Enhancing Edge Detection with an Optimized Canny Algorithm, Hough Transform, and Machine Learning Integration
Keywords:
Geometric Shape Recognition , Noise Reduction , Threshold Optimization , Hough Transform , Machine LearningAbstract
This paper addresses the challenge of accurate edge detection in image data, particularly in noisy and complex environments, by proposing a hybrid approach that combines classical and advanced techniques. An optimized Canny edge detection algorithm is first applied to enhance precision by reducing noise and improving edge continuity. Following this, the Hough Transform is used to extract geometric shapes, such as lines, due to its robustness against irregularities. To further improve detection accuracy, machine learning (ML) models are integrated, enabling the system to adapt and learn from the data. Experimental results on 50 images from BSD dataset show that the proposed hybrid method surpasses standalone techniques, achieving an accuracy of 88.47%, a precision of 29.79%, a recall of 38.55%, and an AUC of 0.80 using XGBoost. This combination of classical image processing methods and ML demonstrates significant improvements in edge extraction performance, making it a reliable solution for edge detection in diverse and noisy image datasets.
References
[1] Sun, R.; Lei, T.; Chen, Q.; Wang, Z.; Du, X.; Zhao, W.; and Nandi, A. K.; “Survey of image edge detection”. Front. Signal Process, 2: 826967, 2022.
[2] Gaurav, K.; and Ghanekar, U.; “Image steganography based on Canny edge detection, dilation operator and hybrid coding”. J. Inf. Secure. Appl, 41: 41-51, 2018.
[3] Xie, X.; Ge, S.; Xie, M.; Hu, F.; and Jiang, N.; “An improved industrial sub-pixel edge detection algorithm based on coarse and precise location”. J. Ambient Intell. Humaniz. Comput, 11: 2061-2070, 2020.
[4] Mukhopadhyay, P.; and Chaudhuri, B. B.; “A survey of Hough Transform”. Pattern Recognit., 48(3): 993-1010, 2015.
[5] Hough, P.V. "Method and Means for Recognizing Complex Patterns". U.S. Patent 3,069,654, 1962.
[6] Duda, R. O.; and Hart, P. E.; “Use of the Hough transformation to detect lines and curves in pictures”. Commun. ACM, 15(1): 11-15, 1972.
[7] Ballard, D. H.; “Generalizing the Hough transform to detect arbitrary shapes”. Pattern Recognit., 13(2): 111-122, 1981.
[8] Beltrametti, M. C.; and Robbiano, L.; “An algebraic approach to Hough transforms”. J. Algebra, 371: 669-681, 2012.
[9] Conti, C.; Romani, L.; and Schenone, D.; “Semi-automatic spline fitting of planar curvilinear profiles in digital images using the Hough transform”. Pattern Recognit., 74: 64-76, 2018.
[10] Babu, G.; and Pinjari, A. K.; “A new design of iris recognition using hough transform with K-means clustering and enhanced faster R-CNN”. Cybern. Syst, 55(2): 551-584, 2024.
[11] Bamogo, M.; and Sere, A.; "An application of the Hough transform with convolutional neural networks to detect lines". In: Proceedings of the International Conference on Innovations and Technological Solutions for Sustainable Development (Intersol 2024), Dakar, Senegal, July; Springer, 2024.
[12] Park, K.; Chae, M.; and Cho, J. H.; “Image pre-processing method of machine learning for edge detection with image signal processor enhancement”. Micromachines, 12(1): 73, 2021.
[13] Flores-Vidal, P.; Castro, J.; and Gomez, D.; “Postprocessing of Edge Detection Algorithms with Machine Learning Techniques”. Math. Probl. Eng, 2022(1): 1-12, 2022.
[14] Russel, K. L.; and Zainab, N. S.; “An Optimized Canny Edge Detection with Traditional Machine Learning for Edge Detection Enhancement”. Eng. Technol. (under review).
[15] Martin, D.; Fowlkes, C.; Tal, D.; and Malik, J.; "A Database of Human Segmented Natural Images and Its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics". In: Proceedings of the Eighth IEEE International Conference on Computer Vision (ICCV 2001), Vancouver, Canada, 7–14 July; IEEE, 2001.
[16] Tomasi, C.; and Manduchi, R.; "Bilateral Filtering for Gray and Color Images". In: Proceedings of the Sixth International Conference on Computer Vision (ICCV 1998), Bombay, India, January; IEEE, 1998.
[17] Yang, X.-S.; "Flower Pollination Algorithm for Global Optimization". In: Proceedings of the International Conference on Unconventional Computing and Natural Computation, Orléans, France, 3–7 September; Springer, 2012.
[18] Illingworth, J.; and Kittler, J.; “A survey of the Hough transform”. Comput. Gr. Image Process., 44(1): 87-116, 1988.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Russel K. Lafta, Zainab N. Sultani
This work is licensed under a Creative Commons Attribution 4.0 International License.
.jpg)
