Real-World Anomaly Detection Using
Deep Learning
Unnati Koppikar, C. Sujatha, Prakashgoud Patil and Uma Mudenagudi
Abstract In this paper, we have carried out a comparative study on two deep learning
models for detecting real-world anomalies in surveillance videos. Anomalous event
is the one which deviates from the normal behavior. The anomalies considered are
related to thefts such as robbery, burglary, stealing, and shoplifting. A framework is
set up using supervised learning approach to train the models using the weakly labeled
videos. The deep learning models considered are VGG-16 and inception model which
are trained with both anomalous and normal videos to detect any anomalous activity
in the video frame. UCF-Crime dataset is used which comprises long, untrimmed
surveillance videos. The deep learning models are evaluated using various metrics.
The experimental results show that the Inception V3 model performs significantly
better in detecting the anomalies as compared to the VGG-16 model with an accuracy
of 94.54%.
Keywords Surveillance · Anomaly detection · Theft · Deep learning ·
Convolutional Neural Networks · VGG-16 model · Inception V3 model
U. Koppikar (B ) · C. Sujatha (B )
Department of CSE, KLE Technological University, Hubballi, India
e-mail: unnatikoppikar@gmail.com
C. Sujatha
e-mail: sujata_c@kletech.ac.in
P. Patil
Department of MCA, KLE Technological University, Hubballi, India
e-mail: prpatilji@gmail.com
U. Mudenagudi
Department of ECE, KLE Technological University, Hubballi, India
e-mail: uma@kletech.ac.in
© Springer Nature Singapore Pte Ltd. 2020
V. Bhateja et al. (eds.), Intelligent Computing and Communication,
Advances in Intelligent Systems and Computing 1034,
https://doi.org/10.1007/978-981-15-1084-7_32
333