Road Collisions Research Articles

Characteristics and comparison between e-scooters and bicycle-related trauma: a multicentre cross-sectional analysis of data from a road collision registry

BackgroundUrban mobility has drastically evolved over the last decade and micromobility rapidly became an expanding segment of contemporary daily transportation routines. E-scooter riders and bicyclists may share similar trauma characteristics, but this has been little explored. The objective was to describe and compare the characteristics of e-scooter and bicycle-related trauma.MethodsWe conducted a cross-sectional analysis of data from the Rhône road collision registry (January 1, 2019 to December 31, 2019). We included all e-scooter or bicycle riders injured in traffic collisions during the study period; there were no exclusion criterion.ResultsA total of 2,779 patients were included; 825 (29.7%) were e-scooter riders and 1,954 (70.3%) were bicyclists. E-scooter riders were younger (median [IQR]: 24 [20–32] vs 29 [20–45] years, p < 0.001) and less frequently male (64.2% vs 73.4%, p < 0.001). Most e-scooter and bicycle road collisions were consequent to a fall or loss of vehicle control (74.2% vs 67.7%, p < 0.001). E-scooter riders were less frequently wearing a helmet at the time of the road collision (6.1% vs 30.7%, p < 0.001) and had more frequently head (24.2% vs 19.9%, p = 0.01) and face (30.6 vs 20.5%, p < 0.001) injuries compared to bicyclists. The median injury severity score was 2 [1–4] in both groups with no significant difference (p = 0.77).ConclusionsE-scooter and bicycle-related trauma patients were mainly young males with minor injuries and most of them sustained a road collision with no third-party. However, they suffered from different injury patterns; e-scooter riders suffered more frequently face and head injuries than bicycle riders, which may be at least partly the consequence of less frequent helmet use among e-scooter riders compared to bicyclists. Hence the two groups of users should not be considered as a single trauma entity. This issue should be promptly addressed to bring down the incidence of preventable injuries and avoid healthcare costs.

A decision support tool for evaluating the wildlife corridor design and conservation performance using analytic network process (ANP)

A recent issue highlights the need for a wildlife corridor to mitigate habitat fragmentation and biodiversity degradation caused by intensive urbanization in the country, such as the development of infrastructure, roads, and highways. Indeed, the issues of frequent road and vehicle-animals collisions are ever-increasing in urban corridors. As a result, flora and fauna populations can become isolated, leading to an unbalanced ecosystem. This study developed the Wildlife Corridor Design and Conservation (WCDC) index, assessment model. The study has investigated the wildlife corridor design and conservation features through a systematic literature review, clustered them into three criteria (biophysical, structural, and environmental designs) and twenty-one sub-criteria. The study applied the Analytic Network Process method to measure the weights of features; habitat preservation (WC1.1 = 0.280), species varieties control (WC1.2.=0187), and conduit for movement (WC1.3. = 0.100) play critical roles in wildlife corridor design and conservation. The Weighted Sum Method verified the WCDC model through a case study (Red Earth overpass, Canada). The model assigns certification labels (Gold, Silver, Bronze, or not-certified); accordingly, it ranked the evaluated site as Gold, which means the Red Earth overpass performs intimately in biophysical, structural, and environmental preservation of mammal habitat. The study conducted regression analysis for model validation, determining a significant association between the base and case-study models.

Accident analysis and road safety audit: a case study on NH-76

Road accidents in India are on the rise as the country’s vehicle population grows. As a result, a considerable number of lives and assets have been lost. The formal process for determining causal factors responsible for road accidents in new or improvement of existing road projects without any major modification in the design is known as a road safety audit. This study examines the traffic safety situation on NH-76 from Udaipur to Chittorgarh in the state of Rajasthan, India, intending to identify causal factors responsible for road accidents so that the real damage resulting from road collisions can be reduced to some extent in the future. For this study, a 93.5-kilometer span was chosen and accident data was obtained from the enforcement department via recorded accident FIRs from 2017 to 2019. The accident data was studied to determine the reasons, taking into account yearly, hourly, time-wise, day-wise variation, accident nature-wise, and other factors. Black spots were also identified in the study area and recommendations based on analysis were made to prevent traffic accidents and make the route safer for users. The purpose of this study is to uncover flaws, improve design characteristics, and improve road credibility.

A Fast Markov Decision Process-Based Algorithm for Collision Avoidance in Urban Air Mobility

Multiple aircraft collision avoidance is a challenging problem due to a stochastic environment and uncertainty in the intent of other aircraft. Traditionally a layered approach to collision avoidance has been employed using a centralized air traffic control system, established rules of the road, separation assurance, and last minute pairwise collision avoidance. With the advent of Urban Air Mobility (air taxis), the expected increase in traffic density in urban environments, short time scales, and small distances between aircraft favor decentralized decision making on-board the aircraft. In this paper, we present a Markov Decision Process (MDP) based method, named FastMDP, which can solve a certain subclass of MDPs quickly, and demonstrate using the algorithm online to safely maintain separation and avoid collisions with multiple aircraft (1-on-n) while remaining computationally efficient. We compare the FastMDP algorithm’s performance against two online collision avoidance algorithms that have been shown to be both efficient and scale to large numbers of aircraft: Optimal Reciprocal Collision Avoidance (ORCA) and Monte Carlo Tree Search (MCTS). Our simulation results show that under the assumption that aircraft do not have perfect knowledge of other aircraft intent FastMDP outperforms ORCA and MCTS in collision avoidance behavior in terms of loss of separation and near mid-air collisions while being more computationally efficient. We further show that in our simulation FastMDP behaves nearly as well as MCTS with perfect knowledge of other aircraft intent. Our results show that FastMDP is a promising algorithm for collision avoidance that is also computationally efficient.

Developing and implementing 20-mph speed limits in Edinburgh and Belfast: mixed-methods study

Background Transport initiatives such as 20-mph (≈30-km/h) speed limits are anticipated to result in fewer road casualties and improve perceptions of safety, leading to increases in active travel. Lower speeds may also lead to more pleasant environments in which to live, work and play. Objectives The main objective was to evaluate and understand the processes and effects of developing and implementing 20-mph speed limits in Edinburgh and Belfast. The focus was on health-related outcomes (casualties and active travel) that may lead to public health improvements. An additional objective was to investigate the political and policy factors (conditions) that led to the decision to introduce the new speed limits. Design This was a mixed-methods study that comprised an outcome, process, policy and economic evaluation of two natural experiments. Setting The study was set in Edinburgh, Scotland, and Belfast, Northern Ireland, from 2000 to 2018. Participants The whole population of each city were participants, as well as stakeholders involved in implementation and decision-making processes. Intervention The intervention was the implementation of 20-mph legislation, signage, enforcement, and education and awareness-raising in Edinburgh (citywide) and Belfast (city centre). Main outcome measures The main outcomes measured were speed; number, type and severity of road collisions; perceptions; and liveability. Data sources The following data sources were used – routinely and locally collected quantitative data for speed, volume of traffic, casualties and collisions, and costs; documents and print media; surveys; interviews and focus groups; and Google Street View (Google Inc., Mountain View, CA, USA). Results Collisions and casualties – the overall percentage reduction in casualty rates was 39% (the overall percentage reduction in collision rates was 40%) in Edinburgh. The percentage reduction for each level of severity was 23% for fatal casualties, 33% for serious casualties and 37% for minor casualties. In Belfast there was a 2% reduction in casualties, reflecting differences in the size, reach and implementation of the two schemes. Perceptions – in Edinburgh there was an increase in two factors (support for 20 mph and rule-following after implementation) supported by the qualitative data. Liveability – for both cities, there was a small statistical increase in liveability. Speed – mean and median speeds reduced by 1.34 mph and 0.47 mph, respectively, at 12 months in Edinburgh, with no statistically significant changes in Belfast. History, political context, local policy goals, local priorities and leadership influenced decision-making and implementation in the two cities. Limitations There was no analysis of active travel outcomes because the available data were not suitable. Conclusions The pre-implementation period is important. It helps frame public and political attitudes. The scale of implementation and additional activities in the two cities had a bearing on the impacts. The citywide approach adopted by Edinburgh was effective in reducing speeds and positively affected a range of public health outcomes. The city-centre approach in Belfast (where speeds were already low) was less effective. However, the main outcome of these schemes was a reduction in road casualties at all levels of severity. Future work Future work should develop a statistical approach to public health interventions that incorporates variables from multiple outcomes. In this study, each outcome was analysed independently of each other. Furthermore, population measures of active travel that can be administered simply, inexpensively and at scale should be developed. Study registration This study is registered as ISRCTN10200526. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme and will be published in full in Public Health Research; Vol. 10, No. 9. See the NIHR Journals Library website for further project information.

Dual-V-Sense-Net (DVN): Multisensor Recommendation Engine for Distraction Analysis and Chaotic Driving Conditions

The Advanced Driver Assistance System (ADAS) features provide better safe driving dynamics reducing road accidents and collisions. Typically, these include lane assist, automatic night vision, adaptive cruise control, driver risk analysis primarily seen in high-end vehicles. Furthermore, such a proactive mechanism alerts the drivers in unsafe/ dangerous situations and enable drivers to respond with immediate actions. However, such ADAS-based solutions do not work very well in heterogeneous, less-disciplined, and chaotic road traffic environments, as seen in Asian countries such as India, Bangladesh, Burma, Nepal, etc. In this regard, this paper aims to propose an advanced integrated deep learning framework <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Dual-V-sense-Net (DVN)</i> to provide recommendations for safe driving. A new dataset - Behavioral Understanding of Driver’s-Distraction Dataset for India (BUDDI), is proposed for the behavioral understanding of driver’s distraction. This proposed recommendation framework is designed in a two-step manner. The first step involves an in-vehicle driver distraction analysis with real-time frontal and cross-view naturalistic driver data for driver distraction analysis. Second step is to perform an exhaustive experimental evaluation is carried out for the tasks such as road traffic scene perception, road navigable area prediction with vehicular temporal dynamics, lane assist, safe drivable area navigation and driver distraction alert. A decision block modeling helps in safe/unsafe driving predictions considering the feature importance. We present comprehensive experimental results for road segmentation, vehicular road lane categorization, obstacle recognition, and driver behavioral pattern prediction using the BUDDI and Indian Driving Dataset (IDD) datasets.

Collision Warning System Using Naïve Bayes Classifier

Motor vehicle crashes can lead to traumatic experiences. High-impact collisions usually cause severe injuries or fatalities. A collision warning system that analyzes driving behaviors and warns drivers of impending crashes can prevent road collisions and save lives, so increasing traffic safety. An application of the Naïve Bayes classifier model to determine the potential for rear-end collisions between highway vehicles is presented. The Naïve Bayes classifier is a supervised machine-learning model based on Bayes’s theorem. Two vehicles are utilized, with one vehicle following the other. The parameters studied are speed, distance, acceleration, and deceleration. Training examples involving over 100 potential collision scenarios have been evaluated. Simulation results show that the model successfully responds to and correctly predicts potential collisions.

Effects of built environment in subway stations on pedestrian injuries

ObjectiveIn South Korea, major events that cause many subway injuries due to trains or other facilities and force majeure such as terrorism or natural catastrophes have markedly decreased. However, there are still many other types of cases involving injury inside subway stations. To reduce this damage, it is essential to identify and evaluate the factors that have the potential to lead to event occurrence. In other transportation domains, such as in the area of road traffic, crash or collision frequencies have been estimated using a safety performance function (SPF) with explanatory variables that represent risk factors. We introduce an SPF modeling approach for the injury events inside subway stations to interpret the risks to the users moving between exit and platform for boarding and alighting and put forward implications for the operation and design of stations. MethodsWe conducted SPF modeling based on a negative binomial regression analysis, using 4039 injury event datasets for eight lines and 235 stations in the Seoul Metropolitan Subway network. The number of passengers is an exposure variable, and other explanatory variables consist of spatial attributes inside stations. ResultsThe number of users of each station has a positive and nonlinear correlation with event frequency. Several spatial attributes have positive correlations with events, especially the vertical movement of passengers and the total length of escalators, and the number of wheelchair lifts and floors in each station. Results estimate that 54% more event occurrence in transfer stations than in non-transfer stations. The total length of platforms and the area of shops in each station, on the other hand, are negatively correlated with events. ConclusionsThe results provide quantitative guidance for dealing with risk factors in each corresponding phase of the design, construction, operation, and maintenance of subway stations.

Slow and Steady Wins the Race: A Comparative Analysis of Standing Electric Scooters’ European Regulations Integrated with the Aspect of Forensic Traumatology

Fuel-driven cars are widely considered unsustainable and contrary to the new paradigm of smart growth planning. The need to reform transport behavior, policies, and infrastructure is among the priorities in urban policies around the world. Electric vehicles are an emerging technology that could advance sustainability programs. In the past year, there has been a rapid increase in the diffusion of electric scooters in several European cities, but various states have been unprepared for the rapid spread of green micro-mobility from a regulatory point of view. In addition, in parallel with the spread, there have been numerous road collisions involving standing electric scooters. The aim of this study was to obtain a detailed view of this phenomenon. We focused on the current legislation on electric micro-mobility at the European level to study and summarize the different attitudes adopted by various states whose regulations are present on the web. (It was not possible to evaluate the regulations of all European countries because they are not all available on online platforms.) The elements assessed in the various regulation were age limits, speed limits, compulsory use of helmets, administrative penalties, and the obligation to insure the new e-vehicle (standing scooter). In this study, we analyze the state of the art in electric micro-mobility, highlight the current situation’s limits, and propose new strategies to adequately integrate this new smart vehicle into the urban transport network.

Searching for Predictors of Safe Driving Behaviour as an Important Activity in Achieving an Integrated Traffic System. Research on the Behavioural Criterion of the Psychometric Test Validity for Drivers

Safety in road traffic as a system in which there are dynamic interactions between its different users depends on how integrated it is. In this system, drivers are a favoured group of users. Unfortunately they are also the most dangerous group, as research shows. Studying the causes of dangerous driving behaviour is still important.&#x0D; The paper aims to present the psychometric methodology to define the diagnostic and prognostic validity of some psychometric tests used by transport psychologists. Our statistical analysis included the four experimental groups of professional drivers with motor vehicle accident and one control group of drivers whose road performance had no motor vehicle accident recordings.&#x0D; The novelty of the study presented here is in linking the psychometric tests outcomes of professional road drivers (city bus drivers, school bus drivers, taxi drivers, ambulance drivers, fire trucks drivers, police drivers, military vehicle&#x0D; drivers, special vehicle drivers etc.) and their behavioral safety performance on public roads (i.e. causing road accidents, road collisions, participating in accidents but not causing it, and participating in road collision but not causing it) – which aims to indicate the diagnostic and prognostic validity of these tests for drivers.&#x0D; Discriminatory analysis based on Fisher ĝ-function was used to find discrepancies between the specific test outcomes of the road drivers in the control group and the four risky behaviour groups. The identified discrepancies are interpreted in terms of equivalence between the diagnostic and prognostic validity of the taken into consideration psychometric tools used by psychologists for diagnosing road drivers.

A Novel Fast Responding Driver Assistance Technique with Efficient Lane Detection and Collision Avoidance Using Dynamic Feature Extraction in Any Environment

Road accidents caused by a driver's irresponsibility while driving is becoming increasingly common. Furthermore, if lanes are lacking on the road, even well-trained drivers may find it difficult to keep the lane while driving in low-light conditions. As a result, it is critical to design trustworthy, precise, and efficient mechanisms in the vehicle system that aid the driver in the event of a road collision. Almost every country in the world is attempting to conquer this greatest difficulty. The topic study article focuses on the crossroads and presents a more authentic and efficient diver aid system strategy in terms of lane departure alarm even when lanes are missing by taking prior lane patterns into account. In addition, research is being conducted in order to provide a speedy collision warning with nearly no false alarms. This study develops a fast-response spatial domain approach for detecting lanes on highways, and if lane markers are missing, virtual lanes are constructed using a novel suggested algorithm. In addition, for vehicle collision avoidance, the system estimates the distance, velocity, and direction from the frontal vehicle to itself. The suggested approach is evaluated on real-time videos in all environmental circumstances such as poor or bright sunlight, rain, and twilight, as well as on different road geometries such as straight and curving at various vehicle speeds. In every situation, the system has achieved more than 98.7 percent accuracy. Even if there are no markings on road, the system provides more accurate and reliable experimental results. Finally, the results are compared to several existing algorithms based on accuracy, precision, recall, and F1 score, along with processing time.

Multiagent modeling of pedestrian-vehicle conflicts using Adversarial Inverse Reinforcement Learning

There is a need for a better understanding of the collision avoidance behavior of road users in near misses. Recently, several models of road user behavior in near misses have been proposed. However, despite the multiagent nature of road user interactions, most of these studies modeled their behavior using a single-agent approach. However, this approach is unrealistic and can limit the models’ accuracy. Therefore, this study proposes the Markov-Game (MG) framework for modeling pedestrian-vehicle interactions and their collision avoidance mechanisms. Pedestrian-vehicle conflicts in a mixed traffic environment in China are extracted using computer-vision algorithms. Pedestrian and vehicle reward functions are recovered via the Multiagent Adversarial Inverse-Reinforcement-Learning approach. Road user optimal policies and collision avoidance mechanisms are predicted using multiagent Actor-Critic deep-reinforcement-learning. The results demonstrate the superiority of the multiagent modeling approach in predicting road user behavior, their collision avoidance mechanisms, and the Post-Encroachment-Time (PET) compared to a baseline single-agent model.

Lateral Pre-crash Sensing and Avoidance in Emotion Enabled Cognitive Agent based Vehicle-2-Vehicle Communication System

A novel inter-vehicle communication system based on emotion enabled cognitive (EEC) agent has been anticipated as an intelligent solution to evade the road catastrophe due to hasty decisions by drivers. An input stimulus is processed in human brain using a short route and a long route during any emergency situation. The proposed EEC agent, mounted inside a vehicle, acts like a human brain and is stirred by short route information processing mechanism of human brain in fear condition. The results are acquired for the decisions made by the proposed approach through the EEC agent using short route and human drivers using long route during urgent situations. A pre crash sensing and avoidance algorithm has been proposed as well to mitigate the lateral or side by side collision using EEC agent. Experimental findings reveal that by commencing emotions with cognition, and using formulated L-PCSA algorithm, lateral collisions chances can be sensed and avoided to secure the valued lives of passengers. It has been pragmatic that the new approach is very effectual and useful in shunning the lateral road collisions.

Wildlife‐vehicle collisions in Moroccan Atlantic Sahara: Impact on resident species and Afro‐Palearctic birds for conservation purposes

AbstractIn this study, we described the temporal and spatial patterns of wildlife‐vehicle collisions in Dakhla‐Laayoune regions, using counts from cars for two years (2018–2021). A total of 462 roadkills were documented (monthly average of 0.075 roadkills/km), comprising 16 species, including mammals (residents), reptiles (residents), and birds (residents and migrants). Seasonal abundance and diversity of roadkills revealed a high peak during breeding and migration periods for all species. The generalised linear mixed models indicated that wildlife‐vehicle accidents have a higher probability of occurrence close to breeding and feeding sites and road sections with increased vegetation cover. Similarly, road collisions were higher close to migration routes and during breeding periods. Our study provides the first and only detailed results on wildlife road collisions in Morocco. Data gathered from this study provide valuable information for conservation measures of the most affected Moroccan species and Palearctic migrant birds which are crossing between Europe and Africa. Moreover, our data provide a new window of large‐scale comparative studies addressing the impacts of the infrastructures on biodiversity in the entire Atlantic coasts of Northwest Africa. Detailed studies on the local population dynamics of highly affected species are urgently needed to conclude whether these levels of road mortality are supportable.

Integrating user Psychology in Road Transport Safety: A Model for Low and Middle-Income Countries, Based on Sri Lanka

Low and Middle Income Countries (LMICs) account for 90% of road collisions that occur globally. These collisions account for 17.63 million serious injuries and deaths while costing LMIC economies 1.7 trillion USD amounting to 6.5% of aggregate annual GDP. Though there are a plethora of models on road transport safety, few address the impact of user psychology and the contextual environment in LMICs on road safety. This paper proposes such a model formulated and developed using existing literature and focus group discussions. The model is composed of three key constituents - behaviour, skills and knowledge, and infrastructure. It further interprets behaviour as an integration of three sub-elements—mindset, circumstances and legal system—pertaining to the incident. The interaction between these constructs constitutes user behaviour in a given circumstance. The research also proposes an implementation guideline to authorities that is instrumental in changing society’s perception of transport safety.

Perspectives on Intelligent Road Lighting Control

Road lighting is best known to have the potential to improve night-time safety. The vast majority of research studies have shown that the installation of road lighting has favourable effects on the number and severity of road collisions. In addition, some evidence suggests that current road lighting practices are over-lighting the roadway. Therefore, the rapidly growing concerns about energy consumption and its related costs will force further changes in road lighting technology and practice. To date, there have been no studies that investigated the association between current knowledge and the road lighting guidelines. This article reviews and discusses CIE road lighting guidelines for motorised traffic. It then provides current knowledge in the area of lighting motorised roads leading to possibilities for intelligent lighting control. Finally, suggestions for considering dimming of road lighting based on recent developments in science and technology are provided. The three recommended approaches, static road lighting, semi-dynamic lighting levels, and dynamic lighting levels, could save energy without compromising traffic safety.

Determining prominent factors across system hierarchies to improve road safety in LMICs: A case study of Bangladesh

Road traffic collisions have become a major safety concern in low- and middle-income countries (LMIC). Traditional collision investigation methods usually identify lower system-level factors whereas systems-based sociotechnical approaches address higher system-level factors contributing to crashes. Accimap, one of the most popular systems-based collision analysis method has been used in this study to analyze eight road collisions varying in terms of collision type, severity, and surroundings in Bangladesh, a low-income country, and address both the higher and lower system-level factors. The inter-connected factors contributing to each collision are mapped according to eight system levels starting from environment and equipment to international committees level, and aggregated to create a generic Accimap having 175 unique factors. Further, network analysis metrics i.e., indegree centrality, outdegree centrality, betweenness centrality, sociometric status, clique census, and global efficiency of network have been evaluated to find out the prominent contributory factors. The results from combining Accimap and network analysis show that improperly designed roads, risky overtaking, and delayed emergency response at the lower system levels have prominent contribution to collisions. Additionally, higher system-level factors such as lack of supervision or monitoring, inefficient fund distribution, lack of enforcement, insufficient infrastructures, and lack of planning play a vital role in the causation of road traffic collisions. So, based on the combined approach, improving road infrastructure, enhancing law enforcement, increasing supervision and monitoring, educating drivers, and properly planned fund distribution are identified as the leverage points which can be utilized by the road safety stakeholders to build a safer road system.

Does predation by wolves reduce collisions between ungulates and vehicles in France?

ABSTRACT The recolonization of French forests by wolves is met with mixed feelings. While the spotlight is on predation of wolves on livestock, the potential benefit of the return of wolves has attracted less attention. Previous research has failed to assess the economic value of these benefits and focused only on the cost of predation. Our study proposes a modeling exercise to evaluate one of the services provided by wolves, namely, the reduction of collisions between wildlife and vehicles in France. We estimated that in 2018 the consumption of roe deer and wild boar by wolves saved between 2.4–7.8 million euros by preventing casualties and damage to property in road collisions. Offsetting this benefit with the costs of managing livestock predation could move the debate away from a simple clash between herders and environmentalists toward a more systemic approach and thus improve coexistence between wolves and humans.

Intelligent Deep Data Analytics Based Remote Sensing Scene Classification Model

Latest advancements in the integration of camera sensors paves a way for new Unmanned Aerial Vehicles (UAVs) applications such as analyzing geographical (spatial) variations of earth science in mitigating harmful environmental impacts and climate change. UAVs have achieved significant attention as a remote sensing environment, which captures high-resolution images from different scenes such as land, forest fire, flooding threats, road collision, landslides, and so on to enhance data analysis and decision making. Dynamic scene classification has attracted much attention in the examination of earth data captured by UAVs. This paper proposes a new multi-modal fusion based earth data classification (MMF-EDC) model. The MMF-EDC technique aims to identify the patterns that exist in the earth data and classifies them into appropriate class labels. The MMF-EDC technique involves a fusion of histogram of gradients (HOG), local binary patterns (LBP), and residual network (ResNet) models. This fusion process integrates many feature vectors and an entropy based fusion process is carried out to enhance the classification performance. In addition, the quantum artificial flora optimization (QAFO) algorithm is applied as a hyperparameter optimization technique. The AFO algorithm is inspired by the reproduction and the migration of flora helps to decide the optimal parameters of the ResNet model namely learning rate, number of hidden layers, and their number of neurons. Besides, Variational Autoencoder (VAE) based classification model is applied to assign appropriate class labels for a useful set of feature vectors. The proposed MMF-EDC model has been tested using UCM and WHU-RS datasets. The proposed MMF-EDC model attains exhibits promising classification results on the applied remote sensing images with the accuracy of 0.989 and 0.994 on the test UCM and WHU-RS dataset respectively.

Visual inspection of carcasses to assess the individual condition of European roe deer Capreolus capreolus

The aim of the study was to demonstrate correlation between selected body condition indices of roe deer Capreolus capreolus (KFI, carcass weight) and the health status of the carcass (i.e. pathological changes found through external and internal examination of the carcass) of deer killed by hunting or road traffic. The study material consisted of 34 carcasses of roe deer found dead due to road collisions or obtained through hunting in accordance with the Principles of Individual and Population Selection of Game Animals in Poland. The results showed a significant correlation between the occurrence of disease signs (especially the presence of parasites) and the mean values of indices describing the individual condition of animals (carcass weight, KFI). This confirms that the preliminary examination of game harvested and animals killed by road collisions plays an important role in inhibiting the spread of epizootic diseases and in assessing the individual condition of animals.