Search Results (11)

Property crime is a pressing issue in maintaining social order and urban sustainability, particularly in regions marked by pronounced socioeconomic disparity. While the link between socioeconomic stress and crime is well established, regional variations in Thailand have not been fully examined. Therefore, the purpose of this research was to examine spatial patterns of property crime and identify the potential associations between property crime and socioeconomic environment across Thailand. Using nationally compiled property-crime data from official sources across all provinces of Thailand, we employed geographic information system (GIS) tools to conduct a spatial cluster analysis at the sub-national level across 76 provinces. Both global and local statistical techniques were applied to identify spatial associations between property-crime rates and neighborhood-level socioeconomic conditions. The results revealed that property-crime clusters are primarily concentrated in the south, while low-crime areas dominate parts of the north and northeast regions. To analyze the spatial dynamics of property crime, we used geospatial statistical models to investigate the influence of socioeconomic variables across provinces. We found that property-crime rates were significantly associated with monthly income, areas experiencing high levels of household debt, migrant populations, working-age populations, an uneducated labor force, and population density. Identifying associated factors and mapping geographic regions with significant spatial clusters is an effective approach for determining where issues concentrate and for deepening understanding of the underlying patterns and drivers of property crime. This study offers actionable insights for enhancing safety, resilience, and urban sustainability in Thailand’s diverse regional contexts by highlighting geographies of vulnerability. Full article

The aim of the present study is to identify obesogenic environment profiles to find the obesogenic environment pattern for Belo Horizonte City. The current research followed the ecological approach and was substantiated by data from food shops, public sports venues, crime rates (homicides and robberies) and the rate of accidents with pedestrians. Descriptive analyses and principal component analysis (PCA) were conducted in Stata software, version 14.0. Georeferencing and map plotting were carried out in Qgis software, version 2.10. All neighborhoods in Belo Horizonte City (n = 486) were included in the study. The obesogenic pattern comprised the highest mean number of shops selling ultra-processed food, crime rates, and accidents with pedestrians. The generated latent variable was divided into tertiles, and the second and third tertiles represented the most obesogenic environments. Neighborhoods accounting for the highest obesogenic profile also recorded the largest number of shops selling all food types. Furthermore, neighborhoods in the third tertile recorded the highest mean income (BRL 2352.00) (p = 0.001) and the lowest Health Vulnerability Index (HVI = 54.2; p = 0.001). These findings point towards the need for developing actions, policies and programs to improve these environments, such as tax incentives to open healthy food retailers and public sports venues to promote healthier lifestyles and to prevent diseases in the middle and long term. Full article

The rapid increase in the number of vehicles has led to increasing traffic congestion, traffic accidents, and motor vehicle crime rates. The management of various parking lots has also become increasingly challenging. Vehicle-type recognition technology can reduce the workload of humans in vehicle management operations. Therefore, the application of image technology for vehicle-type recognition is of great significance for integrated traffic management. In this paper, an improved faster region with convolutional neural network features (Faster R-CNN) model was proposed for vehicle-type recognition. Firstly, the output features of different convolution layers were combined to improve the recognition accuracy. Then, the average precision (AP) of the recognition model was improved through the contextual features of the original image and the object bounding box optimization strategy. Finally, the comparison experiment used the vehicle image dataset of three vehicle types, including cars, sports utility vehicles (SUVs), and vans. The experimental results show that the improved recognition model can effectively identify vehicle types in the images. The AP of the three vehicle types is 83.2%, 79.2%, and 78.4%, respectively, and the mean average precision (mAP) is 1.7% higher than that of the traditional Faster R-CNN model. Full article

With the rapid development in technology in recent years, the use of cameras and the production of video and image data have similarly increased. Therefore, there is a great need to develop and improve video surveillance techniques to their maximum extent, particularly in terms of their speed, performance, and resource utilization. It is challenging to accurately detect anomalies and increase the performance by minimizing false positives, especially in crowded and dynamic areas. Therefore, this study proposes a hybrid video anomaly detection model combining multiple machine learning algorithms with pixel-based video anomaly detection (PBVAD) and frame-based video anomaly detection (FBVAD) models. In the PBVAD model, the motion influence map (MIM) algorithm based on spatio–temporal (ST) factors is used, while in the FBVAD model, the k-nearest neighbors (kNN) and support vector machine (SVM) machine learning algorithms are used in a hybrid manner. An important result of our study is the high-performance anomaly detection achieved using the proposed hybrid algorithms on the UCF-Crime data set, which contains 128 h of original real-world video data and has not been extensively studied before. The AUC performance metrics obtained using our FBVAD-kNN algorithm in experiments were averaged to 98.0%. Meanwhile, the success rates obtained using our PBVAD-MIM algorithm in the experiments were averaged to 80.7%. Our study contributes significantly to the prevention of possible harm by detecting anomalies in video data in a near real-time manner. Full article

Geographical mapping has revolutionized data analysis with the help of analytical tools in the fields of social and economic studies, whereby representing statistical research variables of interest as geographic characteristics presents visual insights. This study employed the QGIS mapping tool to create predicted choropleth maps of Visegrád Group countries based on crime rate. The forecast of the crime rate was generated by time series analysis using the ARIMA (autoregressive integrated moving averages) model in SPSS. The literature suggests that many variables influence crime rates, including unemployment. There is always a need for the integration of widespread data insights into unified analyses and/or platforms. For that reason, we have taken the unemployment rate as a predictor series to predict the future rates of crime in a comparative setting. This study can be extended to several other predictors, broadening the scope of the findings. Predictive data-based choropleth maps contribute to informed decision making and proactive resource allocation in public safety and security administration, including police patrol operations. This study addresses how effectively we can utilize raw crime rate statistics in time series forecasting. Moreover, a visual assessment of safety and security situations using ARIMA models in SPSS based on predictor time-series data was performed, resulting in predictive crime mapping. Full article

Forest destruction has been found to be the cause of natural disasters in the form of floods, landslides in the rainy season, droughts in the dry season, climate change, and global warming. The high rate of forest destruction is caused by various factors, including weak law enforcement efforts against forestry crimes, such as illegal logging events. However, in Indonesia, illegal logging is only discovered when the perpetrator has distributed the wood products. The lack of monitoring of the overall condition of the forest has an impact on the current high level of forest destruction. Through this research, the problems related to environmental damage due to illegal logging will be described through remote sensing technology, which is currently mainly developed on the basis of artificial intelligence and machine learning, namely Google Earth Engine (GEE). Monitoring of illegal logging events will be analysed using Sentinel 1 and 2 data. Obtaining satellite imagery with relatively small cloud cover for tropical regions, such as Indonesia, is remarkably difficult. This difficulty is due to the presence of a radar sensor on Sentinel 1 images that can penetrate clouds, allowing for observation of the forest condition even in the presence of clouds. Using the random forest classification algorithm of the GEE platform, data on forest conditions in 2021 were obtained, covering an area of 2,843,938.87 ha or 63% of the total area of Sulawesi Selatan Province. An analysis using a map of the function of forest areas revealed that of the current forest area, 38.46% was non-forest estates and 61.54% was forest areas. The continued identification of illegal logging events also found 1971 spots of forest change events in the vulnerable time of the first period (January–April) with the second period (April–August), and 1680 spots of forest change events in the vulnerable period of the second period (April–August) with the third period (September–December), revealing a total incident area of 7599.28 ha. Full article

Existing research suggests that COVID-19 lockdowns tend to contribute to a decrease in overall urban crime rates. Most studies have compared pre-lockdown and post-lockdown periods to lockdown periods in Western cities. Few have touched on the fine variations during lockdowns. Equally rare are intracity studies conducted in China. This study tested the relationship between violent crime and COVID-19 lockdown policies in ZG City in southern China. The distance from the isolation location to the nearest violent crime site, called “the nearest crime distance”, is a key variable in this study. Kernel density mapping and the Wilcoxon signed-rank test are used to compare the pre-lockdown and post-lockdown periods to the lockdown period. Panel logistic regression is used to test the fine variations among different stages during the lockdown. The result found an overall decline in violent crime during the lockdown and a bounce-back post-lockdown. Violent crime moved away from the isolation location during the lockdown. This outward spread continued for the first two months after the lifting of the lockdown, suggesting a lasting effect of the lockdown policy. During the lockdown, weekly changes in COVID-19 risk ratings at the district level in ZG City also affected changes in the nearest crime distance. In particular, an increase in the risk rating increased that distance, and a drop in the risk rating decreased that distance. These findings add new results to the literature and could have policy implications for joint crime and pandemic prevention and control. Full article

Lanzhou’s rapid development has raised new security challenges, and improving public safety in areas under the jurisdiction of police stations is an effective way to address the problem of public security in urban areas. Unfortunately, the existing studies do not consider how factors such as future land changes, building functions, and characteristics of criminal behavior influence the choice of areas for police stations and the optimization of police stations with respect to traffic congestion. To solve these problems, we apply multiple methods and multi-source geospatial data to optimize police station locations. The proposed method incorporates a big data perspective, which provides new ideas and technical approaches to site selection models. First, we use the central city of Lanzhou as the study area and erase the exclusion areas from the initial layer to identify the undeveloped areas. Second, historical crime data, point of interest, and other data are combined to assess the potential crime risk. We then use the analytic hierarchy process to comprehensively assess undeveloped areas based on potential crime hotspots and on socioeconomic drivers and orography. In addition, according to China’s Road Traffic Safety Law and the current traffic congestion in the city, a minimum speed is determined, so that the target area can be reached in time even in congested traffic. Finally, we draw the spatial coverage map of police stations based on the location-allocation model and network analysis and optimize the map by considering the coverage rate of high-risk areas and building construction, in addition to maintenance and other objectives. The result shows that crime concentrates mainly in densely populated areas, indicating that people and wealth are the main drivers of crime. The differences in the spatial distribution of crime hotspots and residential areas at different spatial scales mean that the ratio of public security police force to household police force allocated to different police stations is spatially nonuniform. The method proposed herein reduces the overlap of police station service areas by 22.8% and increases the area coverage (12.01%) and demand point coverage (7.25%). The area coverage means an area potentially accessible within five minutes, and point coverage implies an effective drive. Within reasonable optimization, this allows us to eventually remove 13 existing police stations and add 24 candidate police stations. Full article

The recent development in the area of IoT technologies is likely to be implemented extensively in the next decade. There is a great increase in the crime rate, and the handling officers are responsible for dealing with a broad range of cyber and Internet issues during investigation. IoT technologies are helpful in the identification of suspects, and few technologies are available that use IoT and deep learning together for face sketch synthesis. Convolutional neural networks (CNNs) and other constructs of deep learning have become major tools in recent approaches. A new-found architecture of the neural network is anticipated in this work. It is called Spiral-Net, which is a modified version of U-Net fto perform face sketch synthesis (the phase is known as the compiler network C here). Spiral-Net performs in combination with a pre-trained Vgg-19 network called the feature extractor F. It first identifies the top n matches from viewed sketches to a given photo. F is again used to formulate a feature map based on the cosine distance of a candidate sketch formed by C from the top n matches. A customized CNN configuration (called the discriminator D) then computes loss functions based on differences between the candidate sketch and the feature. Values of these loss functions alternately update C and F. The ensemble of these nets is trained and tested on selected datasets, including CUFS, CUFSF, and a part of the IIT photo–sketch dataset. Results of this modified U-Net are acquired by the legacy NLDA (1998) scheme of face recognition and its newer version, OpenBR (2013), which demonstrate an improvement of 5% compared with the current state of the art in its relevant domain. Full article

Given the scarcity of annotated datasets, learning the context-dependency of anomalous events as well as mitigating false alarms represent challenges in the task of anomalous activity detection. We propose a framework, Deep-network with Multiple Ranking Measures (DMRMs), which addresses context-dependency using a joint learning technique for motion and appearance features. In DMRMs, the spatial-time-dependent features are extracted from a video using a 3D residual network (ResNet), and deep motion features are extracted by integrating the motion flow maps’ information with the 3D ResNet. Afterward, the extracted features are fused for joint learning. This data fusion is then passed through a deep neural network for deep multiple instance learning (DMIL) to learn the context-dependency in a weakly-supervised manner using the proposed multiple ranking measures (MRMs). These MRMs consider multiple measures of false alarms, and the network is trained with both normal and anomalous events, thus lowering the false alarm rate. Meanwhile, in the inference phase, the network predicts each frame’s abnormality score along with the localization of moving objects using motion flow maps. A higher abnormality score indicates the presence of an anomalous event. Experimental results on two recent and challenging datasets demonstrate that our proposed framework improves the area under the curve (AUC) score by $6.5\%$ compared to the state-of-the-art method on the UCF-Crime dataset and shows AUC of $68.5\%$ on the ShanghaiTech dataset. Full article

With the development of Geographic Information Systems (GIS), crime mapping has become an effective approach for investigating the spatial pattern of crime in a defined area. Understanding the relationship between crime and its surrounding environment reveals possible strategies for reducing crime in a neighborhood. The relationship between vegetation density and crime has long been under debate. The convenience of a road network is another important factor that can influence a criminal’s selection of locations. This research is conducted to investigate the correlations between tree coverage and property crime, and road density and property crime in the City of Vancouver. High spatial resolution airborne LiDAR data and road network data collected in 2013 were used to extract tree covered areas for cross-sectional analysis. The independent variables were inserted into Ordinary Least-Squares (OLS) regression, Spatial Lag regression, and Geographically Weighted Regression (GWR) models to examine their relationships to property crime rates. The results of the cross-sectional analysis provide statistical evidence that there are negative correlations between property crime rates and both tree coverage and road density, with the stronger correlations occurring around Downtown Vancouver. Full article