Table2indicates the improvement in performance for higher order filter banks in comparison with recently available methods given in [4,5,11,18–20]. For N = 48, keeping Ep, Es, and Etalmost at the same level, the percentage improvements in As
are 17.27%, 15.38%, 7.63%, 23.04%, 2.68%, and 4.02%, respectively. In addition to As, large amount of reduction in PRE are noticed using proposed method which are 65.62%, 38.8%, 67.10%, 99.53%, 7.55%, and 42.7%, respectively.
5 Conclusions
In this paper, a new algorithm based on multi-swarm PSO is presented to design the linear phase prototype filter of QMF bank. To avoid the premature convergence in higher order filter design, SA is hybridized with multi-swarm PSO. The perfor- mance of the proposed SAMCPSO algorithm is compared with that of different PSO variants, ABC, BAT algorithm, and other well-known recently reported methods.
Table1PerformanceparametersobtainedusingproposedSAMCPSOforhigherorderfilterbanks NαβEpEsEteamPRE(dB)As(dB) 480.810−063.9×10−091.6×10−081.4×10−112.1×10−030.0030559.28 560.610−061.6×10−091.6×10−094.8×10−121.2×10−030.0013567.16 640.510−061.1×10−096.5×10−101.9×10−119.1×10−040.0010577.135 760.510−072.7×10−106.4×10−117.5×10−127.7×10−040.00096102.83 820.310−081.4×10−112.7×10−131.1×10−186.8×10−040.00082112.88 900.0110−091.4×10−119.6×10−147.7×10−195.7×10−040.00042119.46 1000.0110−092.9×10−128.6×10−142.7×10−194.8×10−040.00033122.24
Table2PerformancecomparisonofproposedSAMCPSOforhigherorder(N=48and82)filterbank NMethodAs(dB)eamEpEsEtPRE(dB) 48[4]50.55–2.54×10−096.62×10−083.87×10−270.0089 [5]51.38–2.70×10−088.91×10−081.93×10−220.0050 ABC[11]55.08–3.16×10−102.95×10−08–0.0093 BAT[18]48.180.16184.40×10+029.00×10+026.490.6445 COABC[19]57.733.87×10−035.08×10−101.77×10−083.17×10−130.00331 CPSO[20]56.993.9×10−033.58×10−101.76×10−102.26×10−130.00534 SAMCPSO59.282.08×10−033.93×10−091.63×10−081.36×10−110.00306 82[4]96.16–1.31×10−138.30×10−132.01×10−200.0037 [5]96.54–3.37×10−114.18×10−122.07×10−140.0010 ABC[11],BAT[18],COABC[19],CPSO[20]Nofeasiblesolutionfound SAMCPSO112.886.86×10−041.37×10−112.66×10−131.37×10−180.000829
Based on the significantly encouraging simulation results, it can be concluded that the proposed SAMCPSO method results in a better quality prototype filter. Fur- ther, SAMCPSO algorithm exhibits a better balancing ability between exploration and exploitation, and it can be applied effectively in solving other complex real-life problems. The design of multiplier-less two-dimensional digital filters and higher order Hilbert transformer using the proposed algorithm are left as future work.
References
1. Jain V, Crochiere R (1984) Quadrature mirror filter design in time domain. IEEE Trans Acoust Speech Signal Process 32(2):353–361
2. Rafi SM, Kumar A, Singh GK (2013) An improved particle swarm optimization method for multirate filter bank design. J Franklin Inst 350(4):757–769
3. Kumar A, Rafi SM, Singh GK (2012) A hybrid method for designing linear-phase quadrature mirror filter bank. Digit Signal Process 22(3):453–462
4. Kuldeep B, Singh VK, Kumar A, Singh GK (2015) Design of two-channel filter bank using nature inspired optimization based fractional derivative constraints. ISA Trans 54:101–116 5. Baderia K, Kumar A, Singh GK (2015) Design of quadrature mirror filter bank using polyphase
components based on optimal fractional derivative constraints. Int J Electron Commun (AEĩ) 69(9):1254–1264
6. Sahu OP, Soni MK, Talwar IM (2006) Marquardt optimization method to design two-channel quadrature mirror filter banks. Digital Signal Process 16(6):870–879
7. Dhabal S, Venkateswaran P (2017) An efficient gbest-guided Cuckoo Search algorithm for higher order two channel filter bank design. Swarm Evol Comput 33:68–84
8. Upender J, Gupta CP, Singh GK (2010) Design of two-channel quadrature mirror filter bank using particle swarm optimization. Signal Process 20(2):304–313
9. Misra D, Dhabal S, Chakrabarti R, Venkateswaran P (2012) Canonical signed digit repre- sentation of quadrature mirror filter using genetic algorithm. In: Proceedings of international conference on communications, devices and intelligent systems (CODIS), Kolkata, pp 65–68 10. Oliveira HA, Petraglia A, Petraglia M R (2009) Design of FIR quadrature mirror-image fil- ter banks using fuzzy adaptive simulated annealing. In: Proceedings of 13th digital signal processing workshop and 5th IEEE signal processing education workshop, Marco Island, pp 485–489
11. Agrawal SK, Sahu OP (2015) Artificial bee colony algorithm to design two-channel quadrature mirror filter banks. Swarm Evol Comput 21:24–31
12. Kumar A, Singh GK, Anand RS (2010) An improved method for designing quadrature mirror filter banks via unconstrained optimization. J Math Model Algorithms 9(1):99–111
13. Jamili A, Mohammad AS, Moghaddam RT (2011) A hybrid algorithm based on particle swarm optimization and simulated annealing for a periodic job shop scheduling problem. Int J Adv Manuf Technol 54(1):309–322
14. Niu B, Zhu Y, He X, Wu H (2007) MCPSO: a multi-swarm cooperative particle swarm opti- mizer. Appl Math Comput 185(2):1050–1062
15. Zhang J, Ding X (2011) A multi-swarm self-adaptive and cooperative particle swarm optimiza- tion. Eng Appl Artif Intell 24(6):958–967
16. Xu X, Tang Y, Li J, Hua C, Guan X (2015) Dynamic multi-swarm particle swarm optimizer with cooperative learning strategy. Appl Soft Comput 29:169–183
17. Hasanzadeh M, Meybodi MR, Ebadzadeh MM (2013) Adaptive cooperative particle swarm optimizer. Appl Intell 39(2):397–420
18. Yang XS (2013) Bat algorithm and cuckoo search: a tutorial. In: Artificial intelligence, evolu- tionary computing and metaheuristics. Studies in computational intelligence, vol 427. Springer, Berlin, Heidelberg
19. Luo J, Wang Q, Xiao X (2013) A modified artificial bee colony algorithm based on converge- onlookers approach for global optimization. Appl Math Comput 219(20):10253–10262 20. Petrovi´c M, Miti´c M, Vukovi´c N, Miljkovi´c Z (2016) Chaotic particle swarm optimization
algorithm for flexible process planning. Int J Adv Manuf Technol 85(9–12):2535–2555
of Image Noise with the Combination of Median and Average Filters
Sayantan Gupta and Sukanya Roy
Abstract Noise in an image is undesirable to us as it disrupts and degrades the quality of the image. Noise removal is always a difficult task so as edge preservation when the intensity of the disrupted noise in the original image is high. In this paper, we proposed the Medav Filter which is a combination of mean and adaptive median filter that optimally adjusts the level of mask operations according to the noise density.
The median filter has good noise removal qualities, but its complexity is undesirable.
While the mean filter is unable to remove heavy tailored noise, we see its complexity increases in the presence of noise which is dependent upon the signal. In the Medav Filter, we proposed an algorithm to improve the peak signal–to-noise ratio (PSNR) which eventually improved the signal-to-noise ratio (SNR). We also described an efficient model for image restoration. The analysis of the algorithm and the Medav Filter shows that the complexity, as well as the performance, is improved as compared to the primitive filters.
Keywords Digital Image ProcessingãNoise removalãImpulse noise Dual threshold median filterãAverage filterãMedav FilterãImage restoration Computer vision