermore, in order to improve the performance of PSO on complex uncertain environments, the type-2 fuzzy concept was incorporated to PSO to construct T2FPSO searching algorithm, in which the interval type-2 fuzzy inertia weight was designed using an Interval Type-2 Fuzzy Logic System (IT2FLS). Experimental studies indicate that the T2FPSO-T2FSVM approach is effective to deal with uncertainties for scene classification, when scene images are corrupted by the hybrid noises or captured by different view angels and light conditions.

Recommended Citation

Xu, Shuqiong and Yuan, Conggui (2020) "Effective T2FPSO-Based T2FSVM Scene Classification Algorithm," Journal of System Simulation: Vol. 28: Iss. 12, Article 7.
DOI: 10.16182/j.issn1004731x.joss.201612007
Available at: https://dc-china-simulation.researchcommons.org/journal/vol28/iss12/7

First Page

2925

Revised Date

2016-05-24

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201612007

Last Page

2933

CLC

TP391.9

Recommended Citation

Xu Shuqiong, Yuan Conggui. Effective T2FPSO-Based T2FSVM Scene Classification Algorithm[J]. Journal of System Simulation, 2016, 28(12): 2925-2933.

DOI

10.16182/j.issn1004731x.joss.201612007