Adaptif 2-D LMS Filtre Gömülü Kenar Algılama Uygulaması

Year 2020, Ejosat Special Issue 2020 (HORA), 343 - 351, 15.08.2020

Ufuk Paralı

https://doi.org/10.31590/ejosat.780103

Abstract

Bu çalışmada, iki-boyutlu en küçük ortalama kareler (TDLMS) adaptif filtresinin etkisi çeşitli kenar algılama sistemlerinin içerisine gömülerek tartışılmıştır. TDLMS ve kenar algılama modülleri sistem şeması içerisinde seri sırada çalışacak şekilde yerleştirilmiştir. TDLMS algoritması birçok değişik görüntü işleme uygulamalarında yaygın olarak kullanılmaktadır. Özellikle filtre katsayılarının herhangi bir öncül varsayıma ihtiyaç duymadan güncellenebiliyor olması, TDLMS filtresine iki boyutlu sinyal işleme uygulamalarında çok üstün avantajlar sağlamaktadır. Bu çalışmada, literatürde sıkça kullanılan Canny, Sobel, Prewitt, Roberts ve LoG kenar algılama algoritmaları üzerindeki TDLMS’in sağladığı performans artışı incelenmiştir. Gerçekleştrilen benzetimlerde, TDLMS’in yüksek SNR değerine sahip görüntülerden ziyade, düşük SNR değerine sahip görüntülerde kenar algılama algoritmalarının performasında artışa sebep olduğu gözlemlenmiştir. Özellikle, düşük SNR durumunda Sobel, Prewitt ve Roberts algoritmalarındaki performans artışı, Canny ve LoG algoritmalarındaki performans artışına göre daha fazla olmuştur. Etkisi fazla olmamakla birlikte yüksek SNR durumunda ise Canny algoritmasındaki performans artışı Sobel, Prewitt, Roberts ve LoG algoritmalarına kıyasla daha fazla olduğu görülmüştür.

Keywords

2-D Adaptif Fitre, TDLMS, Kenar Algılama Algoritması, Canny, Sobel, Prewitt, Roberts, LoG

Adaptive 2-D LMS Filter Embedded Edge Detection Application

Abstract

In this numerical study the effect of embedding two-dimensional least mean square (TDLMS) adaptive filter into various edge detection systems is discussed. TDLMS and edge detection modules are arranged in the system scheme in a manner such that they work sequentially. TDLMS algorithm is commonly used in many various image processing applications. Due to its ability of updating filter coefficients without needing any a priori assumptions, TDLMS provides superior advantegeous in 2-D signal processing applications. We investigated the performance increment of TDLMS especially on the commonly used edge detection algortihms in the literature such as Canny, Sobel, Prewitt, Roberts and LoG (Laplacian of Gaussian). It is observed that embedding TDLMS is particularly useful in edge detection for low SNR images comparing to high SNR images. The simulation results clearly show TDLMS filter provides significant improvement for the edge detection implementation on a relatively lower SNR case comparing to a higher SNR case. Especially, TDLMS embedded Sobel, Prewitt and Roberts implementations have relatively better results than TDLMS embedded Canny and LoG implementations for a low SNR image. On the other hand, for relatively higher SNR case, embedding TDLMS filter into the edge detection system does not provide as much significant improvement as in relatively lower SNR case. But still, for a high SNR case, TDLMS embedded Canny implementation have relatively better results than TDLMS embedded Sobel, Prewitt, Roberts and LoG implementations.

Keywords

2-D Adaptive Filter, TDLMS, Edge Detection Algorithm, Canny, Sobel, Prewitt, Roberts, LoG

References

• Bae, T. W., Zhang, F., & Kweon, I. S. (2012). Edge directional 2D LMS filter for infrared small target detection. Infrared Physics & Technology, 55(1), 137-145.
• Canny, J. (1986). A computational approach to edge detection. IEEE Transactions on pattern analysis and machine intelligence, (6), 679-698.
• Gupta, S., & Mazumdar, S. G. (2013). Sobel edge detection algorithm. International journal of computer science and management Research, 2(2), 1578-1583.
• Hadhoud, M. M., & Thomas, D. W. (1987). Image averaging using the adaptive two-dimensional least mean square filter. Journal of Modern Optics, 34(1), 79-89.
• Hadhoud, M. M., & Thomas, D. W. (1988). The two-dimensional adaptive LMS (TDLMS) algorithm. IEEE Transactions on Circuits and Systems, 35(5), 485-494.
• Hadhoud, M. M., & Thomas, D. W. (1989). The effect of the image local mean on the two-dimensional least mean square algorithm weight convergence. Journal of Modern Optics, 36(4), 545-549.
• Haralick, R. M., & Shapiro, L. G. (1992). Computer and robot vision (Vol. 1, pp. 28-48). Reading: Addison-Wesley.
• Kaur, A., Malhotra, R., & Kaur, R. (2015, June). Performance evaluation of non-iterative adaptive median filter. In 2015 IEEE International Advance Computing Conference (IACC) (pp. 1117-1121). IEEE.
• Kumar, M., & Saxena, R. (2013). Algorithm and technique on various edge detection: A survey. Signal & Image Processing, 4(3), 65.
• Lin, J. N., Nie, X., & Unbehauen, R. (1993). Two-dimensional LMS adaptive filter incorporating a local-mean estimator for image processing. IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, 40(7), 417-428.
• Liu, T. S., Liu, R. X., & Pan, S. W. (2014). Improved Canny algorithm for edge detection of core image. The Open Automation and Control Systems Journal, 6(1).
• Plataniotis, K. N., & Venetsanopoulos, A. N. (2013). Color image processing and applications. Springer Science & Business Media.
• Praneeth, C., Rao, S., & Srinivas, K. (2011). Image Edge Detection using Adaptive Filter. International Journal of ComputerScience and Information Technologies, 2(4), 1581-1583.
• Pratt, W. K. (1991). Digital image processing John Wiley & Sons. Inc., New York.
• Shrivakshan, G. T., & Chandrasekar, C. (2012). A comparison of various edge detection techniques used in image processing. International Journal of Computer Science Issues (IJCSI), 9(5), 269.
• Smith, C., & Campbell, D. R. (1995). The two-dimensional LMS algorithm and its application to sub-band filtering.
• Soni, T., Rao, B. D., Zeidler, J. R., & Ku, W. H. (1991, May). Enhancement of images using the 2-D LMS adaptive algorithm. In Proceedings IEEE International Conf. on Acoustics Speech and Signal Processing (Vol. 4, pp. 3029-3032).
• Trucco, E., & Verri, A. (1998). Introductory techniques for 3-D computer vision (Vol. 201). Englewood Cliffs: Prentice Hall.
• Verma, K., Singh, B. K., & Thoke, A. S. (2015). An enhancement in adaptive median filter for edge preservation. Procedia Computer Science, 48(C), 29-36.
• Yan, Y., Shin, W. I., Pang, Y. X., Meng, Y., Lai, J., You, C., ... & Pang, C. H. (2020). The first 75 days of novel coronavirus (SARS-CoV-2) outbreak: Recent advances, prevention, and treatment. International journal of environmental research and public health, 17(7), 2323.