Deploy a powerful traffic monitoring model trained on a massive 880-image dataset for comprehensive urban surveillance. This project features a pre-trained computer vision model optimized to detect 10 distinct classes, including cars, buses, emergency vehicles, and pedestrians, providing the scale needed for advanced smart city infrastructure.

Ways to Use Traffic Detection Dataset

1. Traffic Flow Analysis: The dataset can be used in machine learning models to analyze traffic flow in cities. It can identify the type of vehicles on the city roads at different times of the day, helping in planning and traffic management.

2. Vehicle Class Based Toll Collection: Toll booths can use this model to automatically classify and charge vehicles based on their type, enabling a more efficient and automated system.

3. Parking Management System: Parking lot owners can use this model to easily classify vehicles as they enter for better space management. Knowing the vehicle type can help assign it to the most suitable parking spot.

4. Traffic Rule Enforcement: The dataset can be used to create a computer vision model to automatically detect any traffic violations like wrong lane driving by different vehicle types, and notify law enforcement agencies.

5. Smart Ambulance Tracking: The system can help in identifying and tracking ambulances and other emergency vehicles, enabling traffic management systems to provide priority routing during emergencies.

The dataset represents synthetic traffic data for a certain location over a one-year period. It includes information about the traffic volume, weather conditions, and special events that may affect traffic.

Features:

Timestamp: The date and time of the observation.Weather: The weather condition at the time of the observation (e.g., Clear, Cloudy, Rain, Snow).

Events: A binary variable indicating whether there was a special event affecting traffic at the time of the observation (True or False).

Traffic Volume: The volume of traffic at the location at the time of the observation.

The dataset is intended for use in analyzing traffic patterns and trends, as well as for developing and testing models related to traffic prediction and management.

The main aim of this dataset is to enable detection of traffic congestion from surveillance cameras using one-stage object detectors. The dataset contains congested and uncongested traffic scenes with their respective labels. This dataset is collected from different surveillance cameras video footage. To prepare the dataset frames are extracted from video sources and resized to a dimension of 500 x 500 with .jpg image format. To Annotate, the image LabelImg software has used. The format of the label is .txt with the same name as the image. The dataset is mainly prepared for YOLO Models but it can be converted to other models format.

Daily utilization metrics for data.lacity.org and geohub.lacity.org. Updated monthly

Traffic Dataset - 500 Videos

Dataset comprises 500 videos of urban traffic captured by surveillance cameras, providing real-time traffic data enriched with bounding box annotations for vehicles and pedestrians. Designed for traffic monitoring and safety research, the dataset supports tasks like vehicle detection, traffic flow analysis, and accident prediction. By leveraging this dataset, researchers and engineers can advance real-time object detection, traffic surveillance systems… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/real-time-traffic-video-dataset.

Historically, research in traffic and incidents has proceeded along two distinct but intrinsically linked tracks. The traffic domain has focused on enhancing deep learning models to incrementally improve prediction accuracy, while the incident track has predominantly concentrated on isolated studies of incident risks and patterns. For the first time, our XTraffic dataset integrates these two tracks both spatially and temporally across a comprehensive regional scale, encompassing 16,972 traffic nodes for the entire year of 2023. The dataset includes detailed time-series data on traffic flow, lane occupancy, and average vehicle speed, as well as meticulously aligned records of incidents across seven different classes, synchronized with the traffic data. Each node also features extensive physical and policy-level meta-attributes of lanes.

Feature layer containing authoritative traffic count points for Sioux Falls, South Dakota. The traffic counts listed are 24-hour, weekday, two-directional counts. Traffic counts are normally collected during the summer months, but may be taken any season, as weather permits. The traffic counts are factored by the day of the week as well as by the month of the year to become an Average Annual Daily Total (AADT). Traffic volumes (i.e. count data) can fluctuate depending on the month, week, day of collection; the weather, type of road surface, nearby construction, etc. All of the historical data should be averaged to reflect the "normal" traffic count. More specific count data (time, date, hourly volume) can be obtained from the Sioux Falls Engineering Division at 367-8601.

Traffic data from traffic detectors installed on strategic routes / major roads including traffic volume, traffic speed and road occupancy (Raw Data). Traffic speeds from traffic detectors installed on strategic routes / major roads mapped onto the respective road network segments (Processed Data).

This dataset contains an aggregated version of the San Francisco Traffic dataset used by Lai et al. (2017). It contains 862 weekly time series showing the road occupancy rates on the San Francisco Bay area freeways from 2015 to 2016.

This dataset is designed for urban traffic flow prediction and includes temporal, spatial, and categorical features essential for analyzing traffic patterns.

Key Features: Timestamp: Records the exact date and time in 15-minute intervals, enabling the modeling of temporal dependencies. Location: Identifies the traffic sensor locations (e.g., Sensor_01, Sensor_02), capturing spatial variability. Vehicle_Count: Represents the number of vehicles detected by sensors during each interval. Vehicle_Speed: Measures the average speed of vehicles in km/h, indicating traffic conditions. Congestion_Level: An ordinal variable representing traffic congestion on a scale (e.g., 0 for no congestion, 5 for high congestion). Peak_Off_Peak: Categorical data distinguishing between peak and off-peak hours for better contextual analysis. Target_Vehicle_Count: The predicted vehicle count for the subsequent time interval, serving as the target variable for predictive modeling. Data Overview: Rows: 200 Columns: 7 Temporal Coverage: 2 days and 15 minutes intervals, providing high-resolution data for short-term prediction.

The census count of vehicles on city streets is normally reported in the form of Average Daily Traffic (ADT) counts. These counts provide a good estimate for the actual number of vehicles on an average weekday at select street segments. Specific block segments are selected for a count because they are deemed as representative of a larger segment on the same roadway. ADT counts are used by transportation engineers, economists, real estate agents, planners, and others professionals for planning and operational analysis. The frequency for each count varies depending on City staff’s needs for analysis in any given area. This report covers the counts taken in our City during the past 12 years approximately.

Explore our detailed website traffic dataset featuring key metrics like page views, session duration, bounce rate, traffic source, and conversion rates.

This dataset contains the current estimated speed for about 1250 segments covering 300 miles of arterial roads. For a more detailed description, please go to https://tas.chicago.gov, click the About button at the bottom of the page, and then the MAP LAYERS tab. The Chicago Traffic Tracker estimates traffic congestion on Chicago’s arterial streets (nonfreeway streets) in real-time by continuously monitoring and analyzing GPS traces received from Chicago Transit Authority (CTA) buses. Two types of congestion estimates are produced every ten minutes: 1) by Traffic Segments and 2) by Traffic Regions or Zones. Congestion estimate by traffic segments gives the observed speed typically for one-half mile of a street in one direction of traffic. Traffic Segment level congestion is available for about 300 miles of principal arterials. Congestion by Traffic Region gives the average traffic condition for all arterial street segments within a region. A traffic region is comprised of two or three community areas with comparable traffic patterns. 29 regions are created to cover the entire city (except O’Hare airport area). This dataset contains the current estimated speed for about 1250 segments covering 300 miles of arterial roads. There is much volatility in traffic segment speed. However, the congestion estimates for the traffic regions remain consistent for relatively longer period. Most volatility in arterial speed comes from the very nature of the arterials themselves. Due to a myriad of factors, including but not limited to frequent intersections, traffic signals, transit movements, availability of alternative routes, crashes, short length of the segments, etc. speed on individual arterial segments can fluctuate from heavily congested to no congestion and back in a few minutes. The segment speed and traffic region congestion estimates together may give a better understanding of the actual traffic conditions.

Update NotesMar 16 2024, remove spaces in the file and folder names.Mar 31 2024, delete the underscore in the city names with a space (such as San Francisco) in the '02_TransCAD_results' folder to ensure correct data loading by TransCAD (software version: 9.0).Aug 31 2024, add the 'cityname_link_LinkFlows.csv' file in the '02_TransCAD_results' folder to match the link from input data and the link from TransCAD results (LinkFlows) with the same Link_ID.IntroductionThis is a unified and validated traffic dataset for 20 US cities. There are 3 folders for each city.01 Input datathe initial network data obtained from OpenStreetMap (OSM)the visualization of the OSM dataprocessed node / link / od data02 TransCAD results (software version: 9.0)cityname.dbd : geographical network database of the city supported by TransCAD (version 9.0)cityname_link.shp / cityname_node.shp : network data supported by GIS software, which can be imported into TransCAD manually. Then the corresponding '.dbd' file can be generated for TransCAD with a version lower than 9.0od.mtx : OD matrix supported by TransCADLinkFlows.bin / LinkFlows.csv : traffic assignment results by TransCADcityname_link_LinkFlows.csv: the input link attributes with the traffic assignment results by TransCADShortestPath.mtx / ue_travel_time.csv : the traval time (min) between OD pairs by TransCAD03 AequilibraE results (software version: 0.9.3)cityname.shp : shapefile network data of the city support by QGIS or other GIS softwareod_demand.aem : OD matrix supported by AequilibraEnetwork.csv : the network file used for traffic assignment in AequilibraEassignment_result.csv : traffic assignment results by AequilibraEPublicationXu, X., Zheng, Z., Hu, Z. et al. (2024). A unified dataset for the city-scale traffic assignment model in 20 U.S. cities. Sci Data 11, 325. https://doi.org/10.1038/s41597-024-03149-8Usage NotesIf you use this dataset in your research or any other work, please cite both the dataset and paper above.A brief introduction about how to use this dataset can be found in GitHub. More detailed illustration for compiling the traffic dataset on AequilibraE can be referred to GitHub code or Colab code.ContactIf you have any inquiries, please contact Xiaotong Xu (email: kid-a.xu@connect.polyu.hk).

This dataset contains estimates of the average number of vehicles that used roads throughout the City of Detroit. Each record indicates the Annual Average Daily Traffic (AADT) and Commercial Annual Average Daily Traffic (CAADT) for a road segment, where the road segment is located, and other characteristics. This data is derived from Michigan Department of Transportation's (MDOT) Open Data Portal. SEMCOG was the source for speed limits and number of lanes.The primary measure, Annual Average Daily Traffic (AADT), is the estimated mean daily traffic volume for all types of vehicles. Commercial Annual Average Daily Traffic (CAADT) is the estimated mean daily traffic volume for commercial vehicles, a subset of vehicles included in the AADT. The Route ID is an identifier for each road in Detroit (e.g., Woodward Ave). Routes are divided into segments by features such as cross streets, and Location ID's are used to uniquely identify those segments. Along with traffic volume, each record also states the number of lanes, the posted speed limit, and the type of road (e.g., Trunkline or Ramp) based on the Federal Highway Administration (FHWA) functional classification system.According to MDOT's Traffic Monitoring Program a commercial vehicle would be anything Class 4 and up in the FHWA vehicle classification system. This includes vehicles such as buses, semi-trucks, and personal recreational vehicles (i.e., RVs or campers). Methods used to determine traffic volume vary by site, and may rely on continuous monitoring or estimates based on short-term studies. Approaches to vehicle classification similarly vary, depending on the equipment used at a site, and may consider factors such as vehicle weight and length between axles.For more information, please visit MDOT Traffic Monitoring Program.

SDCC Traffic Congestion Saturation Flow Data for January to June 2023. Traffic volumes, traffic saturation, and congestion data for sites across South Dublin County. Used by traffic management to control stage timings on junctions. It is recommended that this dataset is read in conjunction with the ‘Traffic Data Site Names SDCC’ dataset.A detailed description of each column heading can be referenced below;scn: Site Serial numberregion: A group of Nodes that are operated under SCOOT control at the same common cycle time. Normally these will be nodes between which co-ordination is desirable. Some of the nodes may be double cycling at half of the region cycle time.system: SCOOT STC UTC (UTC-MX)locn: Locationssite: Site numbersday: Days of the week Monday to Sunday. Abbreviations; MO,TU,WE,TH,FR,SA,SU.date: Reflects correct actual Date of when data was collected.start_time: NOTE - Please ignore the date displayed in this column. The actual data collection date is correctly displayed in the 'date' column. The date displayed here is the date of when report was run and extracted from the system, but correctly reflects start time of 15 minute intervals. end_time: End time of 15 minute intervals.flow: A representation of demand (flow) for each link built up over several minutes by the SCOOT model. SCOOT has two profiles:(1) Short – Raw data representing the actual values over the previous few minutes(2) Long – A smoothed average of values over a longer periodSCOOT will choose to use the appropriate profile depending on a number of factors.flow_pc: Same as above ref PC SCOOTcong: Congestion is directly measured from the detector. If the detector is placed beyond the normal end of queue in the street it is rarely covered by stationary traffic, except of course when congestion occurs. If any detector shows standing traffic for the whole of an interval this is recorded. The number of intervals of congestion in any cycle is also recorded.The percentage congestion is calculated from:No of congested intervals x 4 x 100 cycle time in seconds.This percentage of congestion is available to view and more importantly for the optimisers to take into account.cong_pc: Same as above ref PC SCOOTdsat: The ratio of the demand flow to the maximum possible discharge flow, i.e. it is the ratio of the demand to the discharge rate (Saturation Occupancy) multiplied by the duration of the effective green time. The Split optimiser will try to minimise the maximum degree of saturation on links approaching the node. .hidden { display: none }

Traffic Volume (24hr count). Data are updated as needed by the Transportation department (typically in the summer), and subsequently copied to VicMap and the Open Data Portal the following day.Traffic speed and volume data are collected at various locations around the city, from different locations each year, using a variety of technologies and manual counting. Counters are placed on streets and at intersections, typically for 24-hour periods. Targeted information is also collected during morning or afternoon peak period travel times and can also be done for several days at a time to capture variability on different days of the week. The City collects data year-round and in all types of weather (except for extreme events like snowstorms). The City also uses data from our agency partners like Victoria Police, the CRD or ICBC. Speed values recorded at each location represent the 85th percentile speed, which means 85% or less traffic travels at that speed. This is standard practice among municipalities to reduce anomalies due to excessively speedy or excessively slow drivers. Values recorded are based on the entire 24-hour period.The Traffic Volume dataset is linear. The lines can be symbolized using arrows and the "Direction" attribute. Where the direction value is "one", use an arrow symbol where the arrow is at the end of the line. Where the direction value is "both", use an arrow symbol where there are arrows at both ends of the line. Use the "Label" field to add labels. The label field indicates the traffic volume at each location, and the year the data was collected. So for example, “2108(05)” means 2108 vehicles were counted in the year 2005 at that location.Data are automatically copied to the Open Data Portal. The "Last Updated" date shown on our Open Data Portal refers to the last time the data schema was modified in the portal, or any changes were made to this description. We update our data through automated scripts which does not trigger the "last updated" date to change. Note: Attributes represent each field in a dataset, and some fields will contain information such as ID numbers. As a result some visualizations on the tabs on our Open Data page will not be relevant.

Description

This dataset contains hourly data on the traffic volume for westbound I-94, a major interstate highway in the US that connects Minneapolis and St Paul, Minnesota. The data was collected by the Minnesota Department of Transportation (MnDOT) from 2012 to 2018 at a station roughly midway between the two cities.

Key Features

• holiday: a categorical variable that indicates whether the date is a US national holiday or a regional holiday (such as the Minnesota State Fair).
• temp: a numeric variable that shows the average temperature in kelvin.
• rain_1h: a numeric variable that shows the amount of rain in mm that occurred in the hour.
• snow_1h: a numeric variable that shows the amount of snow in mm that occurred in the hour.
• clouds_all: a numeric variable that shows the percentage of cloud cover.
• weather_main: a categorical variable that gives a short textual description of the current weather (such as Clear, Clouds, Rain, etc.).
• weather_description: a categorical variable that gives a longer textual description of the current weather (such as light rain, overcast clouds, etc.).
• date_time: a datetime variable that shows the hour of the data collected in local CST time.
• traffic_volume: a numeric variable that shows the hourly I-94 reported westbound traffic volume.

Potential Use Cases

The dataset can be used for regression tasks to predict the traffic volume based on the weather and holiday features. It can also be used for exploratory data analysis to understand the patterns and trends of traffic volume over time and across different conditions.

a packet sniffer software

Location of traffic counters in York. *Please note that the data published within this dataset is a live API link to CYC's GIS server. Any changes made to the master copy of the data will be immediately reflected in the resources of this dataset.The date shown in the "Last Updated" field of each GIS resource reflects when the data was first published.