Lafta, R., Sultani, Z. (2025). An optimize canny algorithm with traditional machine learning for edge detection enhancement. , 43(4), 253-259. doi: 10.30684/etj.2025.158177.1914
Russel K. Lafta; Zainab N. Sultani. "An optimize canny algorithm with traditional machine learning for edge detection enhancement". , 43, 4, 2025, 253-259. doi: 10.30684/etj.2025.158177.1914
Lafta, R., Sultani, Z. (2025). 'An optimize canny algorithm with traditional machine learning for edge detection enhancement', , 43(4), pp. 253-259. doi: 10.30684/etj.2025.158177.1914
Lafta, R., Sultani, Z. An optimize canny algorithm with traditional machine learning for edge detection enhancement. , 2025; 43(4): 253-259. doi: 10.30684/etj.2025.158177.1914

An optimize canny algorithm with traditional machine learning for edge detection enhancement

Engineering and Technology Journal
Volume 43, Issue 4, April 2025, Pages 253-259    PDF (428.67 K)
Document Type: Research Paper
DOI: 10.30684/etj.2025.158177.1914
Authors
Russel K. Lafta* ; Zainab N. Sultani
Computer Science Dept., College of Science, Al-Nahrain University, Jadriya, Baghdad, Iraq.
Abstract
Edge detection still represents a major challenge in image processing and computer vision because of the complexity and variability of real-world images. The canny algorithm is powerful in detecting edges. However, it is sensitive to noise, and as a result, may produce weak edges. The use of machine learning algorithms has significantly improved the performance of edge detection techniques. In this paper, an improved Canny edge detection algorithm is proposed by replacing the Gaussian filter with a bilateral filter. Also, a new approach for estimating Canny algorithm thresholds has been developed using the Flower Pollination algorithm. Subsequently, the improved Canny algorithm with a machine learning model was integrated to enhance edge detection accuracy. The performance of the improved algorithm was evaluated using 50 images from the Berkeley Computer Vision dataset. The experiment results show that the enhanced algorithm has an AUC of 0.81 for RF (Random Forest) and 0.75 for the LR (Logistic Regression) classifier, which can detect edges more accurately than the traditional Canny algorithm, which has an AUC of about 0.57. The proposed method sets a new standard for edge detection performance.

Highlights
  • Canny edge detection was enhanced using FPA for optimal threshold selection
  • A bilateral filter was used to reduce noise while preserving fine details
  • The method achieved higher accuracy than traditional edge detection techniques
  • Precision and recall were balanced, reducing false edge detection
  • F1-score, precision, and recall were used for quantitative validation
Keywords
Edge detection; Optimized canny; Flower Pollination algorithm; Machine learning
References
  1. C. Lopez-Molina, M. Galar, H. Bustince, B. De Baets, On the impact of anisotropic diffusion on edge detection, Pattern Recognit., 47 (2014) 270–281. https://doi.org/10.1016/j.patcog.2013.07.009
  2. O. Elharrouss, Y. Hmamouche, A. K. Idrissi, B. El Khamlichi, A. El Fallah-Seghrouchni, Refined edge detection with cascaded and high-resolution convolutional network, Pattern Recognit., 138 (2023) 109361. https://doi.org/10.1016/j.patcog.2023.109361
  3. S. Öztürk and B. Akdemir, Comparison of edge detection algorithms for texture analysis on glass production, Procedia-Social Behav. Sci., 195 (2015) 2675–2682. https://doi.org/10.1016/j.sbspro.2015.06.477
  4. P. F. Felzenszwalb, R. B. Girshick, D. McAllester; D. Ramanan, Object detection with discriminatively trained part-based models, IEEE Trans. Pattern Anal. Mach. Intell., 32 (2009) 1627–1645. https://doi.org/10.1109/TPAMI.2009.167
  5. C. L. Zitnick, P. Dollár, Edge boxes: Locating object proposals from edges, in: Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, Proceedings, Part V, Springer, (2014) 391–405. https://doi.org/10.1007/978-3-319-10602-1_26
  6. N. R. Pal, S. K. Pal, A review on image segmentation techniques, Pattern Recognit., 26 (1993) 1277–1294. https://doi.org/10.1016/0031-3203(93)90135-J
  7. R. Sun, T. Lei, Q. Chen, Z. Wang, X. Du, et al., Survey of image edge detection, Front. Signal Process., 2 (2022) 826967. https://doi.org/10.3389/frsip.2022.826967
  8. P. Flores-Vidal, J. Castro, D. Gomez, Postprocessing of edge detection algorithms with machine learning techniques, Math. Probl. Eng., 2022 (2022) 9729343. https://doi.org/10.1155/2022/9729343
  9. V. K. Ha, J. Ren, X. Xu, S. Zhao, G. Xie, V. M. Vorgas, Deep learning-based single image super-resolution: A survey, in: Advances in Brain-Inspired Cognitive Systems: 9th International Conference, BICS 2018, Xi’an, China, Springer, 2018. https://doi.org/10.1007/978-3-030-00563-4_11
  10. Z.-W. He, L. Zhang, F.-Y. Liu, DiscoStyle: Multi-level logistic ranking for personalized image style preference inference, Int. J. Autom. Comput., 17 (2020) 637–651. https://doi.org/10.1007/s11633-020-1244-1
  11. Y.-H. Wu, Y. Liu, L. Zhang, M-M. Cheng, EDN: Salient object detection via extremely-downsampled network, IEEE Trans. Image Process., 31 (2022) 3125–3136. https://doi.org/10.1109/TIP.2022.3164550
  12. C. Zhao, Y. Hao, S. Sui, S. Sui, A new method to detect the license plate in dynamic scene, in: 2018 IEEE 7th Data Driven Control and Learning Systems Conference (DDCLS), IEEE, 2018. https://doi.org/10.1109/DDCLS.2018.8516012
  13. L. Gou, H. Li, H. Zheng, H. Li, X. Pei, Aeroengine control system sensor fault diagnosis based on CWT and CNN, Math. Probl. Eng., 2020 (2020) 5357164. https://doi.org/10.1155/2020/5357146
  14. Jingzhong, H., et al., Strip steel surface defects recognition based on SOCP optimized multiple kernel RVM, Math. Probl. Eng., 2018 (2018) 9298017. https://doi.org/10.1155/2018/9298017
  15. V. Maksimovic, M. Petrovic, D. Savic, B. Jksic, P. Spalevic,. New approach of estimating edge detection threshold and application of adaptive detector depending on image complexity, Optik, 238 (2021) 166476. https://doi.org/10.1016/j.ijleo.2021.166476
  16. H. Yamagiwa, Y. Takase, H. Kambe, R. Nakamoto, Zero-shot edge detection with SCESAME: Spectral clustering-based ensemble for segment anything model estimation, in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024. https://doi.org/10.48550/arXiv.2308.13779
  17. J. Canny, A computational approach to edge detection, IEEE Trans. Pattern Anal. Mach. Intell., 6 (1986) 679–698. https://doi.org/10.1109/TPAMI.1986.4767851
  18. C. Tomasi and R. Manduchi, Bilateral filtering for gray and color images, 6th International Conference on Computer Vision (IEEE Cat. No. 98CH36271), IEEE, 1998. https://doi.org/10.1109/ICCV.1998.710815
  19. Yang, X.-S., Flower Pollination Algorithm for Global Optimization, Lecture Notes in Computer Science, Springer, 7445, 2012. https://doi.org/10.1007/978-3-642-32894-7_27
  20. D. Martin, C.Fowlkes, D. Tal, J. Malik, A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics, in: Proceedings of the 8th IEEE International Conference on Computer Vision (ICCV 2001), IEEE, 2001. https://doi.org/10.1109/ICCV.2001.937655
  21. L. Breiman, Random forests, Mach. Learn., 45 (2001) 5–32. https://doi.org/10.1023/A:1010933404324
  22. A. Kumar, S. Raheja, Edge detection using guided image filtering and enhanced ant colony optimization, Procedia Comput. Sci., 173 (2020) 8-17. https://doi.org/10.1016/j.procs.2020.06.003
Statistics
Article View: 199
PDF Download: 150


Journal Management System. Powered by iJournalPro.com