Journal of Space Science and Technology

Journal of Space Science and Technology

Performance Evaluation of Feature Detection Algorithms and Their Impact on the Accuracy and Efficiency of Visual Odometry

Document Type : Original Research Paper

Authors
1 Schoole of Advanced Technologies, Iran University of Science and Technology, Tehran, Iran
2 Iranian Space Research Institute, Tehran, Iran
Abstract
Feature detection is a critical component of visual odometry, directly influencing position estimation accuracy. This process forms the basis for identifying key points in images, playing a pivotal role in subsequent operations such as feature matching and motion tracking. This study examines the impact of various feature detection algorithms on position estimation accuracy in visual odometry, focusing on a comparative analysis of the Harris, FAST, SIFT, CenSurE, and ORB algorithms. Performance evaluation was conducted based on accuracy and computational efficiency in position estimation. Each algorithm's average errors and processing times were calculated and systematically compared using an image dataset. Results indicate that the CenSurE algorithm is optimal for real-time applications and scenarios demanding rapid processing due to its lower computational cost. Its high-speed feature extraction capability makes it particularly suitable for such use cases. Conversely, despite its higher processing time, the Harris algorithm offers superior accuracy in position estimation and angular measurement, making it a preferred choice when precision is prioritized over speed. The FAST and SIFT algorithms balance accuracy and computational efficiency; the FAST algorithm, with its lower processing time, performs effectively in horizontal orientations, whereas the Harris algorithm excels in precision. The ORB algorithm exhibits moderate speed and acceptable performance but demonstrates reduced accuracy in certain positional features. This study enhances the understanding of the trade-offs between accuracy and efficiency in feature detection for visual odometry, providing a foundation for further research in optimizing algorithm selection for specific applications.
Keywords

Subjects


Article Title Persian

ارزیابی عملکرد الگوریتم‌های استخراج ویژگی و تأثیر آن‌ها بر دقت و کارایی ناوبری بصری

Authors Persian

سیدجواد شجاع الساداتی 1
مهدی نصیری سروی 1
محمد سینجلی 2
1 کارشناسی ارشد، دانشکده فناوری های نوین، دانشگاه علم و صنعت ایران، تهران، ایران
2 استادیار، پژوهشگاه فضایی ایران، تهران، ایران
Abstract Persian

استخراج ویژگی یکی از مراحل اساسی در ناوبری بصری است که تأثیر چشمگیری بر دقت تخمین موقعیت دارد. این مرحله به‌‌عنوان پایه فرآیند شناسایی و تمایز اشیاء و نقاط شاخص در تصاویر عمل کرده و در مراحل بعدی، مانند تطابق ویژگی‌ها و ردیابی حرکت، نقش حیاتی ایفا می‌کند. این مقاله تأثیر الگوریتم‌های استخراج ویژگی بر دقت تخمین موقعیت در ناوبری بصری را بررسی کرده و هدف اصلی آن تحلیل و مقایسه عملکرد الگوریتم‌های Harris، FAST، SIFT، CenSurE و ORB از نظر دقت در تخمین موقعیت و مدت زمان پردازش هر مرحله است. با استفاده از مجموعه داده‌های تصویری، میانگین خطاها و مدت زمان‌های پردازش برای هر الگوریتم محاسبه و نتایج به‌صورت مقایسه‌ای ارائه شده است. نتایج نشان می‌دهند که الگوریتم CenSurE به‌دلیل زمان پردازش کمتر، برای کاربردهای بلادرنگ و سایر کاربردهایی که نیازمند پردازش سریع هستند، ایده‌آل است. سرعت بالای این الگوریتم در پردازش ویژگی‌ها، آن را به گزینه‌ای مناسب برای این نوع کاربردها تبدیل می‌کند. در مقابل، الگوریتم Harris با وجود مدت زمان پردازش بیشتر، دقت بالایی در تخمین موقعیت و سنجش زوایای چرخشی دارد و در سناریوهایی که دقت بر سرعت اولویت دارد، گزینه‌ای کارآمد است. الگوریتم‌های FAST و SIFT نیز ترکیبی از سرعت و دقت را ارائه می‌دهند، الگوریتم FAST با زمان پردازش پایین، عملکرد مناسبی در محورهای افقی دارد و برای کاربردهایی با نیاز به پردازش سریع مناسب است، در حالی‌که Harris دقت بیشتری را در تخمین موقعیت فراهم می‌آورد. الگوریتم ORB نیز با سرعتی متوسط، عملکردی نسبتا قابل قبول را نشان می‌دهد، اما در برخی ویژگی‌های موقعیتی دقت کمتری دارد. این مقاله به درک عمیق‌تری از اثرات الگوریتم‌های استخراج ویژگی بر دقت تخمین موقعیت کمک کرده و بستر مناسبی برای تحقیقات آینده در این حوزه فراهم می‌سازد.

Keywords Persian

ناوبری بصری
الگوریتم‌های آشکار ساز
بینایی ماشین
الگوریتم‌های استخراج ویژگی
تطابق ویژگی‌
[1]   F. Fraundorfer and D. Scaramuzza, "Visual odometry: Part ii: matching, robustness, optimization, and applications," IEEE Robotics & Automation Magazine, vol. 19, no. 2, pp. 78-90, 2012, https://doi.org/10.1109/MRA.2012.2182810.
[2]   P. T. Yap, X. Jiang, and A. C. Kot, "Two-dimensional polar harmonic transforms for invariant image representation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 7, pp. 1259-1270, 2010, https://doi.org/10.1109/TPAMI.2009.119.
[3]   S. Liu and X. Bai, "Discriminative features for image classification and retrieval," Pattern Recognition Letters, vol. 33, no. 6, pp. 744-751, 2012, https://doi.org/10.1016/j.patrec.2011.12.008.
[4]   J. Wang, Y. Li, Y. Zhang, C. Chao Wang, H. Xie, and G. Chen, "Notice of violation of IEEE publication principles: bag-of-features based medical image retrieval via multiple assignment and visual words weighting," IEEE Transactions on Medical Imaging, vol. 30, no. 11, pp. 1996-2011, 2011, https://doi.org/10.1109/TMI.2011.2161673.
[5]   A. Andreopoulos and J. K. Tsotsos, "50 years of object recognition: Directions forward," Computer Vision and Image Understanding, vol. 117, no. 8, pp. 827-891, 2013, https://doi.org/10.1016/j.cviu.2013.04.005.
[6]   B. Kim, H. Yoo, and K. Sohn, "Exact order based feature descriptor for illumination robust image matching," Pattern Recognition, vol. 46, no. 12, pp. 3268-3278, 2013, https://doi.org/10.1016/j.patcog.2013.04.015.
[7]   P. Moreels and P. Perona, "Evaluation of features detectors and descriptors based on 3D objects," International Journal of Computer Vision, vol. 73, pp. 263-284, 2007, https://doi.org/10.1007/s11263-006-9967-1.
[8]   G. Takacs, V. Chandrasekhar, S. Tsai, D. Chen, R. Grzeszczuk, and B. Girod, "Rotation-invariant fast features for large-scale recognition and real-time tracking," Signal Processing: Image Communication, vol. 28, no. 4, pp. 334-344, 2013, https://doi.org/10.1016/j.image.2012.11.004.
[9]   J. M. Del Rincón, D. Makris, C. O. Uruñuela, and J. C. Nebel, "Tracking human position and lower body parts using Kalman and particle filters constrained by human biomechanics," IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 41, no. 1, pp. 26-37, 2010, https://doi.org/10.1109/TSMCB.2010.2044041.
[10] S. Lazebnik, C. Schmid, and J. Ponce, "A sparse texture representation using local affine regions," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 8, pp. 1265-1278, 2005, https://doi.org/10.1109/TPAMI.2005.151.
[11] L. Liu and P. Fieguth, "Texture classification from random features," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, no. 3, pp. 574-586, 2012, https://doi.org/10.1109/TPAMI.2011.145.
[12] A. C. Murillo, J. J. Guerrero, and C. Sagues, "Surf features for efficient robot localization with omnidirectional images," in International Conference on Robotics and Automation, Rome, Italy, 2007, pp. 3901-3907, https://doi.org/10.1109/ROBOT.2007.364077.
[13] N. Farajzadeh, K. Faez, and G. Pan, "Study on the performance of moments as invariant descriptors for practical face recognition systems," IET Computer Vision, vol. 4, no. 4, pp. 272-285, 2010, https://doi.org/10.1049/iet-cvi.2009.0140.
[14] C. Schmid, R. Mohr, and C. Bauckhage, "Comparing and evaluating interest points," in Sixth International Conference on Computer Vision (IEEE Cat. No. 98CH36271), Bombay, India, 1998, pp. 230-235, https://doi.org/10.1109/ICCV.1998.710723.
[15] C. Schmid, R. Mohr, and C. Bauckhage, "Evaluation of interest point detectors," International Journal of Computer Vision, vol. 37, no. 2, pp. 151-172, 2000, https://doi.org/10.1023/A:1008199403446.
[16] S. K. Sharma, K. Jain, and A. K. Shukla, "A comparative analysis of feature detectors and descriptors for image stitching," Applied Sciences, vol. 13, no. 10, 2023, Art. no. 6015, https://doi.org/10.3390/app13106015.
[17] N. Kosaka and G. Ohashi, "Vision-based nighttime vehicle detection using CenSurE and SVM," IEEE Transactions on Intelligent Transportation Systems, vol. 16, no. 5, pp. 2599-2608, 2015, https://doi.org/10.1109/TITS.2015.2413971.
[18] N. Govender, "Evaluation of feature detection algorithms for structure from motion," in 3rd Robotics and Mechatronics Symposium (ROBMECH 2009). Pretoria, South Africa, 2009, p 4.
[19] B. Herbert, "Surf: Speeded up robust features," Computer Vision and Image Understanding, vol. 110, no. 3, pp. 346-359, 2008.
[20] C. G. Harris and M. Stephens, "A combined corner and edge detector," in Alvey Vision Conference, Manchester, UK, 1988.
[21] D. G. Lowe, "Distinctive image features from scale-invariant keypoints," International Journal of Computer Vision, vol. 60, pp. 91-110, 2004, https://doi.org/10.1023/B:VISI.0000029664.99615.94.
[22] E. Rosten and T. Drummond, "Machine learning for high-speed corner detection," in Computer Vision–ECCV 2006: 9th European Conference on Computer Vision, Part I, A. Leonardis, H. Bischof, and A. Pinz, Eds. Springer, 2006, pp. 430-443, https://doi.org/10.1007/11744023_34.
[23] S. Gupta, M. Kumar, and A. Garg, "Improved object recognition results using SIFT and ORB feature detector," Multimedia Tools and Applications, vol. 78, no. 23, pp. 34157-34171, 2019, https://doi.org/10.1007/s11042-019-08232-6.
[24] E. Rublee, V. Rabaud, K. Konolige, and G. Bradski, "ORB: An efficient alternative to SIFT or SURF," in International Conference on Computer Vision, Barcelona, Spain, 2011, pp. 2564-2571, https://doi.org/10.1109/ICCV.2011.6126544.
[25] S. Li, "A review of feature detection and match algorithms for localization and mapping," in IOP Conference Series: Materials Science and Engineering, vol. 2017, 2nd International Seminar on Advances in Materials Science and Engineering, Singapore, 2017, Paper 012003, https://doi.org/10.1088/1757-899X/231/1/012003.
[26] S. Andolfo, F. Petricca, and A. Genova, "Rovers localization by using 3D-to-3D and 3D-to-2D visual odometry," in 8th International Workshop on Metrology for AeroSpace (MetroAeroSpace), Naples, Italy, 2021, pp. 334-339, https://doi.org/10.1109/MetroAeroSpace51421.2021.9511741.
[27] Y. Cheng, M. W. Maimone, and L. Matthies, "Visual odometry on the Mars exploration rovers-a tool to ensure accurate driving and science imaging," IEEE Robotics & Automation Magazine, vol. 13, no. 2, pp. 54-62, 2006, https://doi.org/10.1109/MRA.2006.1638016.
Volume 18, Issue 1
2025
Pages 38-52

  • Receive Date 05 November 2024
  • Revise Date 18 January 2025
  • Accept Date 22 February 2025
  • First Publish Date 01 March 2025