Road Segmentation Dataset

This dataset comprises a collection of images captured through DVRs (Digital Video Recorders) showcasing roads. Each image is accompanied by segmentation masks demarcating different entities (road surface, cars, road signs, marking and background) within the scene.

💴 For Commercial Usage: To discuss your requirements, learn about the price and buy the dataset, leave a request on our website to buy the dataset

The dataset can be utilized for enhancing computer vision algorithms involved in road surveillance, navigation, and intelligent transportation systemsand and in autonomous driving systems.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2Fb0789a0ec8075d9c7abdb0aa9faced59%2FFrame%2012.png?generation=1694606364403023&alt=media" alt="">

DATASETS WITH VEHICLES :

• License Plates - 1 209 438 OCR Plates
• Car Object Detection & Tracking
• Vehicle Images & Segmentation
• Parking Space Detection & Classification Dataset
• Videos around cars

Dataset structure

• images - contains of original images of roads
• masks - includes segmentation masks created for the original images
• annotations.xml - contains coordinates of the polygons, created for the original photo

Data Format

Each image from images folder is accompanied by an XML-annotation in the annotations.xml file indicating the coordinates of the polygons and labels . For each point, the x and y coordinates are provided.

Сlasses:

• road_surface: surface of the road,
• marking: white and yellow marking on the road,
• road_sign: road signs,
• car: cars on the road,
• background: side of the road and surronding objects

Example of XML file structure

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2Fa74a4214f4dd89a35527ef008abfc151%2Fcarbon.png?generation=1694608637609153&alt=media" alt="">

Roads Segmentation might be made in accordance with your requirements.

🧩 This is just an example of the data. Leave a request here to learn more

🚀 You can learn more about our high-quality unique datasets here

keywords: road surface, road scene, off-road, vehicle segmentation dataset, semantic segmentation for self driving cars, self driving cars dataset, semantic segmentation for autonomous driving, car segmentation dataset, car dataset, car images, car parts segmentation, self-driving cars deep learning, cctv, image dataset, image classification, semantic segmentation

This dataset focuses on optimizing the performance of various car brands under different driving conditions and performance parameters. It includes data collected from vehicles under five primary driving scenarios: Urban Driving, Highway Driving, Off-Road Conditions, Extreme Weather, and Heavy Traffic. Each scenario includes factors that influence vehicle performance such as fuel efficiency, engine performance, suspension stability, and safety features.

The dataset contains car brand-specific features designed to optimize performance in these conditions. These features include fuel efficiency, engine power (torque and horsepower), safety technologies, electric range for electric vehicles, driving comfort, and the durability of vehicle components. It is intended to help manufacturers and engineers analyze and improve vehicle performance in real-world driving scenarios.

Key Features:

Driving Scenarios: Urban, Highway, Off-Road, Extreme Weather, Heavy Traffic. Brand-Specific Performance Features: Fuel Efficiency, Engine Performance (Horsepower and Torque), Safety Features, Electric Range (EVs), Driving Comfort and Ride Quality, Reliability and Durability. Performance Metrics: Fuel Consumption, Emissions, Braking Efficiency, Tire Grip, Battery Efficiency, and Vehicle Stability.

The dataset contains year-wise data on population, number of vehicles registered, the total road length of highways and all roads, and number of vehicles per 1,000 population and per 1,00 kilometers of road length.

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".

Leeds Road Traffic Accidents

Number of Vehicles, Casualties and Impact Severity

By data.world's Admin [source]

About this dataset

This dataset provides detailed information on road traffic accidents across Leeds from 2009-2018. By analysing this data, we can gain invaluable insights into the locations, amounts of vehicles and people involved, road surfaces, weather conditions, and severities of any casualties in each accident.

In order to get the most out of this dataset for our projects, it is helpful to utilise Excel pivot tables for summarising large amounts of data in a concise manner. Additionally, certain figures may be repeated due to the format of the report (for example: Reference Number Grid Ref: Easting Grid Ref: Northing Number of vehicles Accident Date Time (24hr)). Please also note that due to poor internet connectivity at times when generating Eastings and Northings needed for accident location points - there could be errors that arise from these coordinates being inaccurate. These should then be directed to accident.studiesleeds.gov.uk.

The columns of this dataset include helpful information such as reference number; easting and northing coordinates; number of vehicles involved; time and date; 1st road class; lighting conditions; weather conditions; casualty classifications(sex & severity); type vehicle - all helping us gain a better understanding not only into why accidents occur but also how they are managed and what makes them more preventable in future situations ahead! Before you delve deeper into this dataset - please read through the guidance document - outlining further details regarding particular categories included in all records!

More Datasets

For more datasets, click here.

Featured Notebooks

• 🚨 Your notebook can be here! 🚨!

How to use the dataset

This dataset contains information regarding road traffic accidents in Leeds from 2009-2018. The data includes the location of each accident, the number of vehicles and people involved, the road surface, weather conditions, and severity of any casualties.

Research Ideas

• Identifying high-risk roads for more concentrated safety/enforcement measures: By analyzing the types of accidents, casualty severity and accident locations in this dataset, governments can identify roads with a higher rate of accidents in order to better allocate resources (such as police patrol and enforcement or increased safety measures) to minimize road fatalities and injuries.
• Predicting which type of vehicles are involved most in accidents: With the data on vehicle type provided in this dataset, researchers can use predictive analytics techniques such as Machine Learning (ML) models to predict which vehicle types have a higher probability of being involved in an accident.
• Examining weather conditions for more effective regulations: With detailed weather data captured from each accident, local governments can define better regulations based on whether the weather at the time was conducive to safe driving or not, especially when it comes to reducing speed limits during extremely humid or foggy days or clear off road obstructions during rainy days

Acknowledgements

If you use this dataset in your research, please credit the original authors. Data Source

License

See the dataset description for more information.

Columns

File: 2016-7.csv | Column name | Description | |:------------------------|:------------------------------------------------------------------| | Reference Number | Unique identifier for each accident. (Integer) | | Number of Vehicles | Number of vehicles involved in the accident. (Integer) | | Expr1 | Unknown. (Integer) | | Time (24hr) | Time of the accident in 24 hour format. (String) | | 1st Road Class | Classification of the road where the accident occurred. (String) | | Road Surface | Type of surface of the road where the accident occurred. (String) | | Lighting Conditions | Lighting conditions at the time of the accident. (String) | | Weather Conditions | Weather conditions at the time of the accident. (String) | | Casualty Class | Classification of the casualty involved in the accident. (String) | | Casualty Severity | Severity of the casualty involved in the accident. (String) | | Sex of Casualty | Gender of the casualty involve...

Annual state-reported motor vehicle registration data for the 50 states and DC reported in Highway Statistics table MV-1.

The data in this dataset comes from the Common Questionnaire for Transport Statistics, developed and surveyed by Eurostat in cooperation between the United Nations Economic Commission for Europe (UNECE) and the International Transport Forum (ITF) at OECD.

The Common Questionnaire is not supported by a legal act, but is based on a gentlemen's agreement with the participating countries; its completeness varies from country to country.

Eurostat’s datasets based on the Common Questionnaire cover annual data for the EU Member States, EFTA states and Candidate countries to the EU. Data for other participating countries are available through the ITF and the UNECE. In total, comparable transport data collected through the Common Questionnaire is available for close to 60 countries worldwide.

The Common Questionnaire collects aggregated annual data on:

• Railway transport
• Road transport
• Inland waterways transport
• Oil pipelines transport
• Road traffic - vehicle-kilometres
• Buses (and coaches)

For each mode of transport, the Common Questionnaire covers some or all of the following sub-modules (the number of questions/variables within each sub-module varies between the different modes of transport):

• Infrastructure (All modes)
• Transport equipment (Rail, Road and Inland Waterways)
• Enterprises, economic performance and employment (All modes)
• Traffic (Rail, Road and Inland Waterways)
• Transport measurement (All modes).

As its name suggests, the theme "Road traffic" focuses on "traffic" only, on road:

• of national vehicle stocks, on national and foreign territories, by type of road; and by vehicle category, fuel type and age. These data should rely on odometer readings;
• of national- and foreign-registered vehicles on national territories. These data should be issued from road censuses.

The theme “Buses and coaches” covers detailed information on road “traffic” (vkm) and “transport measurement” (passengers, passenger-km) performed by buses and coaches.

The data collection on Common Questionnaire was streamlined twice in the recent years:

• In 2013, among other changes, the collection of employment data was terminated;
• In 2019, among other changes, the collection of gas pipeline data was terminated, and the number of “Buses” indicators was reduced.

The Common Questionnaire is completed by the competent national authorities. The responsibility for completing specific modules (e.g. Transport by Rail) or part of modules (e.g. Road Infrastructure) may be delegated to other national authorities in charge of specific fields.

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 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.

Kenya Road Transport: Number of Motor Vehicles: Registered data was reported at 4,973,017.000 Unit in 2024. This records an increase from the previous number of 4,784,156.000 Unit for 2023. Kenya Road Transport: Number of Motor Vehicles: Registered data is updated yearly, averaging 2,210,907.000 Unit from Dec 2004 (Median) to 2024, with 21 observations. The data reached an all-time high of 4,973,017.000 Unit in 2024 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.TA: Road Transport: Number of Motor Vehicles: Registered.

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.

Israel IL: Road Traffic: Thousand Vehicle-km per Road Motor Vehicle data was reported at 16,089.835 Vehicle-km th in 2023. This records a decrease from the previous number of 16,598.463 Vehicle-km th for 2022. Israel IL: Road Traffic: Thousand Vehicle-km per Road Motor Vehicle data is updated yearly, averaging 17,697.554 Vehicle-km th from Dec 2007 (Median) to 2023, with 17 observations. The data reached an all-time high of 19,867.612 Vehicle-km th in 2009 and a record low of 14,964.927 Vehicle-km th in 2020. Israel IL: Road Traffic: Thousand Vehicle-km per Road Motor Vehicle data remains active status in CEIC and is reported by Organisation for Economic Co-operation and Development. The data is categorized under Global Database’s Israel – Table IL.OECD.ITF: Road Traffic and Road Accident Fatalities: OECD Member: Annual. [COVERAGE] ROAD TRAFFIC Road traffic is any movement of a road vehicle on a given road network. When a road vehicle is being carried on another vehicle, only the movement of the carrying (active mode) is considered. VEHICLES A road motor vehicle is a road vehicle fitted with an engine whence it derives its sole means of propulsion, which is normally used for carrying persons or goods or for drawing, on the road, vehicles used for the carriage of persons or goods. [STAT_CONC_DEF] VEHICLES The stock of road motor vehicles is the number of road motor vehicles registered at a given date in a country and licenced to use roads open to public traffic. This includes road vehicles exempted from annual taxes or licence fee; it also includes imported second-hand vehicles and other road vehicles according to national practices. It should not include military vehicles.

Road Classification CountThis data set is sourced from Dundee City Council’s Public Space Camera Surveillance System. It shows a count of road vehicles broken down by vehicle classification type in 8 specified areas across Dundee. The data set shows a snapshot of road vehicles within these areas every Monday, Wednesday and Saturday during the period 1pm-2pm.The vehicle classifications shown on the extract are:• Bus• Car• Motorcycle• Pickup Truck • Truck e.g HGV’s etc• Van This data is experimental and subject to further refinement. Please note that due the nature of CCTV cameras at times data may not be collected as specified above. Therefore, caution should be exercised when analysing data and drawing conclusions from this data set.CCTV datasets contain information on object detections taken from a selection of the CCTV cameras throughout Dundee City. CCTV images are translated into object counts, objects counted include ‘person’, ‘car’, ‘bicycle’, ‘bus’, ‘motorcycle', 'truck, ‘pickup truck 'and ‘van’. The data is generated and owned by Dundee City Council. Copyright © Dundee City Council 2022. This dataset is available for use under the Open Government Licence.Background information about the Dundee CCTV cameras including a map showing the location of the cameras is available on the Dundee City Council website and can be accessed using the following link:https://www.dundeecity.gov.uk/service-area/city-development/sustainable-transport-and-roads/dundees-public-space-camera-surveillance-system

Here are a few use cases for this project:

1. Urban Traffic Management: The "Hong Kong vehicle detection" model can be used in improving the city's traffic management system by identifying the ratio and type of vehicles on the road at any given time. This can aid in congestion prediction and reduction.

2. Public Transport Planning: By analyzing the frequency of taxis, mini buses, trams, and other public transport vehicles, authorities and transportation companies can optimize their service routes and schedules.

3. Smart Parking Solutions: This model can detect and monitor the types of vehicles in parking areas, aiding in the design of parking space allocation according to vehicle classes.

4. Surveillance and Security: The model can be integrated into surveillance systems, helping to identify and track suspicious activities involving vehicles in the city.

5. Autonomous Vehicle Training: The data derived from the model can be used in training autonomous vehicles to detect and recognize different types of vehicles, making self-driving technology safer for urban environments.

Here are a few use cases for this project:

1. Traffic Management Systems: Use the "Vehicle detection" model to regulate traffic flow, identify traffic violations, and recognize different types of vehicles on the road, which could lead to more efficient transit systems and improved road safety.

2. Surveillance and Security: Deploy the model in surveillance systems to detect and track specific types of vehicles, particularly useful for identifying unauthorized or suspicious vehicles in certain areas.

3. Autonomous Vehicles: Implement the model in self-driving cars to enable them to identify and differentiate between various vehicle types for safer navigation and traffic maneuvering.

4. Parking Facility Management: Use the model in parking lots and garages to ascertain the types of parked vehicles, automate the parking process, and efficiently allocate spaces based on vehicle classes.

5. Road Maintenance and Planning: Apply the model for traffic data collection followed by analysis on traffic patterns and transportation planning. It can help identify the predominant type of vehicles on particular routes, assisting in better road maintenance and future infrastructure planning.

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 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.