CCTV Data Research Articles

Implementation of VPN Using the PPTP Method to Access CCTV Monitoring Data at the Bukit Raya District Office

Bukit Raya District faces challenges in monitoring activities across five scattered sub-districts, as well as employees frequently being absent from the office during working hours. To address this issue, the installation of a CCTV system is crucial for monitoring employee performance and office activities. However, remote access to CCTV data is not yet available. This study aims to implement a VPN using the PPTP (Point to Point Tunneling Protocol) method to allow CCTV monitoring data in sub-district offices to be securely and stably accessed from the district office. PPTP was chosen due to its cross-platform support and ease of configuration and maintenance. The implementation was carried out in two sub-districts, Tangkerang Selatan and Simpang Tiga. The results indicate that the PPTP VPN successfully connected the district office to these sub-districts, enabling secure remote access to CCTV data. Quality of Service (QoS) testing demonstrated network performance improvements in terms of throughput, packet loss, delay, and jitter. Therefore, this system effectively enhances operational efficiency and security in Bukit Raya District.

Real-Time Object Detection and Tracking for Occupational Health and Safety (OHS) in Mining Operations Using Streamed CCTV Data

Abstract Occupational Safety and Health (OHS) is very important in the mining industry which is prone to accident risks. Automatic detection of violations through CCTV streaming with the concept of a perimeter area is needed to prevent accidents. The purpose of this research is to create a violation detection system without re-modeling training that can adapt to different objects and regions. The proposed violation detection model includes four computational steps, namely (i) data reading, (ii) data pre-processing, (iii) object detection, and (iv) object tracking and violation detection. The computer vision technique uses the SIFT (Scale Invariant Feature Transform) algorithm and heuristics implemented with OpenCV. We used five CCTV video recordings of mining operations as examples. The results showed an average precision of 61.14%, a precision of 80.41%, and a recall of 69.38%. The average computation time of the model is 0.3 seconds per frame with heuristics that allow the model to be used in real-time. The model can be widely used in various fields and locations without having to retrain the model to improve OHS in the mining industry and other sectors with similar risks. This study will contribute to the development of a flexible, fast, and real-time automatic detection system.

Traffic Congestion Detection Using YOLOv8 Algorithm With CCTV Data

Community development and growth according to data from the Central Java Statistics Agency regarding the number of vehicles in Central Java Province in 2021 is 20 320 743. The increasing growth of society has caused vehicle density which is a serious problem in urban areas. This study developed a congestion detection system using the YOLOv8 algorithm to analyze traffic density from CCTV footage. Automated detection of traffic congestion is a critical challenge in urban transport management. YOLOv8, a fast and accurate object detection algorithm, is used to identify vehicles and count their number in various areas of the highway. This information is then processed to assess road congestion conditions, with the aim of detecting congestion. The data obtained were tested on two road scenarios and traffic conditions to evaluate the performance of the system. The results showed that the proposed system was able to detect congestion with an accuracy level of 59.2% from several experiments. The use of YOLOv8 enables real-time detection with efficient computing resources, making it a potential solution for large-scale deployment. This research shows that the incorporation of advanced object detection algorithms such as YOLOv8 with CCTV data can provide an effective solution for traffic management in large cities. This system is expected to improve response to congestion, help control traffic, and reduce the negative impact of congestion in urban areas

SaLFIA: Semi-automatic Live Feeds Image Annotation Tool for Vehicle Classification Dataset

Deep learning’s reliance on abundant data with accurate annotations presents a significant drawback, as developing datasets is often time-consuming and costly for specific problems. To address this drawback, we propose a semi-automatic live-feed image annotation tool called saLFIA. Our case study utilized CCTV data from Indonesia’s toll roads as one of the sources for live-feed images. The primary contribution of saLFIA is a labeling tool designed to generate new datasets from public source images, focusing on vehicle classification using YOLOv3 and SSD algorithms. The evaluation results indicated that SSD achieved higher accuracy with fewer initial images, while YOLOv3 reached maximum accuracy with larger initial datasets, resulting in 8 misdetections out of 380 objects. The saLFIA tool simplifies the annotation process, presenting a labeling tool for creating annotated datasets in a single operation. saLFIA is available at URL https://github.com/gilangmantara/salfia.

Prediction of queue length due to lane closure with shockwave analysis on Jagorawi toll road (case study: Km 19+600)

This study discusses the prediction of queue length with shockwave analysis through traffic flow characteristics on the Jagorawi toll road KM 19+600. The purpose of this study is to analyze the relationship between traffic characteristics (current, density and speed) on the Jagorawi KM19+600 toll road and evaluate the value of shock waves due to the narrowing of the road on the Jagorawi KM 19+600 toll road when the lane is closed. The method used to achieve the purpose of this study is to use the linear regression method to determine the relationship between traffic variables through the Greenshield model. The data obtained was obtained from CCTV data recording at KM 19 + 600. The results of the current-velocity-density relationship analysis show that the selected model for the relationship is the Greenshield model with a maximum current of 9574 pcu/hour. From the results of the shock wave simulation for the closure of 1 lane, 2 lanes and 3 lanes with the same flow, namely 7200 pcu/hour. It can be known the length of the queue that occurs, where the length of the queue that occurs depends on the volume of traffic, the length of the closure and the number of lanes that are closed.

Feasibility of using electrodermal activity responses to assess level of crowdedness of pedestrian paths

As urban populations grow, it's imperative to evaluate and enhance the quality of pedestrian paths from the user's perspective. Crowdedness, associated with discomfort and safety, is crucial in determining the overall walking quality and user experience. Previously utilized methods for measuring crowdedness, such as travel diaries and floating population surveys, were limited to collecting perceptual data from sporadic surveys with restricted spatial coverage. Similarly, methods based on CCTV or mobile service data have been used but present issues with blind spots and fail to consider pedestrian perspectives. Against this background, this study explores the feasibility of assessing crowdedness levels by measuring subjects' physiological responses in a laboratory setting based on visual images of real and virtual environments. This study hypothesizes that the amount of people or vehicles passing by affects the electrodermal activity (EDA) of pedestrians, indicating the comfort level of using the environment. Experimental EDA data were measured using a wearable device while the subjects were watching videos showing different pedestrian traffic flows. Representative EDA signal features (e.g., skin conductance responses) were extracted after data pre-processing. Noticeable changes in EDA responses are observed when subjects countered specific environmental variations, such as differing volumes of passing people, on pedestrian paths. The findings suggest that EDA data can be instrumental in differentiating crowdedness levels on pedestrian paths. This study contributes to the body of knowledge by demonstrating the potential of EDA data to characterize the crowdedness experienced by pedestrians. This aids in the development of a novel, quantitative method to gauge pedestrian path crowdedness and to discern contributing factors, such as path width.

Comprehensive examination of regional railway passenger behavior and dwell time components: Insights from video-based observations in Victoria, Australia

In understudied regional railway operations, this study explores passenger boarding and alighting patterns and how station design impacts them, particularly dwell times. Despite extensive metropolitan and suburban train research, regional railways have been overlooked. This study investigates regional rail passenger flow and dwell time to bridge this gap. This article studies dwell times and passenger boarding and alighting at two regional stations in Victoria, Australia, using CCTV data. The objective is to identify insights that might improve regional railway services' efficiency and user experience and advocate for sector-specific solutions. Analysis indicates distinct station boarding and alighting features, highlighting the discovery of the ‘blinded phenomenon’ for train conductors particularly in the afternoon peak (PMP). The results of the study showed that PMP services, which prioritise alighting passengers, had higher dwell times than the morning peak (AMP) services, which emphasise boarding passengers. Obstructed views make it difficult for train conductors to monitor passenger alighting, prolonging dwell times. Better human resource strategies, artificial intelligence for crowd surveillance, and strategic CCTV system deployment to streamline operations and improve passenger experience on regional railways are proposed in the paper, laying the groundwork for future research and operational changes in this vital transportation sector.

Attention Mechanism with Spatial-Temporal Joint Deep Learning Model for the Forecasting of Short-Term Passenger Flow Distribution at the Railway Station

Accurate understanding of passenger flow distribution is crucial for effective station crowd management. However, due to the complexity and randomness of passenger flow and the unclear spatial-temporal correlation between functional areas within the station, predicting the spatiotemporal distribution dynamics of inflow and future short-term distribution trends is challenging. Emerging deep learning models offer valuable insights for accurately predicting passenger flow distribution. Thus, we propose a deep learning architecture, named “ST-Bi-LSTM,” which combines a bidirectional long short-term memory network with a spatial-temporal attention mechanism. Initially, we outline the methodologies of Bi-LSTM, the DeepWalk-based spatial attention mechanism, and the temporal attention mechanism. The spatial attention mechanism is employed to extract station spatial network topology information and enhance the representation of passenger flow characteristics in highly correlated areas during the forecasting process. Simultaneously, the temporal attention Bi-LSTM is utilized for capturing temporal correlations. The architecture comprises four branches dedicated to station real-time video monitoring data, spatial network topology, function area attributes, and train timetables. Subsequently, leveraging in-station CCTV data, passenger travel behavior data, and train timetables, we apply the architecture to the Tianjin West High-Speed Railway Station. We conduct a comparative analysis of the prediction performance and time complexity of the proposed architecture against existing baseline models, demonstrating superior performance and robustness exhibited by the ST-Bi-LSTM model (achieving a reduction in RMSE of over 10%). This study facilitates the transition of station management from passive response to active prediction of station passenger flow dynamics.

베이지안 모델을 활용한 서울시 자치구별 범죄예측지도 연구

The issue of police personnel in Korea has been continuously raised. Efforts are being made to this end in various ways, but the police are still suffering from manpower shortages and heavy work due to policy changes such as “Adjustment of Investigative Rights between the Police and the Prosecution”. In relation to efficient police activities, a notable measure is the use of crime map based on crime big data. This study examined crime maps used at home and abroad, predicted crime rates for each district in Seoul based on Bayesian model, and visualized them to implement crime maps for each district in Seoul for contributing efficient police patrol activities and crime prevention, and moreover to help Primary crime prevention of citizens.
 To achieve the research purpose, this study looked at crime maps used in the United Kingdom, United States, and Korea’s and how they were used. Furthermore, this study tried to predict crimes in Seoul by district using CCTV, drinking establishment, schools and foot traffic data. Finally, based on the predicted analysis values, visualization of crime map for each district in Seoul was implemented.
 Unlike overseas crime maps, crime maps currently used in Korea were somewhat closed way(only limited users can use it or the algorithm has not been made public). Based on literature review, this study attempted to implement crime map that can be disclosed to private citizens based on the Bayesian model which is rising in the field of criminal investigation. Accordingly, database was established by collecting CCTV, drinking establishment, schools and foot traffic data for each district in Seoul. The Bayesian model, which predicted crime occurrence using the collected data, showed an average accuracy of about 87% which is higher accuracy than other crime prediction program abroad. Finally, visualized as a crime map based on the analysis results.
 The active use of crime map is highly valuable. It can help police's manpower shortage and heavy work, moreover private citizens too. Currently, Korea has also introduced and started operating crime map, but due to police’s confidentiality, it is difficult for private citizens to feel the efficiency of crime map. Crime map can be more valuable depending on how it is used, so if the police actively release crime map to the public, will get a great response.

야간조명의 조도가 야간침입절도에 미치는 영향

As cities grow and develop, lighting is no longer limited to merely improving visibility in the dark and providing an environment for nighttime activities. It also plays a crucial role in preventing crimes that occur at night, ensuring safety, and reducing fear. Despite various studies conducted to clarify the effect of lighting on crime, little research has been devoted to investigating the impact of nighttime illumination on criminal activities, particularly in conjunction with open-street CCTVs. This study aims to address the limitations of these previous studies.
 To achieve this goal, the study selected Songpa-gu, Seoul, as the research area, collected illuminance data, open-street CCTV data, and data on nighttime trespassing theft, and analyzed them using GIS and statistical methods. The study's findings revealed that the intensity of nighttime lighting, the number of open-street CCTVs, and the interaction between nighttime illumination intensity and the number of open-street CCTVs all had a statistically significant positive effect on nighttime trespassing theft.
 These results appear somewhat contradictory to previous studies, and two explanations are possible. First, contrary to expectations, nighttime illumination intensity and open-street CCTVs might not assist in reducing nighttime trespassing theft; instead, they could act as factors that increase it. Second, the observed outcome might be due to autonomous districts increasing illumination and installing numerous open-street CCTVs primarily in areas with a high risk of nighttime intrusion and theft.
 This study concludes with policy proposals related to these findings, research limitations, and discussions on potential future research directions.

The Use of CCTV Model Analysis for a Smart Environment in Semarang City

This study examined the concept of a smart city and the importance of a smart environment in city management. CCTV analytics has been employed to monitor and manage urban environments, especially in Semarang City. Smart city and CCTV analytics challenges facing the Semarang City Government were investigated. Interviews, document analysis, and field observation provided input for this case study, combined with specific cases to gather detailed explanations from various sources or stakeholders. The installation of a CCTV analytics system in Semarang aims to create a smart environment. The city strategically installed CCTV cameras with analytical technology and a data center to process and store CCTV data. This system monitored high-risk areas such as flood plains. The Quick Response Team handled problems immediately after the early warning system was activated. However, funding, analytical CCTV supply, and system integration were issues. Hence, the government must address these challenges to manage the environment and reduce risks. This research discovered that analytical CCTV in Semarang City provided real-time data and visualization of environmental conditions, helping the government make the right decisions and policies. Access to CCTV footage and decision-making benefited the community. Air pollution and flooding must be addressed as the integration continued. This study emphasizes urban environmental management strategies and the pros and cons of technology integration. The potential of smart environments and the challenges of smart city implementation require more research and development.

Machine Learning-Based Intelligent Video Surveillance in Smart City Framework

The proposed method of using Machine Learning in Motion Detection and Pedestrian Tracking-assisted Intelligent Video Surveillance Systems (ML-IVSS) can be seen as an application of intelligent fusion techniques. ML-IVSS combines the power of motion detection, pedestrian tracking, and machine learning to create a more accurate and efficient surveillance system for smart cities. By fusing these techniques, ML-IVSS can effectively detect unusual behaviors such as trespassing, interruption, crime, or fall-down, and provide accurate depth data from surveillance footage to protect residents. Intelligent fusion techniques can help improve the accuracy and effectiveness of surveillance systems in smart cities, making them safer and more secure for residents. Combination channel models are used at first, and an object area with prominent features is selected for surveillance. Scaled modification and extraction of features are carried out on the presumed object's region. Identifying the low-level characteristic is the first step in incorporating it into neural architectures for deep feature learning. A smart CCTV data set is used to evaluate the proposed method's performance. According to the numerical analysis, the proposed ML-IVSS model outperforms other traditional approaches in terms of abnormal behaviour detection (98.8%), prediction (97.4%), accuracy (96.9%), F1-score (97.1%), precision (95.6%), and recall (96.2%).

Deep learning based missing object detection and person identification: an application for smart CCTV

Security and protection are the most crucial concerns in today’s quickly developing world. Deep Learning methods and computer vision assist in resolving both problems. One of the computer vision subtasks that allows us to recognise things is object detection. Videos are a source that is taken into account for detection, and image processing technology helps to increase the effectiveness of state-ofthe-art techniques. With all of these technologies, CCTV is recognised as a key element. Using a deep convolutional neural network, we accept CCTV data in real time in this article. The main objective is to make content the centre of things. Using the YOLO technique, we were able to detect the missing item with an improvement of 10% sparsity over the current state-of-the-art algorithm in the context of surveillance systems, where object detection is a crucial step. It can be utilised to take immediate additional action.

Bus passenger injury prevention: Learning from onboard incidents

Objective Bus travel is relatively safe: however there remains a lack of understanding of passenger injury incidents onboard buses. The objective of this study was to understand more about onboard passenger incidents to help inform injury mitigation. Methods The UK national STATS19 data and Transport for London bus incident data (IRIS) were used to determine the size of the problem in Greater London. Other data including onboard incident reports from two bus operators and CCTV footage of 70 incidents were used to understand passenger injury in more depth and identify common themes and challenges. Results The STATS19 and IRIS analysis showed that there was a difference between nationally reported bus incidents compared to locally reported bus incidents. Non-collision incidents are prevalent in the data suggesting there is a large problem to tackle. The CCTV and bus incident data identified braking to be the single largest problem in onboard bus passenger injury incidents. Inconsistent reporting of passenger incidents and injury descriptions make it difficult to identify injury patterns and trends. Areas on the bus appear to contribute to higher injury incidents namely those seats facing and closest to the wheelchair area. Other challenges relating to expected passenger and driver behaviors were noted where blame for the incident and outcome can be attributed to both parties. Conclusions This combined analysis of incident reports and CCTV footage has enabled a better understanding of the events leading to on-board passenger injury incidents. Preventing harsh braking would appear to be the most effective way of reducing passenger injuries. Additionally improved data collection would assist both transport authorities and bus operators to identify and monitor the effect of bus safety improvements.

Landscape Design Toward Covid-19 in the Old Town Semarang, Indonesia

A significant increase in Covid-19 cases worldwide disrupted many sectors. The virus transmission through the air requires that people limit outdoor activities and apply social distancing. However, outdoor activities cannot be avoided because it is how people meet their daily needs. The urban landscape design arrangement is crucial, especially in public spaces. Old Town Semarang is a location with the potential for high outdoor activities, especially tourism. CCTV data shows that tourists could reach up to 500 people per day, and the average visitor does a comply with health protocols. This is worsened by the unadopted landscape conditions in this area to the current pandemic conditions. Therefore, a landscape design approach is needed to adapt to visitors' needs and minimize virus transmission. This research examined the effect of landscape design on Covid-19 in the Old Town area. Data were collected on the existing landscape conditions, cultural heritage, and the distribution of positive Covid-19 cases and analyzed using spatial analysis, design simulation, and quantitative descriptive. The results were presented as simulation or landscape design recommendations according to cultural heritage elements and health protocols. The recommendations are a division of activity zones with a visitor quota system, and optimization of health facilities, signage for physical distance, as well as a clean and healthy lifestyle for visitors.

Changes in public attitudes toward CCTV installations in residential areas between 2008 and 2019

Despite a steady decline in crime, CCTV usage in public areas has increased rapidly, spreading to residential areas and even non-metropolitan regions in Japan. However, few studies have focused on public attitudes toward CCTV in residential areas, and little is known about how these attitudes have changed over the years. This study examined how public attitudes toward CCTV in residential areas (i.e., neighborhood streets, parks, and shopping areas) have changed between 2008 and 2019. The study used data from two online cross-sectional questionnaires and multi-group structural equation modeling. Our results suggest that respondents were more supportive of CCTV in all sites and that the difference in attitudes between non-metropolitan and metropolitan residents decreased significantly in 2019. In particular, the percentage of respondents who focused on negative impacts (e.g., invasion of privacy) decreased, percentage who agreed with the potential of CCTV to reduce crime decreased, and percentage who believed in the ability of CCTV to reduce fear increased. Although demands for the use of CCTV data for non-security purposes would increase to create data-driven smart cities, the expansion of CCTV installation needs to be carefully considered according to the different socio-economic contexts of each region, including crime risks.

Evaluation of CCTV Data For Estimating Rainfall Condition

Rainfall characteristics of Indonesia's tropical climate have high variability according to space and time, so to determine the rainfall pattern of a location, an in situ rainfall measuring instrument (AWS = automatic weather station) is needed with high density. The existence of AWS also requires relatively high maintenance costs and a standard placement location (according to the rules of WMO = World Meteorological Organization) which is relatively broad and is not obstructed by other objects that can make the result of rainfall data is not representative. With the concept of computer vision, research will be carried out to estimate the rainfall condition from the CCTV cameras. The CCTV camera data which have qualitative characteristic into rainfall data which have quantitative characteristics. This research is also motivated by the large number of CCTVs that are placed in a lot of locations by local governments along with the Smart City program in districts and cities throughout Indonesia. The preliminary research was conducted in Center for Atmospheric Science and Technology office in Bandung. Rainfall data from AWS was used to validate CCTV data which placed in same location. The process of converting CCTV data into rainfall data goes through 6 stages. The first is reading the image mapping data and AWS (in rainfall accumulation data form). Second, read the image data in grayscale. Third, extract the features. Fourth, split the reference and sample data. Fifth, conducts the K-NN Mapping Reference Image and rainfall accumulation data. Sixth is to praise K-NN Testing. The accuracy is calculate with comparing the estimated number of CCTV cameras that are correct with the total sample size. The evaluation result states that the highest accuracy is obtained with K = 1. When K=1, the accuracy percentage reaching 94.8%. Accuracy decreases with increasing value of K and drastically decreases with K> 2. In the 1-10 days reference data, the highest accuracy is obtained by the number of reference data for 10 days, which is around 97%, stable until the value of K = 8. While the lowest accuracy is obtained when the reference data is 1 day with an accuracy value of about 43%. Based on the results of this study, it can be concluded that rain data from CCTV can be used to estimate the rainfall data. The best result happened when K-value is equal to 1.

Two-Level Blockchain System for Digital Crime Evidence Management.

Digital evidence, such as evidence from CCTV and event data recorders, is highly valuable in criminal investigations, and is used as definitive evidence in trials. However, there are risks when digital evidence obtained during the investigation of a case is managed through a physical hard disk drive until it is submitted to the court. Previous studies have focused on the integrated management of digital evidence in a centralized system, but if a centralized system server is attacked, major operations and investigation information may be leaked. Therefore, there is a need to reliably manage digital evidence and investigation information using blockchain technology in a distributed system environment. However, when large amounts of data—such as evidence videos—are stored in a blockchain, the data that must be processed only within one block before being created increase, causing performance degradation. Therefore, we propose a two-level blockchain system that separates digital evidence into hot and cold blockchains. In the criminal investigation process, information that frequently changes is stored in the hot blockchain, and unchanging data such as videos are stored in the cold blockchain. To evaluate the system, we measured the storage and inquiry processing performance of digital crime evidence videos according to the different capacities in the two-level blockchain system.

Deep Learning-based Vehicle Anomaly Detection using Road CCTV Data

Deep Learning-based Vehicle Anomaly Detection using Road CCTV Data

Using deterioration modelling to simulate sewer rehabilitation strategy with low data availability.

Most cities face the problem of an aging infrastructure in need of extensive and ongoing repair, renovation or replacement. Since the 1980s, CCTV has been the industry standard for sewer system inspection and the main source of information for structural performance evaluation. Due to low inspection rates and the lack of information about sewer condition, deterioration models have been developed to simulate the condition of non-inspected sewers and assess the influence of several rehabilitation scenarios. This paper presents an innovative modelling tool for long-term sewer rehabilitation planning based on the integration of a deterioration and a rehabilitation model. The tool is demonstrated in full scale using CCTV and sewer data of the city of Sofia, in Bulgaria. Results provide tangible proofs of investment needs for sewer rehabilitation and support the utility in the negotiation of budgets with the municipality. Since age is one key variable for deterioration modelling, a new method is proposed to estimate missing construction years in the utility database with a prediction error of less than 7 years.