Journals / International Journal of Innovative Engineering Applications / 2021 / Cilt: 5 - Sayı: 2

Anomaly Detection from Traffic Video Analysis Data and Classification with Differential Evolution Algorithm Extreme Learning Machine

Trafik Video Analiz Verilerinden Anomali Tespit ve Diferansiyel Gelişim Algoritması Uç Öğrenme Makinesi ile Sınıflandırma

Pages
115–124
Publication Date
2022-10-19
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results obtained with the developed method have been compared with the results of other machine learning methods (Naive Bayes, Support Vector Machine and Artificial Neural Network) and the performance was the highest with 100%.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The process of obtaining meaningful and interpretable information about the object by detecting and tracking moving objects in videos obtained by cameras can be defined as video analysis.
📊
Key Findings & Data:The analysis of highway traffic flow videos can be expressed as a detection and warning system of the behavior (anomaly) of moving objects (vehicle and pedestrian) outside the rules determined for the road followed.
💡
Conclusion & Impact:The results obtained with the developed method have been compared with the results of other machine learning methods (Naive Bayes, Support Vector Machine and Artificial Neural Network) and the performance was the highest with 100%.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Akıllı Trafik İzleme Sistemi · Anomali Durum Tespit · Diferansiyel Gelişim Algoritması

Abstract

The process of obtaining meaningful and interpretable information about the object by detecting and tracking moving objects in videos obtained by cameras can be defined as video analysis. The analysis of highway traffic flow videos can be expressed as a detection and warning system of the behavior (anomaly) of moving objects (vehicle and pedestrian) outside the rules determined for the road followed. In this study, the anomaly behaviors of moving objects in highway traffic flow videos have been determined and the data has been classified by machine learning methods. The study consists of three stages. In the first stage, Gaussian Mixing Model, one of the background / foreground segmentation methods, and Kalman Filter-Macar Algorithm methods have been used to track the object on video. In the second step, location, time, velocity and shape data have been obtained by using the coordinate information of the detected object. With the unique algorithm-based feature extraction software, a table of properties including anomaly conditions for objects has been created. In the last stage, the data in the feature table; Classification of the Extreme Learning Machine method has been made with the Differential Evolution Algorithm - Extreme Learning Machine method developed by optimizing the number of neurons in the hidden layer and activation functions with the Differential Evolution Algorithm. The results obtained with the developed method have been compared with the results of other machine learning methods (Naive Bayes, Support Vector Machine and Artificial Neural Network) and the performance was the highest with 100%.

Özet

Kameralar ile elde edilen videolarda hareketli nesne tespit ve takibi ile nesne hakkında anlamlı ve yorumlanabilir bilgi elde edilmesi süreci, video analizi olarak tanımlanabilir. Karayolu trafik akış videolarının analizi ise hareket eden nesnelerin (taşıt ve yaya) izlenen yol için belirlenen kurallar dışında davranış sergilemeleri (anomali) durumunun tespit ve uyarı sistemi olarak ifade edilebilir. Bu çalışmada karayolu trafik akış videolarında hareket eden nesnelerin anomali davranışlarını tespit ederek verileri makine öğrenmesi yöntemleri ile sınıflandırma yapılmıştır. Çalışma üç aşamadan oluşmaktadır. Birinci aşamada hareketli nesnenin tespiti için arka/ön plan bölütleme yöntemlerinden Gauss Karışım Modeli (GKM) ve nesnenin videoda takibi yapılabilmesi için Kalman Filtre-Macar Algoritması yöntemleri kullanılmıştır. İkinci aşamada tespit edilen nesneye ait koordinat bilgilerini kullanarak konum, süre, hız ve şekil verileri elde edilmiştir. Hazırlanan özgün algoritmaya dayalı özellik çıkarma yazılımı ile nesneler için anomali durumlarını içeren özellikler tablosu oluşturulmuştur. Son aşamada ise özellik tablosundaki veriler; Uç Öğrenme Makine (UÖM) yönteminin, gizli katmandaki nöron sayıları ve aktivasyon fonksiyonları Diferansiyel Gelişim Algoritması (DGA) ile optimize edilerek geliştirilen DGAUÖM yöntemi ile sınıflandırma yapılmıştır. Geliştirilen yöntem ile elde edilen sonuçlar diğer makine öğrenme yöntem (Naive Bayes, Destek Vektör Makinesi ve Yapay Sinir Ağı) sonuçları ile karşılaştırılmış ve başarım %100 ile en yüksek çıkmıştır.

🔖

Cite This Article

AVCI, D., SÖNMEZ, Y. (2022). Anomaly Detection from Traffic Video Analysis Data and Classification with Differential Evolution Algorithm Extreme Learning Machine. International Journal of Innovative Engineering Applications, Cilt: 5 - Sayı: 2, 115–124. https://ajindex.com/articles/541210

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (1)

  1. Peker, M., & Zengin, A. 2011. Gerçek zamanlı harekete duyarlı bir görüntü tanıma sistemi. In 6th International Advanced Technologies Symposium. [2]Hsia, C. H., & Guo, J. M. 2014. Efficient modified directional lifting-based discrete wavelet transform for moving object detection. Signal Processing, 96, 138–152. [3] Oral, M., & Deniz, U. Ardışık görüntülerde hareket analizi motion detection in moving pictures. [4] www.nuuo.com/enews/201210_enews/images/2012010_Flyer_IVS_V1.0_A4_en.pdf 15 Ocak 2018. [5] Borji, A., Cheng, M. M., Jiang, H., & Li, J. 2014. Salient Object Detection: A Survey. Eprint arXiv, 1–26. [6] Cheung, S. C. S., & Kamath, C. 2004. Robust techniques for background subtraction in urban traffic video. SPIE (Vol. 5308, No. 1, pp. 881-892). [7] Mahamuni, P. D., Patil, R. P., & Thakar, H. S. 2014. Moving object detection using background subtraction algorithm using Simulink. International Journal of Research in Engineering and Technology (IJRET), 3(6), 594-598. [8] Umut Orhan “Makine Öğrenmesi ders notları”, Çukurova Üniversitesi http://bmb.cu.edu.tr/uorhan/MLearn.htm 02.08.2015. [9] Jodoin, J. P., Bilodeau, G. A., & Saunier, N. 2014. Urban Tracker: Multiple object tracking in urban mixed traffic. In 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014 (pp. 885–892). [10] Mohan, A. S., & Resmi, R. 2014. Video image processing for moving object detection and segmentation using background subtraction. In Computational Systems and Communications (ICCSC), 2014 First International Conference on (pp. 288-292). IEEE. [11] Zhu, Y., Nayak, N. M., & Roy-Chowdhury, a K. 2013. Context-Aware Activity Recognition and Anomaly Detection in Video. Selected Topics in Signal Processing, IEEE Journal of, 7(1), 91–101. [12] Bas, E., Tekalp, a. M., & Salman, F. S. 2007. Automatic Vehicle Counting from Video for Traffic Flow Analysis. 2007 IEEE Intelligent Vehicles Symposium, 392–397. [13] Buch, N., Velastin, S. A., & Orwell, J. 2011. A review of computer vision techniques for the analysis of urban traffic. IEEE Transactions on Intelligent Transportation Systems. [14] Morris, B. T., & Trivedi, M. M. 2008. Learning, modeling, and classification of vehicle track patterns from live video. IEEE Transactions on Intelligent Transportation Systems, 9(3), 425–437. [15] Öz, K., Görgünoğlu, S., 2016 “Video Gözetim Sistemlerinde Anomali Tespiti Üzerine Bir Derleme” El-Cezerî Fen ve Mühendislik Dergisi, 3(3);506-512. [16] Çandar, C., & Gerek, Ö. N. Yol ü