Ship recognition based on Hu invariant moments and convolutional neural network for video surveillance

2020 ◽  
Author(s):  
Yongmei Ren ◽  
Jie Yang ◽  
Qingnian Zhang ◽  
Zhiqiang Guo
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Invariant Moments ◽  
Ship Recognition

scholarly journals Automatic Handgun Detection with Deep Learning in Video Surveillance Images

2021 ◽  
Vol 11 (13) ◽  
pp. 6085
Author(s):  
Jesus Salido ◽  
Vanesa Lomas ◽  
Jesus Ruiz-Santaquiteria ◽  
Oscar Deniz
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Automatic Detection ◽  
Public Spaces ◽  
Detection Methods ◽  
Training Dataset ◽  
Average Precision ◽  
Terrorist Acts

There is a great need to implement preventive mechanisms against shootings and terrorist acts in public spaces with a large influx of people. While surveillance cameras have become common, the need for monitoring 24/7 and real-time response requires automatic detection methods. This paper presents a study based on three convolutional neural network (CNN) models applied to the automatic detection of handguns in video surveillance images. It aims to investigate the reduction of false positives by including pose information associated with the way the handguns are held in the images belonging to the training dataset. The results highlighted the best average precision (96.36%) and recall (97.23%) obtained by RetinaNet fine-tuned with the unfrozen ResNet-50 backbone and the best precision (96.23%) and F1 score values (93.36%) obtained by YOLOv3 when it was trained on the dataset including pose information. This last architecture was the only one that showed a consistent improvement—around 2%—when pose information was expressly considered during training.

Privacy Preserving in Video Surveillance Systems Using Regression Residual Convolutional Neural Network in Private and Public Places

2020 ◽  
Vol 12 (1) ◽  
pp. 39-55
Author(s):  
Hadj Ahmed Bouarara
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Security Policy ◽  
Surveillance Systems ◽  
Surveillance Video ◽  
Public Places ◽  
Private And Public ◽  
The One ◽  
Regression Residual

In recent years, surveillance video has become a familiar phenomenon because it gives us a feeling of greater security, but we are continuously filmed and our privacy is greatly affected. This work deals with the development of a private video surveillance system (PVSS) using regression residual convolutional neural network (RR-CNN) with the goal to propose a new security policy to ensure the privacy of no-dangerous person and prevent crime. The goal is to best meet the interests of all parties: the one who films and the one who is filmed.

scholarly journals Superresolution Reconstruction of Video Based on Efficient Subpixel Convolutional Neural Network for Urban Computing

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Jie Shen ◽  
Mengxi Xu ◽  
Xinyu Du ◽  
Yunbo Xiong
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Optical Flow ◽  
Video Surveillance ◽  
Reconstruction Method ◽  
Motion Information ◽  
Learning Networks ◽  
Urban Computing ◽  
Deep Convolution Network

Video surveillance is an important data source of urban computing and intelligence. The low resolution of many existing video surveillance devices affects the efficiency of urban computing and intelligence. Therefore, improving the resolution of video surveillance is one of the important tasks of urban computing and intelligence. In this paper, the resolution of video is improved by superresolution reconstruction based on a learning method. Different from the superresolution reconstruction of static images, the superresolution reconstruction of video is characterized by the application of motion information. However, there are few studies in this area so far. Aimed at fully exploring motion information to improve the superresolution of video, this paper proposes a superresolution reconstruction method based on an efficient subpixel convolutional neural network, where the optical flow is introduced in the deep learning network. Fusing the optical flow features between successive frames can compensate for information in frames and generate high-quality superresolution results. In addition, in order to improve the superresolution, a superpixel convolution layer is added after the deep convolution network. Finally, experimental evaluations demonstrate the satisfying performance of our method compared with previous methods and other deep learning networks; our method is more efficient.

scholarly journals EARLY ACTION PREDICTION USING VGG16 MODEL AND BIDIRECTIONAL LSTM

2021 ◽  
Vol 9 (1) ◽  
pp. 666-672
Author(s):  
Manju D, Dr. Seetha M, Dr. Sammulal P
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Early Stage ◽  
Early Prediction ◽  
Action Prediction ◽  
Automated Driving ◽  
Time Step ◽  
Driving System ◽  
Early Action

Action prediction plays a key function, where an expected action needs to be identified before the action is completely performed. Prediction means inferring a potential action until it occurs at its early stage. This paper emphasizes on early action prediction, to predict an action before it occurs. In real time scenarios, the early prediction can be very crucial and has many applications like automated driving system, healthcare, video surveillance and other scenarios where a proactive action is needed before the situation goes out of control. VGG16 model is used for the early action prediction which is a convolutional neural network with 16 layers depth. Besides its capability of classifying objects in the frames, the availability of model weights enhances its capability. The model weights are available freely and preferred to used in different applications or models. The VGG-16 model along with Bidirectional structure of Lstm enables the network to provide both backward and forward information at every time step. The results of the proposed approach increased observation ratio ranging from 0.1 to 1.0 compared with the accuracy of GAN model.

scholarly journals Deep Large Margin Nearest Neighbor for Gait Recognition

2021 ◽  
Vol 30 (1) ◽  
pp. 604-619
Author(s):  
Wanjiang Xu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Computational Efficiency ◽  
Nearest Neighbor ◽  
Gait Recognition ◽  
Large Margin ◽  
Extensive Evaluation ◽  
Gait Features ◽  
Gait Feature

Abstract Gait recognition in video surveillance is still challenging because the employed gait features are usually affected by many variations. To overcome this difficulty, this paper presents a novel Deep Large Margin Nearest Neighbor (DLMNN) method for gait recognition. The proposed DLMNN trains a convolutional neural network to project gait feature onto a metric subspace, under which intra-class gait samples are pulled together as small as possible while inter-class samples are pushed apart by a large margin. We provide an extensive evaluation in terms of various scenarios, namely, normal, carrying, clothing, and cross-view condition on two widely used gait datasets. Experimental results demonstrate that the proposed DLMNN achieves competitive gait recognition performances and promising computational efficiency.

Sign in / Sign up

Export Citation Format

Share Document