Robust k-means pixel clustering via dominant pixel filtering

Year 2025, Volume: 9 Issue: 4, 630 - 642, 08.10.2025

Pınar Çivicioğlu

https://doi.org/10.31127/tuje.1719011

Abstract

IImage smoothing is a fundamental operation in various image processing domains, including Remote Sensing, Photogrammetry, and Computer Vision, primarily aimed at simplifying image texture. It is frequently employed to limit fluctuations in local contrast within high-resolution imagery. A critical requirement for any smoothing algorithm is that it must be performed without sacrificing significant structural details, such as edges and corners. Conventional methods, such as the Bilateral filter, often introduce blurring artifacts or impose a substantial computational burden. This paper introduces the Dominant-Pixels based Mean Filter (DMF), a highly efficient method developed for detail-preserving image smoothing. The DMF exhibits a low computational burden and reduced algorithmic complexity compared to classical methods, rendering it computationally efficient. The methodology is based on a conditional mean calculation within a local window, where only pixels with intensities similar to the central pixel, as determined by a threshold, are included in the averaging operation. This mechanism effectively preserves local discontinuities that correspond to fine details. Furthermore, this paper investigates the potential of the DMF to enhance the performance of pixel-based image clustering algorithms. Analysis of the experimental results confirms that the DMF achieves statistically successful outcomes in image smoothing tasks.

Keywords

Image Smoothing , Dominant Pixels , Bilateral Filter , Image Clustering

References

• Aydın, M., & Kurnaz, T. F. (2023). An alternative method for the particle size distribution: Image processing. Turkish Journal of Engineering, 7(2), 108-115. https://doi.org/10.31127/tuje.1053462
• Karataş, L., Alptekin, A., Karabacak, A., & Yakar, M. (2022). Detection and documentation of stone material deterioration in historical masonry buildings using UAV photogrammetry: A case study of Mersin Sarisih Inn. Mersin Photogrammetry Journal, 4(2), 53-61. https://doi.org/10.31127/tuje.637050
• Şasi, A., & Yakar, M. (2017). Photogrammetric modelling of Sakahane Masjid using an unmanned aerial vehicle. Turkish Journal of Engineering, 1(2), 82-87. https://doi.org/10.31127/tuje.316675
• Marangoz, A. M., Sefercik, U. G., & Yüce, D. (2020). Three-dimensional earth modelling performance analysis of Gokturk-2 satellite. Turkish Journal of Engineering, 4(3), 164-168. https://doi.org/10.31127/tuje.650989
• Zeng, L. L., et al. (2023). Edge-aware image smoothing via weighted sparse gradient reconstruction. Signal Image and Video Processing, 17(8), 4285-4293.
• Yu, X. B., et al. (2023). Image edge smoothing method for light-field displays based on joint design of optical structure and elemental images. Optics Express, 31(11), 18017-18025.
• Yang, Y., et al. (2023). Soft clustering based on high- and low-level features for image smoothing. Journal of Electronic Imaging, 32(1), 013028. https://doi.org/10.1117/1.JEI.32.1.013028
• Garg, B., & Sharma, G. K. (2016). A quality-aware energy-scalable Gaussian smoothing filter for image processing applications. Microprocessors and Microsystems, 45, 1-9.
• Toprak, N., & Yalman, Y. (2025). Ship detection from optical satellite images using convolutional neural networks. Turkish Journal of Engineering, 9(2), 342-353.
• He, Y., et al. (2023). Laser ultrasonic imaging of submillimeter defect in a thick waveguide using entropy-polarized bilateral filtering and minimum variance beamforming. Mechanical Systems and Signal Processing, 186, 109863.
• Zhao, C. H., Zhu, W. X., & Feng, S. (2022). Hyperspectral image classification based on kernel-guided deformable convolution and double-window joint bilateral filter. IEEE Geoscience and Remote Sensing Letters, 19, 1-5.
• Yao, R. H., et al. (2022). A compact hardware architecture for bilateral filter with the combination of approximate computing and look-up table. IEEE Transactions on Circuits and Systems II-Express Briefs, 69(7), 3324-3328.
• Wu, L. S., et al. (2022). Deep bilateral filtering network for point-supervised semantic segmentation in remote sensing images. IEEE Transactions on Image Processing, 31, 7419-7434.
• Wang, X. Q., et al. (2022). Using bilateral filtering and autoencoder to defend against adversarial attacks for object detection. Journal of Electronic Imaging, 31(4).
• Riya, B., Gupta, & Lamba, S. S. (2022). Structure-aware adaptive bilateral texture filtering. Digital Signal Processing, 123, 103386.
• Hu, Q. B., et al. (2022). Hyperspectral image classification based on bilateral filter with multispatial domain. IEEE Geoscience and Remote Sensing Letters, 19, 5503805.
• Joel, T., & Sivakumar, R. (2021). Nonsubsampled contourlet transform with cross-guided bilateral filter for despeckling of medical ultrasound images. International Journal of Imaging Systems and Technology, 31(2), 763-777.
• Das, A., & Shylaja, S. S. (2021). CDDA, color-dominant deep autoencoder for faster and efficient bilateral image filtering. Signal Image and Video Processing, 15(6), 1189-1195.
• Yang, H., Zhang, Z. B., & Guan, Y. J. (2019). Rolling bilateral filter-based text image deblurring. Visual Computer, 35(11), 1627-1640.
• Karam, C., & Hirakawa, K. (2018). Monte-Carlo acceleration of bilateral filter and non-local means. IEEE Transactions on Image Processing, 27(3), 1462-1474.
• Jomaa, H., et al. (2018). Denoising of dynamic PET images using a multi-scale transform and non-local means filter. Biomedical Signal Processing and Control, 41, 69-80.
• Li, H. F., et al. (2017). Multifocus image fusion via fixed window technique of multiscale images and non-local means filtering. Signal Processing, 138, 71-85.
• Wang, X. T., et al. (2016). Iterative non-local means filter for salt and pepper noise removal. Journal of Visual Communication and Image Representation, 38, 440-450.
• Ni, W. P., & Gao, X. B. (2016). Despeckling of SAR image using generalized guided filter with Bayesian nonlocal means. IEEE Transactions on Geoscience and Remote Sensing, 54(1), 567-579.
• Chan, S. H., Zickler, T., & Lu, Y. M. (2014). Monte Carlo non-local means, random sampling for large-scale image filtering. IEEE Transactions on Image Processing, 23(8), 3711-3725.
• Li, W. T., et al. (2022). A SAR change detection method based on an iterative guided filter and the log mean ratio. Remote Sensing Letters, 13(7), 663-671.
• Xiang, S. M., & Zhang, H. J. (2016). Efficient edge-guided full-waveform inversion by Canny edge detection and bilateral filtering algorithms. Geophysical Journal International, 207(2), 1049-1061.
• Zang, Y., Huang, H., & Zhang, L. (2015). Guided adaptive image smoothing via directional anisotropic structure measurement. IEEE Transactions on Visualization and Computer Graphics, 21(9), 1015-1027.
• Yilmaz, N., & Sahiner, A. (2021). Generalization of hyperbolic smoothing approach for non-smooth and non-Lipschitz functions. Journal of Industrial and Management Optimization, 18(6), 4511-4526.
• Sahiner, A., & Yilmaz, N. (2024). Double stranded smoothing technique in energy minimization problems. Numerical Algebra, Control and Optimization, 14(1), 80-88.
• Kusak, L., Unel, F. B., Alptekin, A., Celik, M. O., & Yakar, M. (2021). Apriori association rule and K-means clustering algorithms for interpretation of pre-event landslide areas and landslide inventory mapping. Open Geosciences, 13(1), 1226-1244.
• Civicioglu, P., & Besdok, E. (2024). Colony-based search algorithm for numerical optimization. Applied Soft Computing, 151, 111162.
• Civicioglu, P., & Besdok, E. (2023). Bernstein-Levy differential evolution algorithm for numerical function optimization. Neural Computing & Applications, 35(9), 6603-6621.
• Civicioglu, P., & Besdok, E. (2021). Bezier search differential evolution algorithm for numerical function optimization: A comparative study with CRMLSP, MVO, WA, SHADE and LSHADE. Expert Systems with Applications, 165, 113875.
• Civicioglu, P., & Besdok, E. (2019). Bernstain-search differential evolution algorithm for numerical function optimization. Expert Systems with Applications, 138, 112831.
• Civicioglu, P., & Besdok, E. (2018). A plus evolutionary search algorithm and QR decomposition based rotation invariant crossover operator. Expert Systems with Applications, 103, 49-62.