Comparison of handcrafted features extraction techniques for traffic sign recognition

Prestone Simiyu; Raphael Angulu; Daniel Otanga

doi:10.51867/ajernet.6.4.2

Autores

Prestone Simiyu Masinde Muliro University of Science and Technology, Kenya https://orcid.org/0000-0001-5802-0723
Raphael Angulu Masinde Muliro University of Science and Technology, Kenya https://orcid.org/0000-0002-7641-1805
Daniel Otanga Masinde Muliro University of Science and Technology, Kenya https://orcid.org/0000-0001-7212-1088

DOI:

https://doi.org/10.51867/ajernet.6.4.2

Palavras-chave:

Bag of Visual Words (BoW), Computer Vision, Feature Extraction, Gabor Filters, Handcrafted Features, Histogram of Oriented Gradients (HOG), Local Binary Patterns (LBP), Machine Learning, Scale-Invariant Feature Transform (SIFT), Traffic Sign Recognition (TSR)

Resumo

Traffic Sign Recognition (TSR) is a critical computer vision task for the advancement of Autonomous Driver Assistance Systems (ADAS) and autonomous vehicles, directly impacting road safety. While deep learning dominates current research, handcrafted feature extraction techniques remain relevant due to their interpretability, lower computational demands, and suitability for embedded systems. This paper presents a systematic empirical evaluation and comparison of five predominant handcrafted feature extraction methods: Histogram of Oriented Gradients (HOG), Scale-Invariant Feature Transform (SIFT), Bag of Words (BoW), Local Binary Patterns (LBP), and Gabor Filters for traffic sign recognition. The techniques are assessed on the German Traffic Sign Recognition Benchmark (GTSRB) using three classifiers: Support Vector Machine (SVM), Random Forest, and Decision Tree. Experimental results demonstrate that the choice of feature extractor and classifier significantly impacts performance, with HOG combined with an SVM classifier achieving the highest accuracy (95.2%). The study provides a clear performance hierarchy, revealing that HOG and LBP offer the best balance of accuracy and computational efficiency for this domain. We conclude that the optimal selection of a handcrafted feature extraction strategy is problem-dependent and provide concrete recommendations for implementing effective and efficient TSR systems.

Downloads

Não há dados estatísticos.

Referências

Almaskati, D., Kermanshachi, S., & Pamidimukkala, A. (2023). Autonomous vehicles and traffic accidents. International Scientific Conference "The Science and Development of Transport," 17(3), 1-8. ELSEVIER. https://www.researchgate.net/publication/374947350_Autonomous_vehicles_and_traffic_accidents

Azhar, R., Tuwohingide, D., Kamudi, D., Suciati, N., & Sarjana, S. (2015). Batik image classification using SIFT feature extraction, bag of features and support vector machine. Procedia Computer Science, 72, 24-30. https://www.researchgate.net/publication/289735014_Batik_Image_Classification_Using_SIFT_Feature_Extraction_Bag_of_Features_and_Support_Vector_Machine

https://doi.org/10.1016/j.procs.2015.12.101 DOI: https://doi.org/10.1016/j.procs.2015.12.101

Ellahyani, A., & El Ansari, M. (2021). Traffic sign detection for intelligent transportation systems: A survey. E3S Web of Conferences, 229, 01040. https://doi.org/10.1051/e3sconf/202122901040 DOI: https://doi.org/10.1051/e3sconf/202122901006

Fu, M.-Y., & Huang, Y.-S. (2010, December). A survey of traffic sign recognition. In 2010 International Conference on Wavelet Analysis and Pattern Recognition (pp. 119-124). IEEE.

https://doi.org/10.1109/ICWAPR.2010.5576425 DOI: https://doi.org/10.1109/ICWAPR.2010.5576425

Jaiswal, A. K., & Banka, H. (2017). Local pattern transformation-based feature extraction techniques for classification of epileptic EEG signals. Biomedical Signal Processing and Control, 34, 81-92. https://www.sciencedirect.com/science/article/abs/pii/S174680941730006X?via%3Dihub

https://doi.org/10.1016/j.bspc.2017.01.005 DOI: https://doi.org/10.1016/j.bspc.2017.01.005

John, V., Boyali, A., & Mita, S. (2017). Gabor filter and Gershgorin disk-based convolutional filter constraining for image classification. International Journal of Machine Learning and Computing, 7(4), 55-60. https://doi.org/10.18178/ijmlc.2017.7.4.620 DOI: https://doi.org/10.18178/ijmlc.2017.7.4.620

Khalid, S., Khalil, T., & Nasreen, S. (2014, November). A survey of feature selection and feature extraction techniques in machine learning. In 2014 Science and Information Conference (pp. 372-378). IEEE. https://doi.org/10.1109/SAI.2014.6918213 DOI: https://doi.org/10.1109/SAI.2014.6918213

Kiran, C. G., Prabhu, L., Varikkottil, A., & Kumaraswamy, R. (2009). Traffic sign detection and pattern recognition using support vector machine. In Proceedings (pp. 87-90). https://www.researchgate.net/publication/220781211_Traffic_Sign_Detection_and_Pattern_Recognition_Using_Support_Vector_Machine

https://doi.org/10.1109/ICAPR.2009.58 DOI: https://doi.org/10.1109/ICAPR.2009.58

Kuo, W. J., & Chiu, C. C. (2012). Content-based image retrieval with bag of visual words. Journal of Information Science and Engineering, 28(3), 655-668.

Luus, F. P. S., Salmon, B. P., van den Bergh, F., & Maharaj, B. T. J. (2015). Multiview deep learning for land-use classification. IEEE Geoscience and Remote Sensing Letters, 12(12), 2448-2452. https://doi.org/10.1109/LGRS.2015.2483680 DOI: https://doi.org/10.1109/LGRS.2015.2483680

Maletzky, A., Hofer, N., Thumfart, S., Bruckmüller, K., & Kasper, J. (2023). Traffic sign detection and classification on the Austrian highway traffic sign data set. Data, 8(1), 16.

https://doi.org/10.3390/data8010016 DOI: https://doi.org/10.3390/data8010016

Mikolajczyk, K., & Schmid, C. (2005). A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(10), 1615-1630. https://doi.org/10.1109/TPAMI.2005.188 DOI: https://doi.org/10.1109/TPAMI.2005.188

Nanni, L., & Lumini, A. (2012). Local binary patterns variants as texture descriptors for medical image analysis. Artificial Intelligence in Medicine, 56(1), 25-35. https://doi.org/10.1016/j.artmed.2012.07.001 DOI: https://doi.org/10.1016/j.artmed.2012.07.001

Ojala, T., Pietikäinen, M., & Harwood, D. (1996). A comparative study of texture measures with classification based on feature distributions. Pattern Recognition, 29(1), 51-59. https://doi.org/10.1016/0031-3203(95)00067-4 DOI: https://doi.org/10.1016/0031-3203(95)00067-4

Omid-Zohoor, A., Almeida, B., Taori, R., Murmann, B., & Chandraiah, P. (2018). Toward always-on mobile object detection: Energy versus performance tradeoffs for embedded HOG feature extraction. IEEE Transactions on Circuits and Systems for Video Technology, 28(5), 1102-1115. https://www.researchgate.net/publication/312476432_Toward_Always-On_Mobile_Object_Detection_Energy_Versus_Performance_Tradeoffs_for_Embedded_HOG_Feature_Extraction

https://doi.org/10.1109/TCSVT.2017.2653187 DOI: https://doi.org/10.1109/TCSVT.2017.2653187

Park, K. H., Wang, S.-W., & Wang, W.-H. (2014). Frequency-cepstral features for bag of words based acoustic context awareness. The Journal of the Acoustical Society of Korea, 33(4), 248-254. https://doi.org/10.7776/ASK.2014.33.4.248 DOI: https://doi.org/10.7776/ASK.2014.33.4.248

Peng, F., Qian, L., Yang, J., Zhang, C., & Zuo, W. (2014). Feature coding in image classification: A comprehensive study. IEEE Transactions on Multimedia, 16(11), 2828-2840. https://www.researchgate.net/publication/239943155_Feature_Coding_in_Image_Classification_A_Comprehensive_Study

Rangayyan, R. M., Banik, S., & Desautels, J. E. (2010). Computer-aided detection of architectural distortion in prior mammograms of interval cancer. Journal of Digital Imaging, 23(5), 564-577. https://link.springer.com/article/10.1007/s10278-009-9257-x DOI: https://doi.org/10.1007/s10278-009-9257-x

Saadna, Y., & Behloul, A. (2017). An overview of traffic sign detection and classification methods. International Journal of Multimedia and Information Retrieval, 6(3), 193-206. https://doi.org/10.1007/s13735-017-0129-8 DOI: https://doi.org/10.1007/s13735-017-0129-8

Sadeghi, B., Jafari, A. M., Vahdani, K., & Shahri, A. M. (2018). 2DIGH: A polar invariant local image descriptor based on joint histogram. The Visual Computer, 34, 1579-1595. https://www.researchgate.net/publication/319661560_2DIGH_a_polar_invariant_local_image_descriptor_based_on_joint_histogram

https://doi.org/10.1007/s00371-017-1433-2 DOI: https://doi.org/10.1007/s00371-017-1433-2

Singh, B. (2016). Iris recognition using curvelet transformation based on Gabor filter & SVM. International Journal of Engineering and Computer Science, 5(8), 17731-17735. https://doi.org/10.18535/ijecs/v5i8.32 DOI: https://doi.org/10.18535/ijecs/v5i8.32

Sugimura, D., Fujimura, T., & Hakura, T. (2016). Enhanced cascading classifier using multi-scale HOG for pedestrian detection from aerial images. International Journal of Pattern Recognition and Artificial Intelligence, 30(3), 1655009. https://www.researchgate.net/publication/283617136_Enhanced_Cascading_Classifier_Using_Multi-Scale_HOG_for_Pedestrian_Detection_from_Aerial_Images

https://doi.org/10.1142/S0218001416550090 DOI: https://doi.org/10.1142/S0218001416550090

Wali, S. B. (2015). Comparative survey on traffic sign detection and recognition: A review. Przeglad Elektrotechniczny, 91(10), 40-44. https://www.researchgate.net/publication/283617136_Enhanced_Cascading_Classifier_Using_Multi-Scale_HOG_for_Pedestrian_Detection_from_Aerial_Images

https://doi.org/10.15199/48.2015.12.08 DOI: https://doi.org/10.15199/48.2015.12.08

Wang, H., & Schmid, C. (2013, December). Action recognition with improved trajectories. In 2013 IEEE International Conference on Computer Vision (pp. 3551-3558). IEEE. https://doi.org/10.1109/ICCV.2013.441 DOI: https://doi.org/10.1109/ICCV.2013.441

Zhang, Z., Gao, S., & Zhao, B. (2013). Multi-scale local binary pattern for texture classification. Journal of Computational Information Systems, 9(7), 2561-2568.

Zhao, S. (2013). Image bag generator based on bag of visual words. Journal of Information and Computational Science, 10(5), 1453-1462. https://doi.org/10.12733/jics20101532 DOI: https://doi.org/10.12733/jics20101532