Connected pixels-based image smoothing filter

Abstract

Digital image processing heavily relies on the connectivity of pixels, as it is a vital component for accurate object identification and analysis within an image. Grouping together pixels with similar features such as colour and intensity, allows for the formation of meaningful patterns or objects, which is essential for object recognition and segmentation. This approach is particularly valuable in photogrammetric imaging, video surveillance, deep learning as it facilitates the isolation of regions of interest and object tracking. Image smoothing is also a crucial aspect in enhancing visual quality by reducing noise and enhancing details, especially in applications such as aerial mapping, medical imaging, video compression, image resizing and computer vision. The absence of connected pixels and image smoothing would make image processing tasks more challenging and less reliable, making them fundamental to digital image processing and critical to various applications in diverse fields. This paper introduces a novel image smoothing filter called Connected Pixels Based Image Smoothing Filter (CPF), which is based on gray connected pixels. The success of the CPF was compared to that of the Non-Local Means Filter (NLMF) in terms of Structural Similarity Index (SSIM) for the same Mean Squared Error (MSE). The experimental results showed that CPF has a better ability to preserve image details compared to NLMF.

Keywords

• Connected Pixels
• Image Smoothing
• Bezier Search Optimization Algorithm

References

1. Perri, S., Spagnolo, F., & Corsonello, P. (2020). A parallel connected component labeling architecture for heterogeneous systems-on-chip. Electronics, 9(2), 292.
2. Chen, J., Nonaka, K., Sankoh, H., Watanabe, R., Sabirin, H., & Naito, S. (2018). Efficient parallel connected component labeling with a coarse-to-fine strategy. IEEE Access, 6, 55731-55740.
3. Narasimhan, R. S., Vengadarajan, A., & Ramakrishnan, K. R. (2017, March). Design of connected component analysis-based clustering of CFAR image in pulse Doppler radars. In 2017 IEEE Aerospace Conference (pp. 1-6). IEEE.
4. Civicioglu, P., & Alci, M. (2003, December). CCII based analog circuit for the edge detection of MRI images. In 2003 46th Midwest Symposium on Circuits and Systems (Vol. 1, pp. 341-344). IEEE.
5. Jeong, J. W., Lee, G. B., Lee, M. J., & Kim, J. G. (2016). A single-pass connected component labeler without label merging period. Journal of Signal Processing Systems, 84(2), 211-223.
6. Besdok, E. (2004). Impulsive noise suppression from images with a modified two-step iterative-median filter. Journal of Electronic Imaging, 13(4), 714-719.
7. Çivicioğlu, P., & Alçı, M. (2004). Edge detection of highly distorted images suffering from impulsive noise. AEU-International Journal of Electronics and Communications, 58(6), 413-419.
8. Wang, Y., & Bhattacharya, P. (1996). On parameter-dependent connected components of gray images. Pattern Recognition, 29(8), 1359-1368.

9. Beşdok, E., Çivicioğlu, P., & Alçı, M. (2004). Impulsive noise suppression from highly corrupted images by using resilient neural networks. In Artificial Intelligence and Soft Computing-ICAISC 2004: 7th International Conference, Zakopane, Poland, June 7-11, 2004. Proceedings 7 (pp. 670-675). Springer Berlin Heidelberg.
10. Wang, Y., & Bhattacharya, P. (1996, November). Gray connected components and image segmentation. In Applications of Digital Image Processing XIX (Vol. 2847, pp. 118-129). SPIE.
11. Donato, M., Hansen, K., Kalavakuru, P., Kirchgessner, M., Kuster, M., Porro, M., ... & Turcato, M. (2017). First functionality tests of a 64× 64 pixel DSSC sensor module connected to the complete ladder readout. Journal of Instrumentation, 12(03), C03025.
12. Tang, J. W., Shaikh-Husin, N., Sheikh, U. U., & Marsono, M. N. (2018). A linked list run-length-based single-pass connected component analysis for real-time embedded hardware. Journal of Real-Time Image Processing, 15, 197-215.
13. Ray, V., & Goyal, A. (2016, January). Image-based fuzzy c-means clustering and connected component labeling subsecond fast fully automatic complete cardiac cycle left ventricle segmentation in multi frame cardiac MRI images. In 2016 International Conference on Systems in Medicine and Biology (ICSMB) (pp. 36-40). IEEE.
14. Ito, Y., & Nakano, K. (2010). Low-latency connected component labeling using an FPGA. International Journal of Foundations of Computer Science, 21(03), 405-425.
15. Chang, W. Y., & Chiu, C. C. (2014, June). Directional Connected Components Algorithm Based on Gradient Information. In 2014 International Symposium on Computer, Consumer and Control (pp. 280-283). IEEE.
16. Beşdok, E., & Yüksel, M. E. (2005). Impulsive noise suppression from images with Jarque-Bera test based median filter. AEU-International Journal of Electronics and Communications, 59(2), 105-110.
17. Çivicioğlu, P., Alçı, M., & Beşdok, E. (2004). Impulsive noise suppression from images with the noise exclusive filter. EURASIP Journal on Advances in Signal Processing, 2004(16), 2434-2440
18. Yeong, L. S., Ang, L. M., & Seng, K. P. (2010, July). Efficient connected component labelling using multiple-bank memory storage. In 2010 3rd International Conference on Computer Science and Information Technology (Vol. 9, pp. 75-79). IEEE.
19. Gunen, M. A., Civicioglu, P., & Beşdok, E. (2016). Differential search algorithm-based edge detection. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 41(B7), 667-670
20. Bieniek, A., & Moga, A. (1998). A connected component approach to the watershed segmentation. Computational Imaging and Vision, 12, 215-222.
21. Sun, Y., Sun, C., & Wang, W. (2000, August). Color images segmentation using new definition of connected components. In WCC 2000-ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000 (Vol. 2, pp. 863-868). IEEE.
22. Ma, D., Liu, S., & Liao, Q. (2017). Run-based connected components labeling using double-row scan. In Image and Graphics: 9th International Conference, ICIG 2017, Shanghai, China, September 13-15, 2017, Revised Selected Papers, Part III 9 (pp. 264-274). Springer International Publishing.
23. Asano, T. (2012). In-place algorithm for erasing a connected component in a binary image. Theory of Computing Systems, 50(1), 111-123.
24. Bekhtin, Y. S., Gurov, V. S., & Zavalishin, S. S. (2015, June). A run equivalence algorithm for parallel connected component labeling on CPU. In 2015 4th Mediterranean Conference on Embedded Computing (MECO) (pp. 276-279). IEEE.
25. Yunfeng, G., Feiyang, W., & Xiaotian, H. (2014, August). Connected components labeling algorithm based on run-length table searching. In 2014 9th International Conference on Computer Science & Education (pp. 700-704). IEEE.
26. Rasmusson, A., Sørensen, T. S., & Ziegler, G. (2013). Connected components labeling on the GPU with generalization to voronoi diagrams and signed distance fields. In Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings, Part I 9 (pp. 206-215). Springer Berlin Heidelberg.
27. Il, H. J., Kim, H. K., & Oh, W. G. (2015, January). Fast text line detection by finding linear connected components on Canny edge image. In 2015 21st Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV) (pp. 1-4). IEEE.
28. Kowalczyk, M., & Kryjak, T. (2021, September). A Connected Component Labelling algorithm for a multi-pixel per clock cycle video stream. In 2021 24th Euromicro Conference on Digital System Design (DSD) (pp. 43-50). IEEE.
29. Rouabeh, H., Abdelmoula, C., & Masmoudi, M. (2016, December). A new efficient connected component labeling algorithm and its VHDL circuit. In 2016 28th International Conference on Microelectronics (ICM) (pp. 105-108). IEEE.
30. Bailey, D. G. (2020, November). History and evolution of single pass connected component analysis. In 2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ) (pp. 1-6). IEEE.
31. Wang, Y., & Bhattacharya, P. (1996, October). Image analysis and segmentation using gray connected components. In 1996 IEEE International Conference on Systems, Man and Cybernetics. Information Intelligence and Systems (Cat. No. 96CH35929) (Vol. 1, pp. 444-449). IEEE.
32. Kang, S. M., Kim, J. H., Yuan, Z., Song, S. H., & Cho, J. D. (2014, June). A fast region expansion labeling of connected components in binary image. In The 18th IEEE International Symposium on Consumer Electronics (ISCE 2014) (pp. 1-2). IEEE.
33. Flatt, H., Blume, S., Hesselbarth, S., Schunemann, T., & Pirsch, P. (2008, July). A parallel hardware architecture for connected component labeling based on fast label merging. In 2008 International Conference on Application-Specific Systems, Architectures and Processors (pp. 144-149). IEEE.
34. Krämer, M., Afzal, M. Z., Bukhari, S. S., Shafait, F., & Breuel, T. M. (2012, November). Robust stereo correspondence for documents by matching connected components of text-lines with dynamic programming. In Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012) (pp. 734-737). IEEE.
35. Yapa, R. D., & Koichi, H. (2007, March). A connected component labeling algorithm for grayscale images and application of the algorithm on mammograms. In Proceedings of the 2007 ACM symposium on Applied computing (pp. 146-152).
36. Zhang, D., Ma, H., & Pan, L. (2019). A gamma-signal-regulated connected components labeling algorithm. Pattern Recognition, 91, 281-290.
37. Shim, J., Yoon, M., & Lee, Y. (2019). Feasibility of fast non local means filter in pediatric chest x-ray for increasing of pulmonary nodule detectability with 3D printed lung nodule phantom. Journal of Radiological Protection, 39(3), 872-889
38. Shim, J., Yoon, M., & Lee, Y. (2018). Feasibility of newly designed fast non local means (FNLM)-based noise reduction filter for X-ray imaging: A simulation study. Optik, 160, 124-130.
39. Civicioglu, P., & Besdok, E. (2019). Bernstain-search differential evolution algorithm for numerical function optimization. Expert Systems with Applications, 138, 112831.
40. Perona, P., & Malik, J. (1990). Scale-space and edge detection using anisotropic diffusion. IEEE Transactions on pattern analysis and machine intelligence, 12(7), 629-639.
41. Civicioglu, P., & Besdok, E. (2022). Contrast stretching based pansharpening by using weighted differential evolution algorithm. Expert Systems with Applications, 208, 118144.