CC BY
1I EURASIAN|OUWt£L_
I
________,
EURASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES
Innovative Academy Research Support Center UIF = 8.3 | SJIF = 7.906 www.in-academy.uz
DEVELOPMENT OF A NOVEL EDGE DETECTION METHOD
Mirzayan Kamilov
Academician of the Academy of Sciences of Uzbekistan, Doctor of Technical Sciences, Professor, Digital Technologies and Artificial Intelligence Research Institute, Tashkent, Uzbekistan Khabibullo Nosirov Professor, Department of TV and Radio Broadcasting Systems, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, Uzbekistan Shohruh Begmatov Mukhriddin Arabboev Associate Professor, Department of TV and Radio Broadcasting Systems, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, Uzbekistan https://doi.org/10.5281/zenodo.10935595
ABSTRACT
ARTICLE INFO
Received: 31th March 2024 Accepted: 05th April 2024 Online: 06th April 2024
KEYWORDS Edge detection; single-channel; multichannel; RGB.
Accurate edge detection is fundamental to various image processing tasks. This paper presents a novel edge detection method that goes beyond traditional approaches. We propose a multichannel edge detection method that can be superior to the single-channel edge detection method.
Introduction
Edge detection is a fundamental technique in image processing [1-2] that plays a crucial role in tasks such as object recognition, image segmentation, and content analysis [3-4]. Traditional methods for edge detection rely on filtering techniques that can identify abrupt changes in intensity between pixels [5]. However, these methods can sometimes be affected by issues such as sensitivity to noise, blurring of edges, and difficulty in capturing intricate details. In recent years, various researchers have conducted research on the development of edge detection methods [6-9].
The proposed method
EURASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES
Innovative Academy Research Support Center UIF = 8.3 | SJIF = 7.906 www.in-academy.uz
Figure 1. Novel edge detection method
The given method in Figure 1 works with following mathematical expressions: rows cols 3 3
Rx = R * Sx = YY
i=0 j=0 33
i=0 j=0
I Is VV J Ll/lJ
= ^ ^ Rxk,l = ^ ^ Rk+i,k+j^xi,j
k=0 1=0 rows cols
(1)
Ry = R*Sy=^^ Rykil = ^ ^ Rk+iik+jSyiJ
k=0 1=0
(2)
rows cols
Re=^ ^Rek,l =Rxk,l\Ryk,l
k=0 1=0
(3)
edge = Re & Ge\Re & Be \ Ge & Be
(4)
Here: Sx, Sy - Sobel filters in matrix form, R- red channel pixel values in matrix form, Rx, Ry - Sx and Sy filtered matrixes, Re, Ge, Be - edges by R, G, B channels. Results
1B1
i b V*
iM
U gui p.
I
«"=: N
lb)
jypi*
S Évl
d)
e)
Figure 2. The proposed multichannel edge detection method
EURASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES
Innovative Academy Research Support Center UIF = 8.3 | SJIF = 7.906 www.in-academy.uz
According to the figure (see Figure 2) shown above each letter means the following: a) input image; b) Edge obtained using R channel; c) Edge obtained using G channel; d) Edge obtained using B channel; e) Final result
Figure 3. Edge obtained using single-channel edge detection method
Figure 4. Comparison of the obtained results on single-channel edge detection and the proposed method
It can be seen from Figure 4 that the proposed method detects edges that are missed by the single-channel approach.
Conclusion
EURASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES
Innovative Academy Research Support Center UIF = 8.3 | SJIF = 7.906 www.in-academy.uz
In this paper, a novel edge detection method is developed. Our method outperforms an existing single-channel edge detection method. In future work, we will focus on the following: the use of other popular image datasets; getting better results; and applying the proposed method in various fields.
References:
1. M. Arabboev et al., "Development of a Novel Method for Image Resizing Using Artificial Neural Network," in In: Zaynidinov, H., Singh, M., Tiwary, U.S., Singh, D. (eds) Intelligent Human Computer Interaction. IHCI 2022. Lecture Notes in Computer Science, vol 13741.
2. M. Arabboev, S. Begmatov, K. Nosirov, J. C. Chedjou, and K. Kyamakya, "Development of a novel method of adaptive image interpolation for image resizing using artificial intelligence," in IVUS 2022: 27th International Conference on Information Technology, pp. 32-38.
3. S. Subhasini and M. Singh, "Color Image Edge Detection: A Survey," Int. J. Innov. Eng. Technol., vol. 8, no. 1, pp. 235-247, 2017.
4. G. M. H. Amer and A. M. Abushaala, "Edge detection methods," 2015 2nd World Symp. Web Appl. Networking, WSWAN 2015, no. April, 2015.
5. S. Malik and T. Kumar, "Comparative Analysis of Edge Detection between Gray Scale and Color Image," Commun. Appl. Electron., vol. 5, no. 2, pp. 38-43, 2016.
6. J. Yan, L. Zhang, X. Luo, H. Peng, and J. Wang, "A novel edge detection method based on dynamic threshold neural P systems with orientation," Digit. Signal Process. A Rev. J., vol. 127, no. 62076206, p. 103526, 2022.
7. W. Xu and J. Mulligan, "A fast multi-channel edge detection algorithm for vision-based autonomous spacecraft docking," 2009 Work. Appl. Comput. Vision, WACV 2009, pp. 1-6, 2009.
8. R. Li, D. Han, J. Dezert, and Y. Yang, "A novel edge detector for color images based on MCDM with evidential reasoning," 20th Int. Conf. Inf. Fusion, Fusion 2017 - Proc., 2017.
9. P. A. Flores-Vidal, D. Gómez, J. Castro, G. Villarino, and J. Montero, "A new approach to color edge detection," Proc. 11th Conf. Eur. Soc. Fuzzy Log. Technol. EUSFLAT 2019, vol. 1, no. Eusflat, pp. 376-384, 2020.
