Abnormal Event Detection Using Convolutional Neural Networks and 1-Class SVM classifier

Bouindour, Samir; Hittawe, Mohamad Mazen; Mahfouz, Sandy; Snoussi, Hichem

doi:10.1049/ic.2017.0040

Cited by 22 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Owing to this limitation, there are limited results available for comparison from other deep-learning approaches, applied to similar datasets. We do, however, provide comparisons of our results with one state-of-the-art method proposed by Bouindour et al 39 We also demonstrate the suitability of our classifier based on the GDMD as compared to 1-SVM. One significant advantage of the GDMD-based algorithm proposed in this paper is the flexibility to incorporate additional environmental variables to capture the extreme events that we plan to do in our future work.…”

Section: Discussionmentioning

confidence: 73%

“…The experimentation results suggest that the Gaussian model of Mahalanobis distances outperformed 1-SVM in terms of sensitivity and specificity for ASST detection in the Red Sea. We have further compared the results of the proposed method with one state-of-the-art method 39 to demonstrate the relevance of the proposed approach. Our results show a better performance in term of average EER for ASST localization in the Red Sea.…”

Section: Discussionmentioning

confidence: 99%

“…Our algorithm of ASST localization achieves an average EER of 28.2. We compare our method with the state-of-the-art method proposed by Bouindour et al 39 They used a deep-learning approach based on Alexnet 40 for feature extraction from video sequences, and then used 1-SVM classifier for detecting anomalies. Our method is more performant in terms of average EER (Table 3) and for each of the fiveyear sets from 1985 to 2015 (Fig.…”

Section: Performance Evaluation Of Abnormal Seamentioning

confidence: 99%

See 2 more Smart Citations

Abnormal events detection using deep neural networks: application to extreme sea surface temperature detection in the Red Sea

et al. 2019

View full text Add to dashboard Cite

We present a method based on deep learning for detecting and localizing abnormal/extreme events in sea surface temperature (SST) of the Red Sea images using training samples of normal events only. The method operates in two stages; the first one involves features extraction from each patch of the SST input image using the first two convolutional layers extracted from a pretrained convolutional neural network. In the second stage, two methods are used for training the model from the normal training data. The first method uses oneclass support vector machine (1-SVM) classifier that allows a fast and robust abnormal detection in the presence of outliers in the training dataset. In the second method, a Gaussian model is defined on the Mahalanobis distances between all normal training data. Experimental tests are conducted on satellite-derived SST data of the Red Sea spanning for a period of 31 years (1985-2015). Our results suggest that the Gaussian model of Mahalanobis distances outperformed 1-SVM by providing better performance in terms of sensitivity and specificity.

show abstract

Section: Discussionmentioning

confidence: 73%

Section: Discussionmentioning

confidence: 99%

Section: Performance Evaluation Of Abnormal Seamentioning

confidence: 99%

See 1 more Smart Citation

Abnormal events detection using deep neural networks: application to extreme sea surface temperature detection in the Red Sea

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Latent representation techniques use parts of deep neural network in order to retrieve the latent feature vectors of the input images and then train a one class classifier for anomaly detection purposes. In their work (Bouindour et al, 2017), the authors used two convolutional layers from the convolutional neural network AlexNet to extract features vectors, which are used to train a one class support vector machine. (Chalapathy et al, 2018) used the latent representation of image data produced by the encoder part of a convolutional autoencoder to train a one class neural network that outputs an anomaly score.…”

Section: Related Workmentioning

confidence: 99%

Usage Identification of Anomaly Detection in an Industrial Context

Zoghlami

Kurrek

Jocas

et al. 2019

Proc. Int. Conf. Eng. Des.

View full text Add to dashboard Cite

The use of flexible and autonomous robotics systems is the solution for the automation task of the production and intra-logistics environments. This dynamic context requires the robot to be aware of its surroundings through the whole task, also after accomplishing the gripping action. We present an anomaly detection approach based on unsupervised learning and reconstruction fidelity of image data. We design our method to enhance the dynamic environment perception of robotics systems and apply it in a palletizing robot, in order to perceive and detect changes to its surrounding and process after the gripping step. Our proposed approach achieves the performance targeted by the considered industrial requirements.

show abstract

“…With the development of machine learning studies, various approaches based on deep learning have achieved remarkable progress in abnormal event detection. For example, convolutional neural networks (CNNs) [17], recurrent neural networks [11] and other deep learning models can learn better feature representation than hand-crafted feature modeling. It is conductive to determinate the occurrence of abnormal event in video sequence.…”

Section: Introductionmentioning

confidence: 99%

Abnormal Event Detection via Feature Expectation Subgraph Calibrating Classification in Video Surveillance Scenes

Deng

et al. 2020

IEEE Access

View full text Add to dashboard Cite

At present, the existing abnormal event detection models based on deep learning mainly focus on data represented by a vectorial form, which pay little attention to the impact of the internal structure characteristics of feature vector. In addition, a single classifier is difficult to ensure the accuracy of classification. In order to address the above issues, we propose an abnormal event detection hybrid modulation method via feature expectation subgraph calibrating classification in video surveillance scenes in this paper. Our main contribution is to calibrate the classification of a single classifier by constructing feature expectation subgraphs. First, we employ convolutional neural network and long short-term memory models to extract the spatiotemporal features of video frame, and then construct the feature expectation subgraph for each key frame of every video, which could be used to capture the internal sequential and topological relational characteristics of structured feature vector. Second, we project expectation subgraphs on the sparse vector to combine with a support vector classifier to calibrate the results of a linear support vector classifier. Finally, the experiments on a common dataset named UCSDped1 and a coal mining video dataset in comparison with some existing works demonstrate that the performance of the proposed method is better than several the state-of-the-art approaches. INDEX TERMS Abnormal event detection, feature expectation subgraph, calibrating classification, sequential and topological relational characteristics.

show abstract

Abnormal Event Detection Using Convolutional Neural Networks and 1-Class SVM classifier

Cited by 22 publications

References 0 publications

Abnormal events detection using deep neural networks: application to extreme sea surface temperature detection in the Red Sea

Abnormal events detection using deep neural networks: application to extreme sea surface temperature detection in the Red Sea

Usage Identification of Anomaly Detection in an Industrial Context

Abnormal Event Detection via Feature Expectation Subgraph Calibrating Classification in Video Surveillance Scenes

Contact Info

Product

Resources

About