The data set contains registered vehicle population count by various criteria such as vehicle class, vehicle status, vechicle make, vehicle model, vehicle year, plate class, plate declaration, county, weight related class and other vehicle decriptors.

This dataset shows the number of vehicles that have passed under a gantry on that particular day. This dataset does not show trips, it only shows segments. Segments are compiled to make trips. There are 10 gantries on the InterCounty Connector (ICC) and 5 interchanges. The eastbound gantries are 101, 105, 107, 109, 113, and the westbound gantries are 102, 106, 108, 110, 114. The dataset has a column for each gantry going east and west, then a total for each gantry. The ICC is an all electronic tolling road which opened February 2011. The first opening was a partial opening, with only the first interchange being available for use.There was a free period from February 23, 2011 through March 6, 2011. The rest of the ICC opened in November 2011, and there was another free period from November 22, 2011 through December 4, 2011. There are a few days where a low number of traffic passed under gantries (rows 196,198, 269,271...), these were either testing periods or construction vehicles.

i.e.

We collected and annotated a dataset containing 105,544 annotated vehicle instances from 24700 image frames within seven different videos, sourced online under creative commons license. The video frames are annotated using DarkLabel tool. In the interest of reusability and generalisation of the deep learning model, we consider the diversity within the collected dataset. This diversity includes changes of lighting amongst the video, as well as other factors such as weather conditions, angle of observation, varying speed of the moving vehicles, traffic flow, and road conditions etc. The videos collected obviously include stationary vehicles, to perform the validation of stopped vehicle detection method. It can be noticed that the road conditions (e.g., motorways, city, country roads), directions, data capture timings and camera views, vary in the dataset producing annotated dataset with diversity. the dataset may have several uses such as vehicle detection, vehicle identification, stopped vehicle detection on smart motorways and local roads (smart city applications) and many more.

The feature class indicates the specific types of motorized vehicles allowed on the designated routes and their seasons of use. The feature class is designed to be consistent with the MVUM (Motor Vehicle Use Map). It is compiled from the GIS Data Dictionary data and NRM Infra tabular data that the administrative units have prepared for the creation of their MVUMs. Only roads with a SYMBOL attribute value of 1, 2, 3, 4, 11, and 12 are Forest Service System roads and contain data concerning their availability for OHV (Off Highway Vehicle) use. This data is published and refreshed on a unit by unit basis as needed. Data for each individual unit must be verified and proved consistent with the published MVUMs prior to publication.The Forest Service's Natural Resource Manager (NRM) Infrastructure (Infra) is the agency standard for managing and reporting information about inventory of constructed features and land units as well as the permits sold to the general public and to partners. Metadata

The U.S. auto industry sold nearly ************* cars in 2024. That year, total car and light truck sales were approximately ************ in the United States. U.S. vehicle sales peaked in 2016 at roughly ************ units. Pandemic impact The COVID-19 pandemic deeply impacted the U.S. automotive market, accelerating the global automotive semiconductor shortage and leading to a drop in demand during the first months of 2020. However, as demand rebounded, new vehicle supply could not keep up with the market. U.S. inventory-to-sales ratio dropped to its lowest point in February 2022, as Russia's war on Ukraine lead to gasoline price hikes. During that same period, inflation also impacted new and used car prices, pricing many U.S. consumers out of a market with increasingly lower car stocks. Focus on fuel economy The U.S. auto industry had one of its worst years in 1982 when customers were beginning to feel the effects of the 1973 oil crisis and the energy crisis of 1979. Since light trucks would often be considered less fuel-efficient, cars accounted for about ** percent of light vehicle sales back then. Thanks to improved fuel economy for light trucks and cheaper gas prices, this picture had completely changed in 2020. That year, prices for Brent oil dropped to just over ** U.S. dollars per barrel. The decline occurred in tandem with lower gasoline prices, which came to about **** U.S. dollars per gallon in 2020 - and cars only accounted for less than one-fourth of light vehicle sales that year. Four years on, prices are dropping again, after being the highest on record since 1990 in 2022.

This dataset shows counts of transactions associated with authorizing vehicles to be used on public roads, commonly referred to as “buying tabs” or “buying tags”. The data includes registration activity by fuel type, county, primary use class, and date. This is comparable to the Fee Distribution Report #13, that is titled "Motor Vehicle Registration By Class and County".

Kenya Road Transport: Number of Motor Vehicles: Registered data was reported at 4,588,770.000 Unit in 2022. This records an increase from the previous number of 4,353,891.000 Unit for 2021. Kenya Road Transport: Number of Motor Vehicles: Registered data is updated yearly, averaging 2,011,972.000 Unit from Dec 2004 (Median) to 2022, with 19 observations. The data reached an all-time high of 4,588,770.000 Unit in 2022 and a record low of 711,142.000 Unit in 2004. Kenya Road Transport: Number of Motor Vehicles: Registered data remains active status in CEIC and is reported by Kenya National Bureau of Statistics. The data is categorized under Global Database’s Kenya – Table KE.TA001: Road Transport: Number of Motor Vehicles: Registered.

Vehicle Information Vehicle Type: This column represents the type of vehicle. Possible values include:

Car: A standard passenger vehicle. Truck: A larger vehicle used for transporting goods. Bus: A vehicle designed to carry multiple passengers. Motorcycle: A two-wheeled motor vehicle. Fuel Type: This column indicates the type of fuel the vehicle uses. Possible values are:

Petrol: Also known as gasoline, a common fuel for internal combustion engines. Diesel: A type of fuel used in diesel engines, often found in larger vehicles like trucks and buses. Electric: Vehicles powered by electric batteries. Hybrid: Vehicles that use a combination of an internal combustion engine and electric propulsion. Engine Size: The size of the vehicle's engine, measured in liters. Larger engines typically produce more power and can lead to higher emissions.

Age of Vehicle: The age of the vehicle in years. Older vehicles may have higher emissions due to wear and tear or outdated technology.

Mileage: The total distance the vehicle has traveled, measured in kilometers or miles. Higher mileage can indicate more wear and potentially higher emissions.

Driving Conditions Speed: The average speed of the vehicle during the measurement period, measured in kilometers per hour (km/h) or miles per hour (mph). Vehicle emissions can vary with speed.

Acceleration: The rate at which the vehicle's speed increases, measured in meters per second squared (m/s²). Rapid acceleration can lead to higher emissions.

Road Type: The type of road the vehicle is driving on. Possible values include:

Highway: Major roads designed for fast travel. City: Urban roads with frequent stops and lower speeds. Rural: Country roads that may have varying conditions. Traffic Conditions: The level of traffic during the measurement period. Possible values include:

Free flow: Minimal traffic, allowing for smooth travel. Moderate: Some traffic, but generally steady movement. Heavy: High traffic, often leading to stop-and-go conditions. Environmental Conditions Temperature: The ambient temperature during the measurement period, measured in degrees Celsius (°C) or Fahrenheit (°F). Temperature can affect engine performance and emissions.

Humidity: The relative humidity of the air during the measurement period, measured as a percentage. Humidity can affect the combustion process and emissions.

Wind Speed: The speed of the wind during the measurement period, measured in meters per second (m/s) or kilometers per hour (km/h). Wind can influence the dispersion of emissions.

Air Pressure: The atmospheric pressure during the measurement period, measured in hectopascals (hPa). Air pressure can affect engine efficiency and emissions.

Emission Data CO2 Emissions: The amount of carbon dioxide emitted by the vehicle, measured in grams per kilometer (g/km). CO2 is a major greenhouse gas contributing to climate change.

NOx Emissions: The amount of nitrogen oxides emitted by the vehicle, measured in grams per kilometer (g/km). NOx contributes to air pollution and can cause respiratory problems.

PM2.5 Emissions: The amount of particulate matter with a diameter of 2.5 micrometers or smaller emitted by the vehicle, measured in grams per kilometer (g/km). PM2.5 can penetrate deep into the lungs and cause health issues.

VOC Emissions: The amount of volatile organic compounds emitted by the vehicle, measured in grams per kilometer (g/km). VOCs contribute to the formation of ground-level ozone and smog.

SO2 Emissions: The amount of sulfur dioxide emitted by the vehicle, measured in grams per kilometer (g/km). SO2 can contribute to acid rain and respiratory problems.

Target Variable Emission Level: This column categorizes the overall emission level of the vehicle into three classes: Low: Vehicles with low emissions. Medium: Vehicles with moderate emissions. High: Vehicles with high emissions. Summary Categorical Features: Vehicle Type, Fuel Type, Road Type, Traffic Conditions, Emission Level. Continuous Numerical Features: Engine Size, Age of Vehicle, Mileage, Speed, Acceleration, Temperature, Humidity, Wind Speed, Air Pressure, CO2 Emissions, NOx Emissions, PM2.5 Emissions, VOC Emissions, SO2 Emissions.

Data collected between 2005 to 2007, 3% of sample collected in 2005, 51% in 2006 and 46% in 2007.

This dataset contains a compilation of data collected from different sources.

VOG 06/ VOG 08 (Value of Goods) :
The data is derived from the information collected in the 2006 Ontario Commercial Vehicle Survey. This survey is a roadside intercept survey of truck drivers, which collects information about the trip, commodity and the vehicle. The survey primarily captures intercity trucking activity and under-represents truck flows in urban areas. The value of goods provided in this table is derived from the Commercial Vehicle Survey, but factored up to represent the overall trucking activity on the network segment for 2006 and 2008.

**AADTT 2006 and ****AADTT** 2008:
The data is derived from the Ministry of Transportation's (MTO) inventory of annual traffic data for the Provincial Highways. The commercial volumes are first calculated using the AADT and the Commercial Percentage values for each traffic segment. These values are then adjusted to remove variations between segments caused by fluctuations in AADT.

The volume given for each direction is one-half of the total value. MTO does not maintain volume by direction. For freeway segments with core/collector configuration, the total volume is divided into four equal portions and assigned to each stream.

**Hourly Truck Volumes ( WD00-23 and WN00-23): **
These fields contain estimates of average hourly volumes for a typical weekday and weekend day. The estimates are based on observed hourly distribution at more than 100 directional Commercial Vehicle Survey sites across the province, AADTT and other information.

RD _NAME:
Name of the road

VOG 06:
2006 average daily value of goods assigned to road network link by directions.

VOG 08:
2008 average daily value of goods assigned to road network link by directions.

AADTT 2006:
2006 Annual Average Daily Truck Traffic; it is the truck volume assigned to road network link by directions.

AADTT 2008:
2008 Annual Average Daily Truck Traffic; it is the truck volume assigned to road network link by directions.

WD 00-23:
2008 Weekday ( WD ) hourly truck volume; 00 - 23 represents starting hour of the day (e.g. 12 represents 12 P.M. - 1 P.M.).

WN 00-23:
2008 Weekend ( WN ) hourly truck volume; 00 - 23 represents starting hour of the day (e.g. 12 represents 12 P.M. - 1 P.M.).

*[ WD]: Week day *[VOG]: Value of Goods *[AADTT]: Annual Average Daily Truck Traffic *[WN]: Week end *[RD]: Road *[WD]: Week day *[MTO]: Ministry of Transportation *[AADT]: Annual Average Daily Traffic

This annual release provides a snapshot of the number of active vehicle registration counts of light-duty vehicles and medium-duty vehicles by type of vehicle and fuel type, heavy-duty vehicles, buses, and motorcycles and mopeds. Data are obtained from the administrative files from provincial and territorial governments.

We present the GLOBAL ROADKILL DATA, the largest worldwide compilation of roadkill data on terrestrial vertebrates. We outline the workflow (Fig. 1) to illustrate the sequential steps of the study, in which we merged local-scale survey datasets and opportunistic records into a unified roadkill large dataset comprising 208,570 roadkill records. These records include 2283 species and subspecies from 54 countries across six continents, ranging from 1971 to 2024.Large roadkill datasets offer the advantage ofpreventing the collection of redundant data and are valuable resources for both local and macro-scale analyses regarding roadkill rates, road and landscape features associated with roadkill risk, species more vulnerable to road traffic, and populations at risk due to additional mortality. The standardization of data - such as scientific names, projection coordinates, and units - in a user-friendly format, makes themreadily accessible to a broader scientific and non-scientific community, including NGOs, consultants, public administration officials, and road managers. The open-access approach promotes collaboration among researchers and road practitioners, facilitating the replication of studies, validation of findings, and expansion of previous work. Moreover, researchers can utilize suchdatasets to develop new hypotheses, conduct meta-analyses, address pressing challenges more efficiently and strengthen the robustness of road ecology research. Ensuring widespreadaccess to roadkill data fosters a more diverse and inclusive research community. This not only grants researchers in emerging economies with more data for analysis, but also cultivates a diverse array of perspectives and insightspromoting the advance of infrastructure ecology.MethodsInformation sources: A core team from different continents performed a systematic literature search in Web of Science and Google Scholar for published peer-reviewed papers and dissertations. It was searched for the following terms: “roadkill* OR “road-kill” OR “road mortality” AND (country) in English, Portuguese, Spanish, French and/or Mandarin. This initiative was also disseminated to the mailing lists associated with transport infrastructure: The CCSG Transport Working Group (WTG), Infrastructure & Ecology Network Europe (IENE) and Latin American & Caribbean Transport Working Group (LACTWG) (Fig. 1). The core team identified 750 scientific papers and dissertations with information on roadkill and contacted the first authors of the publications to request georeferenced locations of roadkill andofferco-authorship to this data paper. Of the 824 authors contacted, 145agreed to sharegeoreferenced roadkill locations, often involving additional colleagues who contributed to data collection. Since our main goal was to provide open access to data that had never been shared in this format before, data from citizen science projects (e.g., globalroakill.net) that are already available were not included.Data compilation: A total of 423 co-authors compiled the following information: continent, country, latitude and longitude in WGS 84 decimal degrees of the roadkill, coordinates uncertainty, class, order, family, scientific name of the roadkill, vernacular name, IUCN status, number of roadkill, year, month, and day of the record, identification of the road, type of road, survey type, references, and observers that recorded the roadkill (Supplementary Information Table S1 - description of the fields and Table S2 - reference list). When roadkill data were derived from systematic surveys, the dataset included additional information on road length that was surveyed, latitude and longitude of the road (initial and final part of the road segment), survey period, start year of the survey, final year of the survey, 1st month of the year surveyed, last month of the year surveyed, and frequency of the survey. We consolidated 142 valid datasets into a single dataset. We complemented this data with OccurenceID (a UUID generated using Java code), basisOfRecord, countryCode, locality using OpenStreetMap’s API (https://www.openstreetmap.org), geodeticDatum, verbatimScientificName, Kingdom, phylum, genus, specificEpithet, infraspecificEpithet, acceptedNameUsage, scientific name authorship, matchType, taxonRank using Darwin Core Reference Guide (https://dwc.tdwg.org/terms/#dwc:coordinateUncertaintyInMeters) and link of the associatedReference (URL).Data standardization - We conducted a clustering analysis on all text fields to identify similar entries with minor variations, such as typos, and corrected them using OpenRefine (http://openrefine.org). Wealsostandardized all date values using OpenRefine. Coordinate uncertainties listed as 0 m were adjusted to either 30m or 100m, depending on whether they were recorded after or before 2000, respectively, following the recommendation in the Darwin Core Reference Guide (https://dwc.tdwg.org/terms/#dwc:coordinateUncertaintyInMeters).Taxonomy - We cross-referenced all species names with the Global Biodiversity Information Facility (GBIF) Backbone Taxonomy using Java and GBIF’s API (https://doi.org/10.15468/39omei). This process aimed to rectify classification errors, include additional fields such as Kingdom, Phylum, and scientific authorship, and gather comprehensive taxonomic information to address any gap withinthe datasets. For species not automatically matched (matchType - Table S1), we manually searched for correct synonyms when available.Species conservation status - Using the species names, we retrieved their conservation status and also vernacular names by cross-referencing with the database downloaded from the IUCNRed List of Threatened Species (https://www.iucnredlist.org). Species without a match were categorized as "Not Evaluated".Data RecordsGLOBAL ROADKILL DATA is available at Figshare27 https://doi.org/10.6084/m9.figshare.25714233. The dataset incorporates opportunistic (collected incidentally without data collection efforts) and systematic data (collected through planned, structured, and controlled methods designed to ensure consistency and reliability). In total, it comprises 208,570 roadkill records across 177,428 different locations(Fig. 2). Data were collected from the road network of 54 countries from 6 continents: Europe (n = 19), Asia (n = 16), South America (n=7), North America (n = 4), Africa (n = 6) and Oceania (n = 2).(Figure 2 goes here)All data are georeferenced in WGS84 decimals with maximum uncertainty of 5000 m. Approximately 92% of records have a location uncertainty of 30 m or less, with only 1138 records having location uncertainties ranging from 1000 to 5000 m. Mammals have the highest number of roadkill records (61%), followed by amphibians (21%), reptiles (10%) and birds (8%). The species with the highest number of records were roe deer (Capreolus capreolus, n = 44,268), pool frog (Pelophylax lessonae, n = 11,999) and European fallow deer (Dama dama, n = 7,426).We collected information on 126 threatened species with a total of 4570 records. Among the threatened species, the giant anteater (Myrmecophaga tridactyla, VULNERABLE) has the highest number of records n = 1199), followed by the common fire salamander (Salamandra salamandra, VULNERABLE, n=1043), and European rabbit (Oryctolagus cuniculus, ENDANGERED, n = 440). Records ranged from 1971 and 2024, comprising 72% of the roadkill recorded since 2013. Over 46% of the records were obtained from systematic surveys, with road length and survey period averaging, respectively, 66 km (min-max: 0.09-855 km) and 780 days (1-25,720 days).Technical ValidationWe employed the OpenStreetMap API through Java todetect location inaccuracies, andvalidate whether the geographic coordinates aligned with the specified country. We calculated the distance of each occurrence to the nearest road using the GRIP global roads database28, ensuring that all records were within the defined coordinate uncertainty. We verified if the survey duration matched the provided initial and final survey dates. We calculated the distance between the provided initial and final road coordinates and cross-checked it with the given road length. We identified and merged duplicate entries within the same dataset (same location, species, and date), aggregating the number of roadkills for each occurrence.Usage NotesThe GLOBAL ROADKILL DATA is a compilation of roadkill records and was designed to serve as a valuable resource for a wide range of analyses. Nevertheless, to prevent the generation of meaningless results, users should be aware of the followinglimitations:- Geographic representation – There is an evident bias in the distribution of records. Data originatedpredominantly from Europe (60% of records), South America (22%), and North America (12%). Conversely, there is a notable lack of records from Asia (5%), Oceania (1%) and Africa (0.3%). This dataset represents 36% of the initial contacts that provided geo-referenced records, which may not necessarily correspond to locations where high-impact roads are present.- Location accuracy - Insufficient location accuracy was observed for 1% of the data (ranging from 1000 to 5000 m), that was associated with various factors, such as survey methods, recording practices, or timing of the survey.- Sampling effort - This dataset comprised both opportunistic data and records from systematic surveys, with a high variability in survey duration and frequency. As a result, the use of both opportunistic and systematic surveys may affect the relative abundance of roadkill making it hard to make sound comparisons among species or areas.- Detectability and carcass removal bias - Although several studies had a high frequency of road surveys,the duration of carcass persistence on roads may vary with species size and environmental conditions, affecting detectability. Accordingly, several approaches account for survey frequency and target speciesto estimate more

The dataset is outcome of a paper "Floating Car Data Map-matching Utilizing the Dijkstra Algorithm" accepted for 3rd International Conference on Data Management, Analytics & Innovation held in Kuala Lumpur, Malaysia in 2019.

The floating car data (FCD representing movement of cars with their position in time) is produced by the traffic simulator software (further referred to as Simulator) published in [1] and can be used as an input for data processing and benchmarking. The dataset contains FCD of various quality levels based on the routing graph of the Czech Republic derived from Open Street Map openstreetmap.org.

Should the dataset be exploited in scientific or other way, any acknowledgement or references to our paper [1] and dataset are welcomed and highly appreciated.

Archive contents

The archive contains following folders.

city_oneway and city_roadtrip - FCD from the city of Brno, Czech Republic where FCD is based on Origin-Destination in case of oneway and Origin-Destination-Origin in case of a road trip

intercity_oneway and intercity_roadtrip - FCD from cities of Brno, Ostrava, Olomouc and Zlin, all Czech Republic where FCD is based on Origin-Destination in case of oneway and Origin-Destination-Origin in case of a road trip

Content explanation

All four of mentioned folders contain raw FCD as they come from our Simulator, post-processed FCD enriching Simulator FCD, and obfuscated raw FCD (of both low and high obfuscation level). In the both obfuscated data sets, each measured point was moved in a random direction a number of meters given by drawing a number from a Gaussian distribution. We utilized two Gaussian distributions, one for the roads outside the city (N(0,10) for the lower and N(0,20) for the higher obfuscation level) and one for the roads inside the city (N(0,15) and N(0,30) respectively). Then some predefined number of randomly chosen points were removed (3% in our case). This approach should roughly represent real conditions encountered by FCD data as described by El Abbous and Samanta [2].

In case of post-processed road trip data, there is one extra dataset with "cache" suffix representing the very same dataset limited to a 5-minute session memoization. This folder also contains a picture of processed FCD represented on a map.

Data format Standard UTF-8 encoded CSV files, separated by a semicolon with the following columns:

RAW

Header

session_id;timestamp;lat;lon;speed;bearing;segment_id

Data

session_id: (Type: unsigned INT) - session (car) identifier timestamp: (Type: datetime) - timestamp in UTC lat: (Type: unsigned long) - latitude as used in Google maps lon: (Type: unsigned long) - longitude as used in Google maps speed: (Type: unsigned INT) - actual speed in kmh bearing: (Type: unsigned INT) - actual bearing in angles 0-360 segment_id: (Type: unsigned long) - unique edge identifier

POST-PROCESSED

Header

gid;car_id;point_time;lat;lon;segment_id;speed_kmh;speed_avg_kmh;distance_delta_m;distance_total_m;speedup_ratio;duration;segment_changed;duration_segment;moved;duration_move;good;duration_good;bearing;interpolated

Data

gid: (Type: unsigned long) - global identifier of a record car_id: (Type: unsigned INT) - session (car) identifier point_time: (Type: datetime) - timestamp with timezone lat: (Type: unsigned long) - latitude as used in Google maps lon: (Type: unsigned long) - longitude as used in Google maps segment_id: (Type: unsigned long) - unique edge identifier speed: (Type: unsigned INT) - actual speed in kmh speed_avg_kmh: (Type: unsigned long) - actual average speed of a car in kmh distance_delta_m: (Type: unsigned long) - actual distance delta in metres distance_total_m: (Type: unsigned long) - actual total distance of a car in metres speedup_ratio: (Type: unsigned long) - actual speed-up ratio of a car duration: (Type: time) - actual duration of a car segment_changed: (Type: boolean) - signals if actual segment of a car differs from the previous one duration_segment: (Type: time) - actual duration on a segment of a car moved: (Type: boolean) - signals if actual position of a car differs from the previous one duration_move:(Type: time) - actual duration of a car since moving good: signals if actual record values satisfies all data constraints (all true as derived from Simulator) duration_good: actual duration of a car since when all constraints conditions satisfied bearing: (Type: unsigned INT) - actual bearing in angles 0-360 interpolated: (Type: boolean) - signals if actual segment identifier is calculated (all false as derived from Simulator)

References

[1] V. Ptošek, J. Ševčík, J. Martinovič, K. Slaninová, L. Rapant, and R. Cmar, Real-time traffic simulator for self-adaptive navigation system validation, Proceedings of EMSS-HMS: Modeling & Simulation in Logistics, Traffic & Transportation, 2018.

[2] A. El Abbous and N. Samanta. A modeling of GPS error distri-butions, In proceedings of 2017 European Navigation Conference (ENC), 2017.

Motor Trend Car Road TestsDescriptionThe data was extracted from the 1974 Motor Trend US magazine, and comprises fuel consumption and 10 aspects of automobile design and performance for 32 automobiles (1973–74 models).FormatA data frame with 32 observations on 11 variables.1 - mpg: Miles/(US) gallon2 - cyl: Number of cylinders3 - disp: Displacement (cu.in.)4 - hp: Gross horsepower5 - drat: Rear axle ratio6 - wt: Weight (1000 lbs)7 - qsec: 1/4 mile time8 - vs: Engine shape (0 = v-shaped, 1 = straight)9 - am: Transmission (0 = automatic, 1 = manual)10 - gear: Number of forward gears11 - carb: Number of carburettorsSourceHenderson and Velleman (1981), Building multiple regression models interactively. Biometrics, 37, 391–411.

Number of vehicles authorized to drive in Quebec, both for road vehicles and for vehicles designed for off-road traffic. The data has been revised to comply with the new provisions of Bill 25 protecting the privacy of Quebecers.

The datasets contain the subject ADAS-equipped vehicle’s trajectory collected in naturalistic traffic conditions in central Ohio. The instrumented subject vehicle was either a discreet or readily-identifiable ADAS-equipped vehicle with SAE L2 capabilities. The dataset also contains trajectories for adjacent vehicles in traffic (observed by the subject vehicle’s onboard sensors).

15 smart sensors were installed on Mill Road and surrounding streets to record numbers of pedestrians, bicycles, cars and other vehicles. The data being collated and analysed by the Smart Cambridge programme will help the Greater Cambridge Partnership understand how people use the road network.

Data will be released monthly for these locations until the end of 2020. Please note that due to the level of insight that can be gained from these sensors, additional sensors in more locations have been installed in Cambridge since the summer of 2019. Some sensors will remain beyond 2020 in strategic locations and the network is expected to grow. Data for those more permanent sites, outside of the Mill Road project will be published here: https://data.cambridgeshireinsight.org.uk/dataset/cambridge-city-smart-s...

Mill Road Bridge was closed for eight weeks from 1 July 2019 for crucial work being carried out to improve rail services. Pedestrians and cyclists will still be able to cross the railway for most of the working time.

A high concentration of sensors were installed for approximately 18 months to gather data before the closure, during the time when there is no vehicle traffic coming over Mill Road Bridge and then after the bridge is re-opened. This has allowed engineers to see the impact of the closure on surrounding roads, including on air quality. Keeping the sensors in place for this long has also allowed teams to make greater comparisons, by taking in to account daily, weekly, monthly and annual variations in traffic levels.

The below data release offers counts for each sensor over 1 hour periods. The curent data covers the period 03/06/2019 to 13/12/2020.

Hourly counts are broken down by inbound and outbound journeys. .

Counts are also broken down by vehicle type. This includes:

Pedestrians Cyclists Buses LGV OGV 1 OGV 2 The release also includes a full list of sensor sites with geographic point location data.

Context

The concept of searching and localizing vehicles from live traffic videos based on descriptive textual input has yet to be explored in the scholarly literature. Endowing Intelligent Transportation Systems (ITS) with such a capability could help solve crimes on roadways. While artificial intelligence (AI) can be a powerful tool for this data intensive application, existing state-of-the-art AI models struggle with fine-grain vehicle recognition. Typically, only reporting model performance on still input image data, often captured at high resolution and at pristine quality. These settings are not reflective of real-world operating conditions and thus, recognition accuracies typically cannot be replicated on video data. One major impediment to the advancement of fine-grain vehicle recognition models is the lack of video testbench datasets with annotated ground-truth data. Additionally, to the best of our knowledge, no metrics currently exist for evaluating the robustness and performance efficiency of a vehicle recognition model on live videos, and even less so for vehicle search and localization models. In this paper, we address these challenges by proposing V-Localize, a novel artificial intelligence framework for vehicle search and continuous localization captured from live traffic videos based on input textual descriptions. An efficient hashgraph algorithm is introduced to process input text (such as a sentence, paragraph, or report) to extract detailed target information used to query the recognition and localization model. This work further introduces two novel datasets that will help advance AI research in these challenging areas. These datasets include: a) the most diverse and large-scale Vehicle Color Recognition (VCoR) dataset with 15 colors classes -- twice as many as the number of color classes in the largest existing such dataset -- to facilitate finer-grain recognition with color information; and b) a Vehicle Recognition in Video (VRiV) dataset, which is a first of its kind video test-bench dataset for evaluating the performance of vehicle recognition models in live videos rather than still image data. The VRiV dataset will open new avenues for AI researchers to investigate innovative approaches that were previously intractable due to the lack of a traffic vehicle recognition annotated test-bench video dataset. Finally, to address the gap in the field, 5 novel metrics are introduced in this paper for adequately accessing the performance of vehicle recognition models in live videos. Ultimately, the proposed metrics could also prove intuitively effective at quantitative model evaluation in other video recognition applications. The novel metrics and VRiV test-bench dataset introduced in this paper are specifically aimed at advancing state-of-the-art research for vehicle recognition in videos. Likewise, the proposed novel vehicle search and continuous localization framework could prove assistive in cases such as of amber alerts or hit-and-run incidents. One major advantage of the proposed system is that it can be integrated into intelligent transportation system software to help aid law-enforcement.

Image Acquisition

The proposed Vehicle Recognition in Video (VRiV) dataset is the first of its kind and is aimed at developing, improving, and analyzing performance of vehicle search and recognition models on live videos. The lack of such a dataset has limited performance analysis of modern fine-grain vehicle recognition systems to only still image input data, making them less suitable for video applications. The VRiV dataset is introduced to help bridge this gap and foster research in this direction. The proposed VRiV dataset consists of up to 47 video sequences averaging about 38.5 seconds per video. The videos are recorded in a traffic setting focusing on vehicles of volunteer candidates whose ground truth make, model, year and color information are known. For security reasons and safety of participants, experiments are conducted on streets/road with low traffic density. For each video, there is a target vehicle with known ground truth information, and there are other vehicles either moving in traffic or parked on side streets, to simulate real-world traffic scenario. The goal is for the algorithm to be able to search, recognize and continuously localize just the specific target vehicle of interest for the corresponding video based on the search query. It is worth noting that the ground truth information about other vehicles in the videos are not known. The 47 videos in the testbench dataset are distributed across 7 distinct makes and 17 model designs as shown in Figure 10. The videos are also annotated to include ground truth bounding boxes for the specific target vehicles in corresponding videos. The dataset includes more than 46k annotated frames averaging about 920 frames per video. This dataset will be made available on Kaggle, and new videos will be added as they become available.

Content

There is one main zip file available for download. The zip file contains 94 files. 1) 47 video files 2) 47 ground-truth annotated files which identifies locations where the vehicle of interest is in the frame. Each video file is labelled with the corresponding vehicle brand name, model, year, and color information.

Terms and Conditions

• Videos provided in this dataset are freely available for research and education purposes only. Please be sure to properly credit the authors by citing the article below.
• Be sure to upvote this dataset if you find it useful by scrolling up and clicking the ^ sign at the top-right corner of the cover image of this page.
• Be sure to blur out all plate numbers before publishing any of the contents available in this dataset.

Acknowledgements

Any publication using this database must reference to the following journal manuscript:

• Panetta, Karen, Landry Kezebou, Victor Oludare, James Intriligator, and Sos Agaian. 2021. "Artificial Intelligence for Text-Based Vehicle Search, Recognition, and Continuous Localization in Traffic Videos" AI 2, no. 4: 684-704. https://doi.org/10.3390/ai2040041
• open access : https://www.mdpi.com/2673-2688/2/4/41

Note: if the link is broken, please use http instead of https.

In Chrome, use the steps recommended in the following website to view the webpage if it appears to be broken https://www.technipages.com/chrome-enabledisable-not-secure-warning

VCoR dataset: https://www.kaggle.com/landrykezebou/vcor-vehicle-color-recognition-dataset VRiV dataset: https://www.kaggle.com/landrykezebou/vriv-vehicle-recognition-in-videos-dataset

For any enquires regarding the VCoR dataset, contact: landrykezebou@gmail.com