Video anomaly detection in the surveillance video is one of the essential components of the intelligent video surveillance system. However, anomaly detection remains an ill-defined problem, despite the diverse applications due to its rareness and equivocal nature. A Long Short Term Memory - Variational Autoencoder (LSTM-VAE) model is proposed to detect video anomalies. The model consists of a spatial encoder comprised of convolutional layers, a temporal encoder as well as a decoder comprised of Convolutional LSTM (ConvLSTM), and a spatial decoder consisting of transposed convolution layers. The generative model is trained only on normal video clips with the objective of minimizing the reconstruction error. Then, the trained model is tested on the test video sequences comprised of both normal and abnormal incidents. The reconstruction error corresponding to the test frame sequences having video anomalies will be very high as the model is not trained to reconstruct them. Subsequently, the corresponding frames will have a low regularity score. An appropriate threshold regularity score is set to segregate the anomaly frames from the normal ones. Frames having a regularity score less than the set threshold value are considered as anomalous frames. The model is developed by using one of the publicly available bench-marked video anomaly datasets, i.e., UCSD Ped2. The performance metrics of the proposed model are promising.

A smart university is supposed to be a safe university. At this moment we observe multiple cameras in different locations in the Hall University and rooms to detect suspicious behavior such as violation, larceny or persons in a state of alcohol or drug intoxication. Samples of the video footage is monitored 24/7 by operators in control rooms. Currently the recorded videos are visual assessed after a suspicious event has occurred. There is a requirement for realtime surveillance with smart cameras which can detect, track and analyze suspicious behavior over place and time. The expanding number of cameras requires an enormous measure of observing operators. This paper proposes a distributed intelligent surveillance system based on smart cameras. We seek to improve the Quality of Experience QoE operator side or QoEvideo surveillance expressed in function of i- resource availability constraints, ii- false detection of suspicious behavior, iii- define an optimal perimeter for intrusion detection (subset of cameras, network parameters required . . . ).

Smart Surveillance Systems are becoming an important aspect of our lives, reducing man labour and additionally increasing the accuracy of detection by reducing false positives. Specifically for an ATM, Surveillance system is very crucial because of the transactions happening being sensitive along with that drop-box containing confidential documents like cheques and bank forms. Hence, there is a need to develop a fool-proof system which can handle a lot of load and perform various surveillance tasks. Moreover, the systems also need to have network security to protect the data from being illegally traced and changed. In this paper, we will be reviewing and comparing various smart surveillance system methods which involve various technologies.

Understanding dynamic human behavior based on online video has many applications in security control, crime surveillance, sports, and industrial IoT systems. This paper solves the problem of classifying video data recorded on surveillance cameras in order to identify fragments with instances of shoplifting. It is proposed to use a classifier that is a symbiosis of two neural networks: convolutional and recurrent. The convolutional neural network is used for extraction of features from each frame of the video fragment, and the recurrent network for processing the temporal sequence of processed frames and subsequent classification.

With the development of technology, the technological informationization of the security network video surveillance service industry has become the demand of the times. How to improve the functions of the video surveillance system and build an open security integrated monitoring management platform has become the research point of this article. This article intends to build a video surveillance system based on database technology to meet the multi-functional requirements of the surveillance system. This article mainly uses experimental methods to test the data of the monitoring system designed in this article, and then uses the comparative method to compare the speed of the three methods to calculate the data, and the data results are obtained. According to the experiment, the data processing time of the binary algorithm in the video surveillance system is within 15s. Image detection in database technology uses binary algorithms to operate and analyze it more quickly.

Surveillance is an observation of a place, large areas, behavior, or a variety of activities to acquire information, influence, manage, or guide it. When people talk about surveillance solutions, the growing demand for large area monitoring becomes one of the key trends in the security industry. Surveillance video is used in real-time to watch known threats. Suspicious activities through surveillance video are a major topic in image processing and deep learning research.Residential area security is very much important to people nowadays. The proposed system is concerned with the development of a surveillance video framework in the residential area to detect any type of suspicious robbery activity. This system makes effective use of deep learning techniques of yolo, this includes techniques like object detection and eventually identifying the actions required to prevent robberies.Surveillance cameras are used here to remotely monitor a residential area or building by transmitting recorded images or videos to a control station to thwart suspicious activities. As a result, deep learning techniques are employed to achieve outstanding detection of suspicious actions that yielded positive results..

In recent years, in order to continuously promote the construction of safe cities, security monitoring equipment has been widely used all over the country. How to use computer vision technology to realize effective intelligent analysis of violence in video surveillance is very important to maintain social stability and ensure people s life and property safety. Video surveillance system has been widely used because of its intuitive and convenient advantages. However, the existing video monitoring system has relatively single function, and generally only has the functions of monitoring video viewing, query and playback. In addition, relevant researchers pay less attention to the complex abnormal behavior of violence, and relevant research often ignores the differences between violent behaviors in different scenes. At present, there are two main problems in video abnormal behavior event detection: the video data of abnormal behavior is less and the definition of abnormal behavior in different scenes cannot be clearly distinguished. The main existing methods are to model normal behavior events first, and then define videos that do not conform to the normal model as abnormal, among which the learning method of video space-time feature representation based on deep learning shows a good prospect. In the face of massive surveillance videos, it is necessary to use deep learning to identify violent behaviors, so that the machine can learn to identify human actions, instead of manually monitoring camera images to complete the alarm of violent behaviors. Network training mainly uses video data set to identify network training.

As an important component of security systems, the number of video surveillance systems is growing rapidly year by year. However, video surveillance systems often have many network security problems, and there is no perfect solution at present. To address these security issues, we propose a TPM-based security enhancement design for video surveillance systems. We enhance the security of the video surveillance system from the perspective of its own environmental security, video data security and device authentication, combined with the TPM module s trusted metrics, trusted authentication and key management mechanisms. We have developed and implemented a prototype video surveillance system and conducted experiments. The experimental results show that the framework we designed can greatly enhance the security of the video surveillance system while ensuring performance.

Wearables Security 2022 - In the twenty-first century, given the worldwide situation, the first concern of any female is her personal protection. Women Labor Day and night to sustain themselves and their families. These women are more susceptible to attacks and assaults, and their security and safety are paramount issues. This technique proposed several new goods to safeguard women. Among the products that may be employed is a smart jacket for women s safety. The proposed approach also includes features to send alert notification to family members with Geo location live tracking and live camera video streaming placed on the jacket for the emergency attention when women are not secure. This gadget is an appeal to all women to earn the right to a safe and secure planet.

Wearables Security 2022 - In aura and era of the Internet of Things (IoT) and the fourth industrial revolution, modern wearable electronic devices and their communication networks are marching into every corner of modern society and changing every aspect of our daily life. Thus, the progress of digitalization including miniaturization of sensor and wearable technology and its growing importance of physical and psychological wellbeing have a tremendous impact on almost all consumer goods from wearable to nonwearable industries. Different types of signals are used in communication between the devices for wireless transmission of data. such as Radio Frequency, Infrared, and Lightwave Transmissions. Wearable devices are becoming a hot topic in many fields such as medical, fashion, education, etc. Digital dependency of WIoT devices, introduced new security challenges, and vulnerabilities. This research is focused on Fitness Wearable Technology Devices Security and Privacy Vulnerability Analysis and highlights the importance of this topic by revealing the potential security concerns. Fog Computing, Sidera and Blockchain technologies were researched as Security Techniques to enhance security and efficiency while providing access to medical and personal records.

Wearables Security 2022 - Mobile devices such as smartphones are increasingly being used to record personal, delicate, and security information such as images, emails, and payment information due to the growth of wearable computing. It is becoming more vital to employ smartphone sensor-based identification to safeguard this kind of information from unwanted parties. In this study, we propose a sensor-based user identification approach based on individual walking patterns and use the sensors that are pervasively embedded into smartphones to accomplish this. Individuals were identified using a convolutional neural network (CNN). Four data augmentation methods were utilized to produce synthetically more data. These approaches included jittering, scaling, and time-warping. We evaluate the proposed identification model’s accuracy, precision, recall, F1-score, FAR, and FRR utilizing a publicly accessible dataset named the UIWADS dataset. As shown by the experiment findings, the CNN with the timewarping approach operates with very high accuracy in user identification, with the lowest false positive rate of 8.80\% and the most incredible accuracy of 92.7\%.

Wearables Security 2022 - Healthcare has become one of the most important aspects of people s lives, resulting in a surge in medical big data. Healthcare providers are increasingly using Internet of Things (IoT)-based wearable technologies to speed up diagnosis and treatment. In recent years, Through the Internet, billions of sensors, gadgets, and vehicles have been connected. One such example is for the treatment and care of patients, technology—remote patient monitoring—is already commonplace. However, these technologies also offer serious privacy and data security problems. Data transactions are transferred and logged. These medical data security and privacy issues might ensue from a pause in therapy, putting the patient s life in jeopardy. We planned a framework to manage and analyse healthcare large data in a safe manner based on blockchain. Our model s enhanced privacy and security characteristics are based on data sanitization and restoration techniques. The framework shown here make data and transactions more secure.

Wearables Security 2022 - One of the biggest new trends in artificial intelligence is the ability to recognise people s movements and take their actions into account. It can be used in a variety of ways, including for surveillance, security, human-computer interaction, and content-based video retrieval. There have been a number of researchers that have presented vision-based techniques to human activity recognition. Several challenges need to be addressed in the creation of a vision-based human activity recognition system, including illumination variations in human activity recognition, interclass similarity between scenes, the environment and recording setting, and temporal variation. To overcome the above mentioned problem, by capturing or sensing human actions with help of wearable sensors, wearable devices, or IoT devices. Sensor data, particularly one-dimensional time series data, are used in the work of human activity recognition. Using 1D-Convolutional Neural Network (CNN) models, this works aims to propose a new approach for identifying human activities. The Wireless Sensor Data Mining (WISDM) dataset is utilised to train and test the 1D-CNN model in this dissertation. The proposed HAR-CNN model has a 95.2\%of accuracy, which is far higher than that of conventional methods.

Wearables Security 2022 - Interoperability remains one of the biggest challenges facing healthcare organizations today. Despite the advancements made through digital transformation and API that allow increased interoperability, patients still have to contend with a different patient portal for each provider they visit. Several health systems are unable to successfully exchange EHR data. API transfer and consolidate patient information including medical history and treatment records across the disparate healthcare systems. Mobile apps use API to gather data from various medical wearables and add the data to a patient’s health record. However, API exposes application logic and sensitive data information giving patient data a window to the World Wide Web and has thus increasingly become a target for attackers. As the need for tighter API security grows, managing APIs becomes more important than ever. The goal of this paper is to provide an overview and discuss research questions that can aid in understanding and building the knowledge base on API data integration and interoperability.

Wearables Security 2022 - As it becomes easier to obtain various data from wearable devices, it is known that biometric and behavioral information must be handled with care. On the other hand, data on the surrounding environment, such as outside temperature, is seen as having a weak relationship with the wearer, and data handling is considered to be a chore. We believe that even data with weak relationships have the potential to infer information about the wearer if a large amount of data is acquired. In this paper, we verify whether it is possible to estimate the wearer’s location from time series data of outside air temperature using only the temperature sensor. We calculated the average absolute error between the temperature data from the wearable device and the same time-series data obtained from the Japan Meteorological Agency, and we evaluated the wearer’s position estimation. It was found that the location where the temperature was sampled appeared at the top of the estimation ranking, and that cities near the sampling location were estimated to be at the high ranking. It was also found that the number of data to be used can be a factor that increases the estimation ranking.

Wearables Security 2022 - As 5G is deployed and applied, a large number of mobile devices have been increasingly deployed on the network. Scenarios such as smartphones, smart car, smart transportation, smart wearable devices, and smart industry are increasingly demanding for networks. And the Internet of Things (IoT), as a new and high technology, will play an important role and generate huge economic benefits. However, IoT security also faces many challenges due to the inherent security vulnerabilities in multiple device interactions and the data also needs more accurate processing. Big data and deep learning have been gradually applied in various industries. Therefore, we have summarized and analyzed the use of big data and deep learning technology to solve the hidden dangers of the IoT security under the consideration of some suggestions and thinking for industry applications.

Wearables Security 2022 - Wearable devices are becoming increasingly popular since they are used in a variety of services. A variety of personal data is collected by the smartwatch. Although devices can give convenience to consumers, there are additional security threats that warn of cybersecurity risks, device penetration, and exploiting vulnerabilities. Devices are prone to attack, and hacking might reveal the acquired data. The lack of authentication and location monitoring, as well as Bluetooth issues and security holes, are all problems in these devices. Although there are security recommendations for such devices, consumers are typically unaware of the risks. The goal of this study is to provide awareness regarding cybersecurity to the common people while using the wearable device.

Wearables Security 2022 - In recent years, technological industry has made a large investment in the design of wearable devices. Wearable devices are attractive to a variety of different age groups within the majority of population. The main reasons for such popularity are related to ease of wear and friendly use, affordable prices with competitive products, as well as providing different services. Usually, wearable devices are collecting different kinds of data. Some of these data are sensitive and personal data of the wearer/user. Such data can be attacked, leaked, misused or edited. Therefore, privacy and security issues are among the main important issues to be considered carefully and discussed clearly when wearable devices are designed and used. Presenting unclear privacy and security strategies will lead the user to mistrust wearable technology with its application. In this research, we present our proposed wearable security protocol to create a secure environment of wearable data and their processing. The main idea of our protocol is to secure the identity of people as well as hiding their sensitive and personal data. Meanwhile, it does not affect the quality of data when applying their mining techniques. The protocol can be used for any kind of data with any application while keeping the user’s privacy and security properties. Moreover, it can be easily understood, implemented, and processed, in addition to any update might be needed.

Vulnerability Detection 2022 - Cross-site scripting attacks, as a means of attack against Web applications, are widely used in phishing, information theft and other fields by unscrupulous people because of their wide targeting and hidden implementation methods. Nevertheless, cross-site scripting vulnerability detection is still in its infancy, with plenty of challenges not yet fully explored. In this paper, we propose Crawler-based Cross Site Scripting Detector, a tool based on crawler technology that can effectively detect stored Cross Site Scripting vulnerabilities and reflected Cross Site Scripting vulnerabilities. Subsequently, in order to verify the effectiveness of the tool, we experim ented this tool with existing tools such as XSSer and Burp Suite by selecting 100 vulnerable websites for the tool s efficiency, false alarm rate and underreporting rate. The results show that our tool can effectively detect Cross Site Scripting vulnerabilities.

Vulnerability Detection 2022 - The power industrial control system is an important part of the national critical Information infrastructure. Its security is related to the national strategic security and has become an important target of cyber attacks. In order to solve the problem that the vulnerability detection technology of power industrial control system cannot meet the requirement of non-destructive, this paper proposes an industrial control vulnerability analysis technology combined with dynamic and static analysis technology. On this basis, an industrial control non-destructive vulnerability detection system is designed, and a simulation verification platform is built to verify the effectiveness of the industrial control non-destructive vulnerability detection system. These provide technical support for the safety protection research of the power industrial control system.

Vulnerability Detection 2022 - Aiming at the fact that the existing source code vulnerability detection methods did not explicitly maintain the semantic information related to the vulnerability in the source code, which made it difficult for the vulnerability detection model to extract the vulnerability sentence features and had a high detection false positive rate, a source code vulnerability detection method based on the vulnerability dependency graph is proposed. Firstly, the candidate vulnerability sentences of the function were matched, and the vulnerability dependency representation graph corresponding to the function was generated by analyzing the multi-layer control dependencies and data dependencies of the candidate vulnerability sentences. Secondly, abstracted the function name and variable name of the code sentences node and generated the initial representation vector of the code sentence nodes in the vulnerability dependency representation graph. Finally, the source code vulnerability detection model based on the heterogeneous graph transformer was used to learn the context information of the code sentence nodes in the vulnerability dependency representation graph. In this paper, the proposed method was verified on three datasets. The experimental results show that the proposed method have better performance in source code vulnerability detection, and the recall rate is increased by 1.50\%\textasciitilde22.27\%, and the F1 score is increased by 1.86\%\textasciitilde16.69\%, which is better than the existing methods.

Vulnerability Detection 2022 - Aiming at the problems of low detection accuracy and poor robustness of the existing zero-speed detection methods, an effective gait cycle segmentation method is adopted and a Bayesian network model based on inertial sensor measurements and kinematics knowledge is introduced to infer the zero-speed interval; The method can effectively reduce the ambiguity of the zero velocity (ZV) boundary. S upport vector machine has the advantages of versatility, simple calculation, high operation efficiency and perfect theory. It is a relatively mature and efficient algorithm in the current network security situation algorithm. And a looped Bayesian network model for probabilistic safety assessment of simple feedback control systems is established.

Vulnerability Detection 2022 - With the booming development of deep learning and machine learning, the use of neural networks for software source code security vulnerability detection has become a hot pot in the field of software security. As a data structure, graphs can adequately represent the complex syntactic information, semantic information, and dependencies in software source code. In this paper, we propose the MPGVD model based on the idea of text classification in natural language processing. The model uses BERT for source code pre-training, transforms graphs into corresponding feature vectors, uses MPNN (Message Passing Neural Networks) based on graph neural networks in the feature extraction phase, and finally outputs the detection results. Our proposed MPGVD, compared with other existing vulnerability detection models on the same dataset CodeXGLUE, obtain the highest detection accuracy of 64.34\%.

Vulnerability Detection 2022 - For the last few decades, the number of security vulnerabilities has been increasing with the development of web applications. The domain of Web Applications is evolving. As a result, many empirical studies have been carried out to address different security vulnerabilities. However, an analysis of existing studies is needed before developing new security vulnerability testing techniques. We perform a systematic mapping study documenting state-of-the-art empirical research in web application security vulnerability detection. The aim is to describe a roadmap for synthesizing the documented empirical research. Existing research and literature have been reviewed using a systematic mapping study. Our study reports on work dating from 2001 to 2021. The initial search retrieved 150 papers from the IEEE Xplore and ACM Digital Libraries, of which 76 were added to the study. A classification scheme is derived based on the primary studies. The study demonstrates that vulnerability detection in web applications is an ongoing field of research and that the number of publications is increasing. Our study helps illuminate research areas that need more consideration.

Vulnerability Detection 2022 - Vulnerability detection has always been an essential part of maintaining information security, and the existing work can significantly improve the performance of vulnerability detection. However, due to the differences in representation forms and deep learning models, various methods still have some limitations. In order to overcome this defect, We propose a vulnerability detection method VDBWGDL, based on weight graphs and deep learning. Firstly, it accurately locates vulnerability-sensitive keywords and generates variant codes that satisfy vulnerability trigger logic and programmer programming style through code variant methods. Then, the control flow graph is sliced for vulnerable code keywords and program critical statements. The code block is converted into a vector containing rich semantic information and input into the weight map through the deep learning model. According to specific rules, different weights are set for each node. Finally, the similarity is obtained through the similarity comparison algorithm, and the suspected vulnerability is output according to different thresholds. VDBWGDL improves the accuracy and F1 value by 3.98\% and 4.85\% compared with four state-of-the-art models. The experimental results prove the effectiveness of VDBWGDL.