Low-Altitude Aerial Video Surveillance via One-Class SVM Anomaly Detection from Textural Features in UAV Images

Avola, Danilo; Cinque, Luigi; Mambro, A. Di; Diko, Anxhelo; Fagioli, Alessio; Foresti, Gian Luca; Marini, Marco Raoul; Mecca, Alessio; Pannone, Daniele

doi:10.3390/info13010002

Cited by 24 publications

(14 citation statements)

References 108 publications

(114 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM is used to train and recognize the extracted HOG features. Since SVM has high recognition accuracy even for small sample training results, it avoids a large number of manually labeled image categories, and, the fast-training speed saves a lot of time and improves efficiency [11,12]. In addition, HOG is used as a training feature, and as a gradient feature, HOG features themselves have high discriminative properties [13,14].…”

Section: Geographic Image Feature Training and Recognitionmentioning

confidence: 99%

Automatic Geographic Information Classification Method Based on SVM for Site Selection of Stretch Field

Jing

Wang²,

Yang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

In tension stringing construction, the site selection of the stretch field is related to the efficiency, safety and cost of the whole wire construction. Surrounding geographic information impacts the site selection extremely seriously. Accurate grasp of the surrounding geographic information along the transmission line is the key to ensure the optimal site selection of the stretch field. Aiming at the geographic image data along the transmission lines collected by the UAV, firstly, the map data is blocked using the image gradient information and the geometric topological relationship in the image; then, the histogram features of the gradient direction are extracted from the blocked data; finally, the support vector machine is used to train the extracted features, and the map data is classified according to the model obtained from the training, and finally the automatic classification and recognition of the geographic information along the transmission lines is obtained. The proposed method can achieve high recognition accuracy by using small samples for training. Through practice, the proposed method can accurately classify map data, thus providing a reliable guarantee for determining the surrounding geographic influences when selecting a site for a stretch field.

show abstract

Section: Geographic Image Feature Training and Recognitionmentioning

confidence: 99%

Automatic Geographic Information Classification Method Based on SVM for Site Selection of Stretch Field

Jing

Wang²,

Yang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…Considering that the scene exhibits an evident distinction when a UAV is present, using abnormality detection for UAV identification is a novel approach. There has been research using abnormal detection to identify UAVs [ 23 ], but it is still relatively scarce. However, there is extensive research on abnormal detection and these results can be used as an important reference.…”

Section: Related Workmentioning

confidence: 99%

CUDM: A Combined UAV Detection Model Based on Video Abnormal Behavior

Cai

Song

Xiong

et al. 2022

Sensors

View full text Add to dashboard Cite

The widespread use of unmanned aerial vehicles (UAVs) has brought many benefits, particularly for military and civil applications. For example, UAVs can be used in communication, ecological surveys, agriculture, and logistics to improve efficiency and reduce the required workforce. However, the malicious use of UAVs can significantly endanger public safety and pose many challenges to society. Therefore, detecting malicious UAVs is an important and urgent issue that needs to be addressed. In this study, a combined UAV detection model (CUDM) based on analyzing video abnormal behavior is proposed. CUDM uses abnormal behavior detection models to improve the traditional object detection process. The work of CUDM can be divided into two stages. In the first stage, our model cuts the video into images and uses the abnormal behavior detection model to remove a large number of useless images, improving the efficiency and real-time detection of suspicious targets. In the second stage, CUDM works to identify whether the suspicious target is a UAV or not. Besides, CUDM relies only on ordinary equipment such as surveillance cameras, avoiding the use of expensive equipment such as radars. A self-made UAV dataset was constructed to verify the reliability of CUDM. The results show that CUDM not only maintains the same accuracy as state-of-the-art object detection models but also reduces the workload by 32%. Moreover, it can detect malicious UAVs in real-time.

show abstract

“…We adopted the classical support vector machine (SVM) method [30] and gradient boosting decision tree (GBDT) algorithm of XGBoost [31] and LightGBM [32] to construct prediction models, and finally compared and analyzed the prediction results of different models. Support vector regression (SVR) is an application of SVMs in regression analysis, which can perform regression analysis and prediction on time series and is widely used in time series prediction prob-lems such as change and anomaly detection from textural features [33], fake stereo audio identification [34], or drowsiness estimation [35]. XGBoost is an integrated decision tree algorithm in which new trees can correct the results of existing trees in the model so that the model can be made satisfactory by continuously adding decision trees.…”

Section: Classical Machine Learning Modelsmentioning

confidence: 99%

DNS Request Log Analysis of Universities in Shanghai: A CDN Service Provider’s Perspective

et al. 2022

View full text Add to dashboard Cite

Understanding the network usage patterns of university users is very important today. This paper focuses on the research of DNS request behaviors of university users in Shanghai, China. Based on the DNS logs of a large number of university users recorded by CERNET, we conduct a general analysis of the behavior of network browsing from two perspectives: the characteristics of university users’ behavior and the market share of CDN service providers. We also undertake experiments on DNS requests patterns for CDN service providers using different prediction models. Firstly, in order to understand the university users’ Internet access patterns, we select the top seven universities with the most DNS requests and reveal the characteristics of different university users. Subsequently, to obtain the market share of different CDN service providers, we analyze the overall situation of the traffic distribution among different CDN service providers and its dynamic evolution trend. We find that Tencent Cloud and Alibaba Cloud are leading in both IPv4 and IPv6 traffic. Baidu Cloud has close to 15% in IPv4 traffic, but almost no fraction in IPv6 traffic. Finally, for the characteristics of different CDN service providers, we adopt statistical models, traditional machine learning models, and deep learning models to construct tools that can accurately predict the change in request volume of DNS requests. The conclusions obtained in this paper are beneficial for Internet service providers, CDN service providers, and users.

show abstract

Low-Altitude Aerial Video Surveillance via One-Class SVM Anomaly Detection from Textural Features in UAV Images

Cited by 24 publications

References 108 publications

Automatic Geographic Information Classification Method Based on SVM for Site Selection of Stretch Field

Automatic Geographic Information Classification Method Based on SVM for Site Selection of Stretch Field

CUDM: A Combined UAV Detection Model Based on Video Abnormal Behavior

DNS Request Log Analysis of Universities in Shanghai: A CDN Service Provider’s Perspective

Contact Info

Product

Resources

About