Trafik İşaret Levhası Tespiti için Derin Öğrenme Yöntemi
Year 2020,
Volume: 3 Issue: 2, 140 - 157, 30.11.2020
Mert Çetinkaya
,
Tankut Acarman
Abstract
Bu çalışmada araç üzerinde bulunan kamera algılayıcısı ile çekilen trafik sahne resimleri üzerinde trafik işaret levhası tespiti için yöntem öneriyoruz. Yöntemimiz derin öğrenmeye dayalı olmakla birlikte işlemsel hız ve tespit başarısını artırmak üzere istatistiksel yaklaşımlardan da barındırmaktadır. Yöntemimiz oluşturan algoritmamız 3 adımdan oluşmaktadır. İlk olarak derin öğrenmeye dayalı bir görüntü segmentasyonu yaptık ve görüntüden bazı bölge önerileri çıkardık. Sonra, hipotez testleri yaptık ve resim içerisinde trafik levhası olma ihtimali düşük olan bölge önerilerinden bazılarını eledik. Son adımda ise uygun şekilde eğitilmiş olan Konvolüsyonel Sinir Ağları (CNN) sınıflandırıcımızı kalan bölge önerileri üzerinde kullandık ve sınıflandıracımız tarafından o bölgede trafik levhası olduğu güçlü bir şekilde saptandı ise ilgili bölgeyi kabul ettik. Burada, ilk adımda verimli ve hızlı çalışan bir görüntü segmentasyonu yaklaşımı kullandık ve ikinci adımda da hipotez testi gibi basit bir yaklaşım ile hızlıca bazı elemeler yaptık. Algoritmamızı Alman Trafik İşareti Algılama Karşılaştırması (German Traffic Sign Detection Benchmark (GTSDB)) veriseti üzerinde test ettik ve algoritmamızın kullandığımız metriklere göre literatürde yayınlanmış başka yöntemlere göre daha başarılı sonuçlara ulaştığını ve aynı zamanda işleme maliyeti bakımından da düşük olduğunu sonucuna ulaştık.
References
- Abedin, Z., Dhar, P., Hossenand, M. K., ve Deb, K. (2017). Traffic Sign Detection and Recognition Using Fuzzy Segmentation Approach and Artificial Neural Network Classifier Respectively. International Conference on Electrical, Computer and Communication Engineering, (s. 518-523). Cox's Bazar.
- Agrawal, S., ve Chaurasiya, R. K. (2017). Automatic Traffic Sign Detection and Recognition Using Moment Invariants and Support Vector Machine. International conference on Recent Innovations is Signal Processing and Embedded Systems.
- Arcos-García, A., Álvarez-García, J. A., ve Soria-Morillo, L. M. (2018). Evaluation of deep neural networks for traffic sign detection systems. Neurocomputing, 316, 332-344.
- Berkaya, S. K., Gunduz, H., Ozsen, O., Akinlar, C., ve Gunal, S. (2016). On circular traffic sign detection and recognition. Expert Systems With Applications, 48, 67-75.
- Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., . . . Schiele, B. (2016). The Cityscapes Dataset for Semantic Urban Scene Understanding. IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
- Ellahyani, A., Ansari, M. E., ve Jaafari, I. E. (2016). Traffic sign detection and recognition based on random forests. Applied Soft Computing, 46, 805-815.
- Fleyeh, H., ve Dougherty, M. (2005). Road and traffic sign detection and recognition. 16th Mini-EURO Conference and 10th Meeting of EWGT, (s. 644-653).
- Ghimire, S., ve Wang, H. (2012). Classification of image pixels based on minimum distance and hypothesis testing. Computational Statistics & Data Analysis, 56, 2273-2287.
- Houben, S., Stallkamp, J., Salmen, J., Schlipsing, M., ve Igel, C. (2013). Detection of Traffic Signs in Real-World Images: The German Traffic Sign Detection Benchmark. International Joint Conference on Neural Networks (IJCNN).
- Jung, S., Lee, U., Jung, J., ve Shim, D. H. (2016). Real-time Traffic Sign Recognition System with Deep Convolutional Neural Network. 13th International Conference on Ubiquitous Robots and Ambient Intelligence. Urai.
- Li, Y., Møgelmose, A., ve Trivedi, M. M. (2016). Pushing the “Speed Limit”: High-Accuracy U.S. Traffic Sign Recognition with Convolutional Neural Networks. IEEE Transactions on Intelligent Vehicles, 167-176.
- Liao, S. M., ve Akritas, M. (2007). Test-based classification: A linkage between classification and statistical testing. Statistics & Probability Letters, 77, 1269-1281.
- Liu, X., Deng, Z., ve Yang, Y. (2019). Recent progress in semantic image segmentation. Artificial Intelligence Review, 1089–1106.
- Malik, Z., ve Siddiqi, I. (2014). Detection and Recognition of Traffic Signs from Road Scene Images. 12th International Conference on Frontiers of Information Technology, (s. 330-335).
- Mathias, M., Timofte, R., Benenson, R., ve Van Gool, L. (2013). Traffic sign recognition — How far are we from the solution? International Joint Conference on Neural Networks (IJCNN), (s. 1-8). Dallas.
- Paszke, A., Chaurasia, A., Kim, S., ve Culurciello, E. (2016). Enet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation. arXiv: 1606.02147.
- Rahman, Q. M., Sünderhauf, N., ve Dayoub, F. (2019). Did You Miss the Sign? A False Negative Alarm System for Traffic Sign Detectors. arXiv:1903.06391.
- Serna, C. G., ve Ruichek, Y. (2018). Classification of Traffic Signs: The European Dataset. IEEE Access, 4, 78136-78148.
- Song, S., Que, Z., Hou, J., Du, S., ve Song, Y. (2019). An efficient convolutional neural network for small traffic sign detection. Journal of Systems Architecture, 97, 269-277.
- Stallkamp, J., Schlipsing, M., Salmen, J., ve Igel, C. (2011). The German Traffic Sign Recognition Benchmark: A multi-class classification competition. International Joint Conference on Neural Networks, (s. 1453-1460).
- Stallkamp, J., Schlipsing, M., Salmen, J., ve Igel, C. (2012). Man vs. Computer: Benchmarking Machine Learning Algorithms for Traffic Sign Recognition. Neural networks, 323–332.
- Sun, D., ve Ho, M. (2011). Image Segmentation via Total Variation and Hypothesis Testing Methods.
- Suzuki, S., ve Abe, K. (1985). Topological structural analysis of digitized binary images by border following. Computer Vision, Graphics, and Image Processing, 30, 32–46.
- Temel, D., Alshawi, T., Chen, M.-H., ve AlRegib, G. (2019). Challenging Environments for Traffic Sign Detection: Reliability Assessment under Inclement Conditions. arXiv:1902.06857.
- Torres, L. T., Paixao, T. M., Berriel, R. F., De Souza, A. F., Badue, C., Sebe, N., ve Oliveira-Santos, T. (2019). Effortless Deep Training for Traffic Sign Detection Using Templates and Arbitrary Natural Images. International Joint Conference on Neural Networks. Budapest.
- Wong, A., Shafiee, M. J., ve Jules, M. S. (2018). MicronNet: A Highly Compact Deep Convolutional Neural Network Architecture for Real-Time Embedded Traffic Sign Classification. IEEE Access, 6, 59803–59810.
- Xu, X., Jin, J., Zhang, S., Zhang, L., Pu, S., ve Chen, Z. (2019). Smart data driven traffic sign detection method based on adaptive color threshold and shape symmetry. Future Generation Computer Systems, 94, 381-391.
- Yuan, X., Guo, J., Hao, X., ve Chen, H. (2015). Traffic Sign Detection via Graph-Based Ranking and Segmentation Algorithms. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 45, 1509-1521.
Year 2020,
Volume: 3 Issue: 2, 140 - 157, 30.11.2020
Mert Çetinkaya
,
Tankut Acarman
References
- Abedin, Z., Dhar, P., Hossenand, M. K., ve Deb, K. (2017). Traffic Sign Detection and Recognition Using Fuzzy Segmentation Approach and Artificial Neural Network Classifier Respectively. International Conference on Electrical, Computer and Communication Engineering, (s. 518-523). Cox's Bazar.
- Agrawal, S., ve Chaurasiya, R. K. (2017). Automatic Traffic Sign Detection and Recognition Using Moment Invariants and Support Vector Machine. International conference on Recent Innovations is Signal Processing and Embedded Systems.
- Arcos-García, A., Álvarez-García, J. A., ve Soria-Morillo, L. M. (2018). Evaluation of deep neural networks for traffic sign detection systems. Neurocomputing, 316, 332-344.
- Berkaya, S. K., Gunduz, H., Ozsen, O., Akinlar, C., ve Gunal, S. (2016). On circular traffic sign detection and recognition. Expert Systems With Applications, 48, 67-75.
- Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., . . . Schiele, B. (2016). The Cityscapes Dataset for Semantic Urban Scene Understanding. IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
- Ellahyani, A., Ansari, M. E., ve Jaafari, I. E. (2016). Traffic sign detection and recognition based on random forests. Applied Soft Computing, 46, 805-815.
- Fleyeh, H., ve Dougherty, M. (2005). Road and traffic sign detection and recognition. 16th Mini-EURO Conference and 10th Meeting of EWGT, (s. 644-653).
- Ghimire, S., ve Wang, H. (2012). Classification of image pixels based on minimum distance and hypothesis testing. Computational Statistics & Data Analysis, 56, 2273-2287.
- Houben, S., Stallkamp, J., Salmen, J., Schlipsing, M., ve Igel, C. (2013). Detection of Traffic Signs in Real-World Images: The German Traffic Sign Detection Benchmark. International Joint Conference on Neural Networks (IJCNN).
- Jung, S., Lee, U., Jung, J., ve Shim, D. H. (2016). Real-time Traffic Sign Recognition System with Deep Convolutional Neural Network. 13th International Conference on Ubiquitous Robots and Ambient Intelligence. Urai.
- Li, Y., Møgelmose, A., ve Trivedi, M. M. (2016). Pushing the “Speed Limit”: High-Accuracy U.S. Traffic Sign Recognition with Convolutional Neural Networks. IEEE Transactions on Intelligent Vehicles, 167-176.
- Liao, S. M., ve Akritas, M. (2007). Test-based classification: A linkage between classification and statistical testing. Statistics & Probability Letters, 77, 1269-1281.
- Liu, X., Deng, Z., ve Yang, Y. (2019). Recent progress in semantic image segmentation. Artificial Intelligence Review, 1089–1106.
- Malik, Z., ve Siddiqi, I. (2014). Detection and Recognition of Traffic Signs from Road Scene Images. 12th International Conference on Frontiers of Information Technology, (s. 330-335).
- Mathias, M., Timofte, R., Benenson, R., ve Van Gool, L. (2013). Traffic sign recognition — How far are we from the solution? International Joint Conference on Neural Networks (IJCNN), (s. 1-8). Dallas.
- Paszke, A., Chaurasia, A., Kim, S., ve Culurciello, E. (2016). Enet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation. arXiv: 1606.02147.
- Rahman, Q. M., Sünderhauf, N., ve Dayoub, F. (2019). Did You Miss the Sign? A False Negative Alarm System for Traffic Sign Detectors. arXiv:1903.06391.
- Serna, C. G., ve Ruichek, Y. (2018). Classification of Traffic Signs: The European Dataset. IEEE Access, 4, 78136-78148.
- Song, S., Que, Z., Hou, J., Du, S., ve Song, Y. (2019). An efficient convolutional neural network for small traffic sign detection. Journal of Systems Architecture, 97, 269-277.
- Stallkamp, J., Schlipsing, M., Salmen, J., ve Igel, C. (2011). The German Traffic Sign Recognition Benchmark: A multi-class classification competition. International Joint Conference on Neural Networks, (s. 1453-1460).
- Stallkamp, J., Schlipsing, M., Salmen, J., ve Igel, C. (2012). Man vs. Computer: Benchmarking Machine Learning Algorithms for Traffic Sign Recognition. Neural networks, 323–332.
- Sun, D., ve Ho, M. (2011). Image Segmentation via Total Variation and Hypothesis Testing Methods.
- Suzuki, S., ve Abe, K. (1985). Topological structural analysis of digitized binary images by border following. Computer Vision, Graphics, and Image Processing, 30, 32–46.
- Temel, D., Alshawi, T., Chen, M.-H., ve AlRegib, G. (2019). Challenging Environments for Traffic Sign Detection: Reliability Assessment under Inclement Conditions. arXiv:1902.06857.
- Torres, L. T., Paixao, T. M., Berriel, R. F., De Souza, A. F., Badue, C., Sebe, N., ve Oliveira-Santos, T. (2019). Effortless Deep Training for Traffic Sign Detection Using Templates and Arbitrary Natural Images. International Joint Conference on Neural Networks. Budapest.
- Wong, A., Shafiee, M. J., ve Jules, M. S. (2018). MicronNet: A Highly Compact Deep Convolutional Neural Network Architecture for Real-Time Embedded Traffic Sign Classification. IEEE Access, 6, 59803–59810.
- Xu, X., Jin, J., Zhang, S., Zhang, L., Pu, S., ve Chen, Z. (2019). Smart data driven traffic sign detection method based on adaptive color threshold and shape symmetry. Future Generation Computer Systems, 94, 381-391.
- Yuan, X., Guo, J., Hao, X., ve Chen, H. (2015). Traffic Sign Detection via Graph-Based Ranking and Segmentation Algorithms. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 45, 1509-1521.