In Pittsburgh, Autonomous Vehicle (AV) companies have been testing autonomous vehicles since September 2016. However, the tech is new, and there have been some high-profile behavior that we believe warrants a larger conversation. So in early 2017, we set out to design a survey to see both how BikePGH donor-members, and Pittsburgh residents at large, feel about about sharing the road with AVs as a bicyclist and/or as a pedestrian. Our survey asked participants how they feel about being a fellow road user with AVs, either walking or biking. We also wanted to collect stories about people’s experiences interacting with this nascent technology. We are unaware of any public surveys about people’s feelings or understanding of this new technology. We hope that our results will help add to the body of data and help the public and politicians understand the complexity of possible futures that different economic models AV technology can bring to our cities and towncenters.

We conducted our 2017 survey in two parts. First, we launched the survey exclusively to donor-members, yielding 321 responses (out of 2,900) via email. Once we closed the survey, we launched it again, but allowed the general public to take it. Through promoting it on our website, social media channels, and a few news articles, we yielded 798 responses (mostly from people in the Pittsburgh region), for a combined total of 1,119 responses.

Regarding the 2019 survey: In total, 795 people responded. BikePGH solicited responses from their blog, website, and email list. There were also a few local news articles about the survey. While many questions were kept similar to the 2017 survey, BikePGH wanted to dig a bit deeper into regulations as well as demographics this time around.

The 2019 follow up survey also aims to see how the landscape has changed, and how specifically, Pittsburghers on bike and on foot feel about sharing the road with AVs so that we’re all better prepared to deal with this new reality and help make sure that it is introduced as safely as humanly possible.

Context

Automobile data holds immense importance as it offers insights into the functioning and efficiency of the automotive industry. It provides valuable information about car models, specifications, sales trends, consumer demographics, and preferences, which car manufacturers and dealerships can leverage to optimize their operations and enhance customer experiences. By analyzing data on vehicle reliability, fuel efficiency, safety ratings, and resale values, the automotive industry can identify trends and implement strategies to produce more reliable and environmentally friendly vehicles, improve safety standards, and enhance the overall value of cars for consumers. Moreover, regulatory bodies and policymakers rely on this data to enforce regulations, set emissions standards, and make informed decisions regarding automotive policies and environmental impacts. Researchers and analysts use car purchase data to study market trends, assess the environmental impact of various vehicle types, and develop strategies for sustainable growth within the industry. In essence, car purchase data serves as a foundation for informed decision-making, operational efficiency, and the overall advancement of the automotive sector.

Content

This dataset comprises diverse parameters relating to car purchases and ownership on a global scale. The dataset prominently incorporates fields such as 'First Name', 'Last Name', 'Country', 'Car Brand', 'Car Model', 'Car Color', 'Year of Manufacture', and 'Credit Card Type'. These columns collectively provide comprehensive insights into customer demographics, vehicle details, and payment information. Researchers and industry experts can leverage this dataset to analyze trends in car purchasing behavior, optimize the customer car-buying experience, evaluate the popularity of car brands and models, and understand payment preferences within the automotive industry.

Dataset Glossary (Column-wise)

• First Name - The first name of the car purchaser.
• Last Name - The last name of the car purchaser.
• Country - The country of residence of the car purchaser.
• Car Brand - The brand or manufacturer of the purchased car.
• Car Model - The specific model of the purchased car.
• Car Color - The color of the purchased car.
• Year of Manufacture - The year the car was manufactured.
• Credit Card Type - The type of credit card used for the car purchase.

Structure of the Dataset

https://i.imgur.com/olZpXsT.png" alt="">

Acknowledgement

The dataset provided here is a simulated example and was generated using the online platform found at Mockaroo. This web-based tool offers a service that enables the creation of customizable mock datasets that closely resemble real data. It is primarily intended for use by developers, testers, and data experts who require sample data for a range of uses, including testing databases, filling applications with demonstration data, and crafting lifelike illustrations for presentations and tutorials. To explore further details, you can visit their website.

Cover Photo by: Freepik

Thumbnail by: Car icons created by Freepik - Flaticon

This Eurobarometer survey is about people's perception of robots, autonomous cars and drones. With these technologies becoming more mainstream, it is important to understand what people think and to assess the extent to which people will accept robots performing certain functions. It builds on a previous study conducted in 2012, and looks at ways in which attitudes may have changed over the last two years.

While this development is accelerating, most people appear unaware of how much responsibility we have already outsourced to machines and intelligent algorithms. Thus, the first goal of this study would be to determine the level of awareness among European citizens of this phenomenon.

Secondly, we want to determine to what extent people are open to these technological advances. Are they willing to use autonomous vehicles? Do they mind planes flying more or less on their own? To advance with our research in this area and to faciliate uptake, we have to have a better understanding of what know about autonomous systems and how they feel about them.

While it is admittedly difficult for people to really assess how they are going to react to future technologies, answers to these and similar questions can at least provide an indication of people's feelings.

Thirdly, this study would emphasise DG CONNECT's stake in this important, forward-looking topic. Autonomoy will become a key regulatory challenge in the future, and DG CONNECT can become the worldwide frontrunner in this field.

To reach these objectives, the idea is to conduct a Eurobarometer survey in all 28 EU MS.

The results by volumes are distributed as follows:

• Volume A: Countries
• Volume AA: Groups of countries
• Volume A' (AP): Trends
• Volume AA' (AAP): Trends of groups of countries
• Volume B: EU/socio-demographics
• Volume B' (BP) : Trends of EU/ socio-demographics
• Volume C: Country/socio-demographics ---- Researchers may also contact GESIS - Leibniz Institute for the Social Sciences: https://www.gesis.org/eurobarometer

Differential Privacy for Automotive Data Market Outlook

According to our latest research, the global Differential Privacy for Automotive Data market size reached USD 1.92 billion in 2024. The market is exhibiting robust growth, registering a CAGR of 27.4% during the forecast period. By 2033, the market is projected to reach USD 14.6 billion, driven by increasing demand for secure data sharing and privacy-preserving analytics in the automotive sector. The rapid proliferation of connected and autonomous vehicles, alongside stringent regulatory frameworks for data privacy, is propelling market expansion worldwide.

The primary growth driver for the Differential Privacy for Automotive Data market is the exponential rise in data generated by modern vehicles. With the automotive industry embracing digital transformation, vehicles now produce massive volumes of data from sensors, telematics, infotainment systems, and vehicle-to-everything (V2X) communications. This data is invaluable for enhancing user experiences, predictive maintenance, and optimizing fleet operations. However, the sensitive nature of personal and operational data necessitates robust privacy-preserving technologies. Differential privacy solutions enable automotive stakeholders to extract actionable insights from aggregated datasets while ensuring individuals’ identities and personal information remain protected. The adoption of differential privacy is further accelerated by the increasing integration of artificial intelligence and machine learning in automotive applications, which require access to large, high-quality datasets without compromising privacy.

Another significant factor fueling market growth is the evolving regulatory landscape concerning data privacy and protection. Governments across North America, Europe, and Asia Pacific have enacted stringent regulations such as the General Data Protection Regulation (GDPR), California Consumer Privacy Act (CCPA), and China’s Personal Information Protection Law (PIPL). These regulations mandate automotive OEMs, suppliers, and service providers to implement advanced privacy measures when handling consumer and operational data. Differential privacy, recognized for its mathematical guarantees and scalability, is rapidly emerging as the preferred solution for regulatory compliance. Automotive companies are increasingly investing in privacy-enhancing technologies not only to avoid hefty fines but also to build consumer trust and safeguard their brand reputation in a data-driven marketplace.

Furthermore, the surge in connected and autonomous vehicle adoption is catalyzing the demand for differential privacy solutions. Connected vehicles rely on real-time data exchange for navigation, safety, diagnostics, and infotainment, while autonomous vehicles process vast amounts of sensor data to enable self-driving capabilities. As the automotive ecosystem becomes more interconnected, the risk of data breaches and unauthorized access intensifies. Differential privacy offers a proactive approach to mitigate these risks by introducing controlled noise to datasets, thereby preventing the re-identification of individuals or vehicles. This technology not only addresses privacy concerns but also unlocks new opportunities for data monetization, analytics, and cross-industry collaborations without compromising user confidentiality.

From a regional perspective, North America and Europe are at the forefront of adopting differential privacy solutions in the automotive sector, owing to mature regulatory frameworks and high penetration of connected vehicles. The Asia Pacific region is expected to witness the fastest growth, driven by rapid urbanization, expanding automotive production, and increasing government initiatives for smart mobility and data protection. Latin America and the Middle East & Africa are gradually embracing differential privacy as part of their digital transformation agendas, with a focus on enhancing road safety, fleet management, and mobility services. As automotive data ecosystems continue to evolve, regional disparities in technology adoption and regulatory enforcement will shape the competitive landscape and growth trajectory of the global Differential Privacy for Automotive Data market.

Component Analysis

The Differential Privacy for Automotive Data market is segmented by component into Software, Hardware, and Services. Software solutions dominate the market, accounting for the largest sh

Description:

The "Cars" dataset is specifically designed for vehicle detection tasks, focusing on the identification of various types of transport such as cars, trucks, and buses. This dataset is ideal for anyone looking to build and train machine learning models for object detection, particularly in the context of traffic analysis and autonomous driving applications.

Contents:

XML Files: The dataset contains annotated XML files, providing detailed information about the bounding boxes for each detected vehicle within the images. These annotations are in PASCAL VOC format, making them compatible with a wide range of object detection frameworks.

Image Files: While the primary focus is on the XML annotations, this dataset can be complemented with image data for training and testing purposes. The annotations are highly accurate, ensuring that the models trained on this dataset achieve high precision in detecting vehicles.

Applications:

Transport Detection: Perfect for developing models that can detect and classify different types of vehicles on the road, such as cars, buses, and trucks.

Traffic Analysis: Can be used in traffic management systems to monitor vehicle flow and identify potential traffic jams.

Autonomous Driving: A valuable resource for training autonomous vehicles to detect surrounding traffic accurately.

Usage:

This dataset can be easily integrated into popular machine learning libraries and frameworks such as TensorFlow, PyTorch, and FastAI. The XML annotations are straightforward to parse, and the dataset is structured to facilitate quick deployment into your machine learning pipeline.

License:

The dataset is open for use in academic research, commercial projects, and public domain initiatives, with proper attribution to the source.

In Pittsburgh, Autonomous Vehicle (AV) companies have been testing autonomous vehicles since September 2016. However, the tech is new, and there have been some high-profile behavior that we believe warrants a larger conversation. So in early 2017, we set out to design a survey to see both how BikePGH donor-members, and Pittsburgh residents at large, feel about about sharing the road with AVs as a bicyclist and/or as a pedestrian. Our survey asked participants how they feel about being a fellow road user with AVs, either walking or biking. We also wanted to collect stories about people’s experiences interacting with this nascent technology. We are unaware of any public surveys about people’s feelings or understanding of this new technology. We hope that our results will help add to the body of data and help the public and politicians understand the complexity of possible futures that different economic models AV technology can bring to our cities and towncenters. We conducted our survey in two parts. First, we launched the survey exclusively to donor-members, yielding 321 responses (out of 2,900) via email. Once we closed the survey, we launched it again, but allowed the general public to take it. Through promoting it on our website, social media channels, and a few news articles, we yielded 798 responses (mostly from people in the Pittsburgh region), for a combined total of 1,119 responses.

Self-driving cars have the potential to greatly improve public safety. However, their introduction onto public roads must overcome both ethical and technical challenges. To further understand the ethical issues of introducing self-driving cars, we conducted two moral judgement studies investigating potential differences in the moral norms applied to human drivers and self-driving cars. In the experiments, participants made judgements on a series of dilemma situations involving human drivers or self-driving cars. We manipulated which perspective situations were presented from in order to ascertain the effect of perspective on moral judgements. Two main findings were apparent from the results of the experiments. First, human drivers and self-driving cars were largely judged similarly. However, there was a stronger tendency to prefer self-driving cars to act in ways to minimize harm, compared to human drivers. Second, there was an indication that perspective influences judgements in some situations. Specifically, when considering situations from the perspective of a pedestrian, people preferred actions that would endanger car occupants instead of themselves. However, they did not show such a self-preservation tendency when the alternative was to endanger other pedestrians to save themselves. This effect was more prevalent for judgements on human drivers than self-driving cars. Overall, the results extend and agree with previous research, again contradicting existing ethical guidelines for self-driving car decision making and highlighting the difficulties with adapting public opinion to decision making algorithms.

The Bushy Park Reservoir is a relatively shallow impoundment in a semi-tropical climate and is the principal water supply for the 400,000 people of the City of Charleston and the surrounding areas including the industries in the Bushy Park Industrial Complex. Although there is an adequate supply of freshwater in the reservoir, there are taste-and-odor water-quality concerns. The U.S. Geological Survey (USGS) worked in cooperation with the Charleston Water System to study the hydrology and water-quality of the Bushy Park Reservoir to identify factors affecting water-quality conditions. This data release is for the water-quality data collected with an autonomous underwater vehicle (AUV) for a water-quality study of Bushy Park Reservoir. Sixteen water-quality surveys were collected over the period of September 2013 to May 2015. Data includes water temperature, specific conductance, pH, dissolved oxygen, turbidity, chlorophyll, and blue-green algae. The typical water-quality survey lasted around four hours, collected data every second (or 3 feet) for the seven parameters for a total 100,000 data points. This Data Release is for USGS Scientific Investigations Report entitled Characterization of the Water Quality of Bushy Park Reservoir, South Carolina 2013-2015 (in press).

The Internet of Vehicles (IoV) counts for much in advancing intelligent transportation by connecting people, vehicles, infrastructures, and cloud servers (CS). However, the open-access wireless channels within the IoV are susceptible to malicious attacks. Therefore, an authentication key agreement protocol becomes essential to ensure secure vehicular communications and protect vehicle privacy. Nevertheless, although the vehicles in the group are compromised, they can still update the group key and obtain the communication content in the existing group key agreement protocols. Therefore, it is still challenging to guarantee post-compromise forward security (PCFS). Dynamic key rotation is a common approach to realizing PCFS, which brings a heavy computation and communication burden. To address these issues, an efficient and robust continuous group key agreement (ER-CGKA) scheme with PCFS is designed for IoV. The propose-and-commit flow is employed to support asynchronous group key updates. Besides, the computation cost and communication overhead are significantly reduced based on the TreeKEM architecture. Furthermore, we adopt the threshold mechanism to resist the collusion attacks of malicious vehicles, which enhances the ER-CGKA scheme’s robustness. Security analysis indicates that the proposed scheme satisfies all the fundamental security requirements of the IoV and achieves PCFS. The performance evaluation results show that our ER-CGKA scheme demonstrates a reduction in the computation cost of 18.82% (Client) and 33.18% (CS) approximately, and an increase in communication overhead of around 55.57% since pseudonyms are utilized to achieve conditional privacy-preserving. Therefore, our ER-CGKA scheme is secure and practical.

Synthetic Data for Traffic AI Training Market Outlook

According to our latest research, the global synthetic data for traffic AI training market size reached USD 1.38 billion in 2024, driven by the rapid advancements in artificial intelligence and machine learning applications for transportation. The market is currently expanding at a remarkable CAGR of 34.2% and is forecasted to reach USD 16.93 billion by 2033. This robust growth is primarily fueled by the increasing demand for high-quality, diverse, and privacy-compliant datasets to train sophisticated AI models for traffic management, autonomous vehicles, and smart city infrastructure, as per our latest research findings.

The marketÂ’s strong growth trajectory is underpinned by the burgeoning adoption of autonomous vehicles and advanced driver assistance systems (ADAS) across the globe. As automotive manufacturers and technology companies race to develop safer and more reliable self-driving technologies, the need for vast quantities of accurately labeled, diverse, and realistic traffic data has become paramount. Synthetic data generation has emerged as a transformative solution, enabling organizations to create tailored datasets that simulate rare or hazardous traffic scenarios, which are often underrepresented in real-world data. This capability not only accelerates the development and validation of AI models but also significantly reduces the costs and risks associated with traditional data collection methods. Furthermore, synthetic data allows for precise control over variables and environmental conditions, enhancing the robustness and generalizability of AI algorithms deployed in dynamic traffic environments.

Another critical growth factor for the synthetic data for traffic AI training market is the increasing regulatory scrutiny and privacy concerns surrounding the use of real-world data, especially when it involves personally identifiable information (PII) or sensitive sensor data. Stringent data protection regulations such as GDPR in Europe and CCPA in California have compelled organizations to seek alternative data sources that ensure compliance without compromising on data quality. Synthetic data, generated through advanced simulation and generative modeling techniques, offers a privacy-preserving alternative by eliminating direct links to real individuals while maintaining the statistical properties and complexity required for effective AI training. This shift towards privacy-first data strategies is expected to further accelerate the adoption of synthetic data solutions in traffic AI applications, particularly among government agencies, public sector organizations, and research institutions.

The proliferation of smart city initiatives and the growing integration of AI-powered traffic management systems are also contributing to the expansion of the synthetic data for traffic AI training market. Urban centers worldwide are investing heavily in intelligent transportation infrastructure to address congestion, improve road safety, and optimize traffic flow. These systems rely on robust AI models that require diverse and scalable datasets for training and validation. Synthetic data generation enables cities and solution providers to simulate complex urban traffic patterns, pedestrian behaviors, and multimodal transportation scenarios, supporting the development of more adaptive and efficient traffic management algorithms. Additionally, the ability to rapidly generate data for emerging use cases, such as connected vehicle networks and emergency response simulations, positions synthetic data as a critical enabler of next-generation urban mobility solutions.

Synthetic Data for Computer Vision is revolutionizing the way AI models are trained, particularly in the realm of traffic AI applications. By generating synthetic datasets that replicate complex visual environments, developers can enhance the training of computer vision algorithms, which are crucial for interpreting traffic scenes and making real-time decisions. This approach allows for the simulation of diverse scenarios, including various lighting conditions, weather patterns, and rare events, which are often challenging to capture with real-world data. As a result, synthetic data for computer vision is becoming an indispensable tool for improving the accuracy and robustness of AI models used in traffic management and autonomous driving.

Sunny Day Crossroads Dash Cam Video Dataset

Introduction

The "Sunny Day Crossroads Dash Cam Video Dataset" offers a focused glimpse into the dynamic interactions at crossroads under sunny conditions, a critical aspect of urban driving for autonomous vehicles. Captured with high-resolution driving recorders at over 1920 x 1080 pixels and a frame rate surpassing 34 fps, this dataset provides clear and fluid visuals necessary for detailed analysis and AI training. It encompasses bounding boxes and tags for over 10 typical urban object categories, including humans, cars, electric bicycles, vans, trucks, and more. This rich dataset aims to enhance the decision-making algorithms of self-driving cars at busy intersections, where the complexity of vehicle and pedestrian movements is significantly heightened.

If you has interested in the full version of the datasets, featuring 10000 annotated images, please visit our website maadaa.ai and leave a request.

Specification

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F22149246%2Fcb93848661d198d24d2f8be259210f52%2Fsunnyday.jpg?generation=1724340274264117&alt=media" alt="">

About maadaa.ai

Since 2015, maadaa.ai has been dedicated to delivering specialized AI data services. Our key offerings include:

• Data Collection: Comprehensive data gathering tailored to your needs.

• Data Annotation: High-quality annotation services for precise data labeling.

• Off-the-Shelf Datasets: Ready-to-use datasets to accelerate your projects.

• Annotation Platform: Maid-X is our data annotation platform built for efficient data annotation.

We cater to various sectors, including automotive, healthcare, retail, and more, ensuring our clients receive the best data solutions for their AI initiatives.

Here are a few use cases for this project:

1. Autonomous Driving: Utilize this computer vision model for autonomous vehicles to identify elements such as cars, people, crosswalks, or school zone signs. This would allow the vehicle to react accurately and in real-time to its changing environment, thus enhancing the safety level of autonomous driving especially in school zones.

2. Traffic Monitoring Systems: Incorporate the model into traffic surveillance systems to identify and monitor real-time activities in school zones. This could provide up-to-date information on the number of pedestrians, specific types of cars, and adherence to school zone road markings, helping manage and prevent potential accidents.

3. Advanced Driver Assistance Systems (ADAS): This computer vision model could be used in ADAS to alert drivers in real-time about objects detected in school zones to increase pedestrian safety. Alerts could include proximity to crosswalks or presence of children with different color identifications.

4. Urban Planning: The model could be used to analyze traffic and pedestrian behavior in school zones, providing valuable data for urban planners when designing safer streets and infrastructure around school areas.

5. Safety Training Programs: Use this model within virtual reality-based driver's education or safety training programs to mimic real-life driving scenarios. Users can learn how to appropriately react when encountering different objects or persons in a school zone context.

This study seeks to leverage newly available Connected and Automated Vehicle (CAV) data to improve crash investigation procedures and obtain input from stakeholders, specifically law enforcement. In particular, law enforcement use of existing Event Data Recorders (EDRs), which store vehicle kinematics during a crash, is explored. Crash investigations are currently aided by EDRs, but this aid could be expanded to include the information gathered by Automated Driving System (ADS) technologies such as radar, cameras, LIDAR, infrared, and ultrasonic. This detailed data could improve the fidelity of future crash investigations, with potential new information such as driver/operator state, vehicle automation capabilities, location, objects and people in the immediate area, performance and diagnostic data, and environmental factors. Through text mining analysis of CAV and sensor-related literature and interviews with law enforcement, this study contributes by gathering evidence about crash investigations to pinpoint the contributing factors of a crash. Further we explore law enforcement involvement in the design of the current EDR retrieval process and their knowledge about using ADS data. Broadly, the project applies the safe systems approach by suggesting a framework that integrates CAV data in the new crash investigation procedures.

One measure used to analyze roadway reliability is the Planning Time Index (PTI). It is the ratio of the 95th percentile travel time relative to the free-flow (uncongested) travel time. PTI helps in understanding the impacts of nonrecurring congestion from crashes, weather, and special events. It approximates the extent to which a traveler should add extra time to their trip to ensure on-time arrival at their destination. A value of 1.0 indicates a person can expect free-flow speeds along their route. A 2.0 index value indicates a traveler should expect that the trip could be twice as long as free-flow conditions. PTI values from 2.0 to 3.0 indicate moderate unreliability, and ones greater than 3.0 are highly unreliable.

The data comes from aggregated Global Positioning System probe data—anonymized data from mobile apps, connected vehicles, and commercial fleets—provided to the Probe Data Analytics (PDA) Suite by INRIX, a travel data technology company. The PDA Suite was created by a consortium of sponsors, including the Eastern Transportation Coalition and the University of Maryland.

PTI values by region, subregion, and county are grouped either as highway facilities or local roads. Highways are roadway segments classified as interstates, turnpikes, and expressways in the PDA Suite. Local roads are segments classified as U.S. routes, state routes, parkways, frontages, and others. The PDA Suite reports weekday hourly averages by facility type and direction. Average weekday values are reported by facility type and direction, within the following time periods:

• Morning (AM): 6:00 AM–9:59 AM;
• Midday (MD): 10:00 AM–2:59 PM;
• Evening (PM): 3:00 PM–6:59 PM;
• Nighttime (NT): 7:00 PM–5:59 AM; and
• Daily: 12:00 AM–11:59 PM.

Although INRIX data collection precedes years reported in Tracking Progress, early years of reporting are highly variable based on a lack of facility coverage. The years from 2011 onward show higher stability for highway facilities for most counties and for the region. For local facilities, 2014 and beyond is where values seem most stable due to more widespread facility coverage.

Historic data for the federal Transportation Performance Management (TPM) system performance reporting requirements is shown. These are Level of Travel Time Reliability (LOTTR), Level of Truck Travel Time Reliability (TTTR), and Annual Hours of Peak-Hour Excessive Delay (AHPHED). The entire states of Pennsylvania and New Jersey are included for LOTTR and TTTR, so the region’s figures can be compared with statewide data.

LOTTR is used to calculate the percentage of roadway segments that are considered reliable. A road segment with an LOTTR of less than 1.5 is considered reliable. Reliable segment lengths in miles are multiplied by their Annual average daily traffic (AADT) values times the average number of people in a vehicle. Then, this sum is then divided by the exact same product for all road segments, to get the resulting percentage of roadway that is reliable for the geography.

TTTR measures how consistent travel times are for trucks on interstates. This can be helpful with analyzing goods movement along the region’s interstates. TTTR is calculated by dividing the 95th percentile of travel times by the 50th percentile of travel times, using the highest value over the Morning (AM), Midday (MD), Evening (PM), Nighttime (NT), and weekend. Each interstate segment multiplies its length by the travel time ratio, the results are summed and then divided by total Interstate length in the geography to determine the area’s TTTR value.

AHPHED is the average number of hours per year spent by motorists driving in congestion during peak periods. This can be useful for analyzing the impact of congestion from an individual’s perspective, since it analyzes how many hours the average person spends stuck in congestion. The figures used are based on the 2010 urbanized area boundaries in the Census. In 2020, they were renamed to urban areas. There are only Mercer County PHED values from 2021 onward, because they only apply to the second four-year TPM performance period, when the Trenton, NJ Urban Area was required to track metrics and set performance targets. AHPHED per capita is that figure divided by the urban area’s population during that year.

It is also important to measure PTIs along the roads buses travel, to measure how reliable the roads are that commuters travel on. To calculate the agency and division type combination PTIs, for each route, all their segments’ planning times from 7-8 AM, 8-9 AM, 4-5 PM, and 5-6 PM are first summed. Then, those are divided by the sums of those segments' free-flow travel times for those same time periods, to get one PTI per time period for each route. Then, the highest of those four PTIs is taken to get one maximum peak hour PTI per route. Then, for each agency and division type combination, all of their routes’ maximum peak hour PTIs are averaged for each year to get the PTIs. Since all NJ Transit routes in the DVRPC region are part of their Southern Division, NJ Transit only has one agency and division mode combination. SEPTA has two: “City” and “Suburban”. SEPTA splits their bus routes into their urban routes, all within their City Transit Division, and their suburban routes, which are in their Victory and Frontier divisions. The Victory and Frontier divisions have been grouped into their own “Suburban” division type.

Congestion is susceptible to external forces like the economy. A downturn can reduce congestion, but this reflects fewer and shorter trips for households and businesses during lean times and may not represent an improvement. Therefore, it may be useful to correlate these results with the Miles Driven indicator.

Unlock insights into road scenes with our comprehensive Vehicle Image Captioning Dataset. This dataset comprises a diverse collection of images capturing vehicles in various settings. Each image is accompanied by detailed captions generated and verified by humans.

These captions, following a specific question format, describe every object on the road, including vehicle color, windshield presence, door and window status, vehicle type, visible wheels, number plate details, logos or brands, vehicle and people activity, and background description. With a 60-70 word description, this dataset offers rich contextual information for image understanding and captioning tasks.

Optimized for Generative AI, Visual Question Answering, Image Classification, and LMM development, this dataset provides a strong basis for achieving robust model performance.

Features:

• 1000+ high-resolution images captured across diverse Indian road scenes.
• Detailed captions describing each object on the road from left to right.
• Captions include vehicle color, windshield presence, door and window status, vehicle type, visible wheels, number plate details, logos or brands, vehicle and people activity, and background description.
• Images sourced from various cities and regions across India, covering day and night scenarios, varied distances, different backgrounds, viewpoints, and more.
• Ideal for image captioning, object detection, scene understanding, and AI research tasks.

Applications:

• Image captioning and description generation.
• Object detection and recognition.
• Autonomous vehicle navigation and scene understanding.
• Traffic analysis and management.
• Urban planning and infrastructure development.

Dataset with Bounding Boxes: The dataset also includes bounding box annotation for Indian Vehicles in 15+ classes. To access the dataset, please visit: https://www.kaggle.com/datasets/dataclusterlabs/indian-vehicle-dataset

The images in this dataset are exclusively owned by Data Cluster Labs and were not downloaded from the internet. To access a larger portion of the training dataset for research and commercial purposes, a license can be purchased. For more details contact us at sales@datacluster.ai or visit www.datacluster.ai

OBJECTIVE : To describe the main characteristics of victims, roads and vehicles involved in traffic accidents and the risk factors involved in accidents resulting in death. METHODS A non-concurrent cohort study of traffic accidents in Fortaleza, CE, Northeastern Brazil, in the period from January 2004 to December 2008. Data from the Fortaleza Traffic Accidents Information System, the Mortality Information System, the Hospital Information System and the State Traffic Department Driving Licenses and Vehicle database. Deterministic and probabilistic relationship techniques were used to integrate the databases. First, descriptive analysis of data relating to people, roads, vehicles and weather was carried out. In the investigation of risk factors for death by traffic accident, generalized linear models were used. The fit of the model was verified by likelihood ratio and ROC analysis. RESULTS There were 118,830 accidents recorded in the period. The most common types of accidents were crashes/collisions (78.1%), running over pedestrians (11.9%), colliding with a fixed obstacle (3.9%), and with motorcycles (18.1%). Deaths occurred in 1.4% of accidents. The factors that were independently associated with death by traffic accident in the final model were bicycles (OR = 21.2, 95%CI 16.1;27.8), running over pedestrians OR = 5.9 (95%CI 3.7;9.2), collision with a fixed obstacle (OR = 5.7, 95%CI 3.1;10.5) and accidents involving motorcyclists (OR = 3.5, 95%CI 2.6;4.6). The main contributing factors were a single person being involved (OR = 6.6, 95%CI 4.1;10.73), presence of unskilled drivers (OR = 4.1, 95%CI 2.9;5.5) a single vehicle (OR = 3.9, 95%CI 2,3;6,4), male (OR = 2.5, 95%CI 1.9;3.3), traffic on roads under federal jurisdiction (OR = 2.4, 95%CI 1.8;3.7), early morning hours (OR = 2.4, 95%CI 1.8;3.0), and Sundays (OR = 1.7, 95%CI 1.3;2.2), adjusted according to the log-binomial model. CONCLUSIONS Activities promoting the prevention of traffic accidents should primarily focus on accidents involving two-wheeled vehicles that most often involves a single person, unskilled, male, at nighttime, on weekends and on roads where they travel at higher speeds.

Rainy Dash Cam Video Dataset

Introduction

The "Rainy Dash Cam Video Dataset" is specifically developed for autonomous driving systems to accurately function under rainy conditions, which pose significant visibility and surface traction challenges. Captured with driving recorders at resolutions exceeding 1920 x 1080 pixels and a frame rate of more than 30 fps, this dataset focuses on rainy day scenarios in urban settings, including crossroads, avenues, and paths. It features bounding boxes and tags for over 10 common urban categories such as humans, cars, electric bicycles, vans, and trucks, under the variable and often difficult lighting conditions that accompany rainy weather.

If you has interested in the full version of the datasets, featuring 6.4k annotated images, please visit our website maadaa.ai and leave a request.

Specification

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F22149246%2F621c64d3053b5c42c580caa28c2e477b%2Frainyday.jpg?generation=1724340001918949&alt=media" alt="">

About maadaa.ai

Since 2015, maadaa.ai has been dedicated to delivering specialized AI data services. Our key offerings include:

• Data Collection: Comprehensive data gathering tailored to your needs.

• Data Annotation: High-quality annotation services for precise data labeling.

• Off-the-Shelf Datasets: Ready-to-use datasets to accelerate your projects.

• Annotation Platform: Maid-X is our data annotation platform built for efficient data annotation.

We cater to various sectors, including automotive, healthcare, retail, and more, ensuring our clients receive the best data solutions for their AI initiatives.

As autonomous machines, such as automated vehicles (AVs) and robots, become pervasive in society, they will inevitably face moral dilemmas where they must make decisions that risk injuring humans. However, prior research has framed these dilemmas in starkly simple terms, i.e., framing decisions as life and death and neglecting the influence of risk of injury to the involved parties on the outcome. Here, we focus on this gap and present experimental work that systematically studies the effect of risk of injury on the decisions people make in these dilemmas. In four experiments, participants were asked to program their AVs to either save five pedestrians, which we refer to as the utilitarian choice, or save the driver, which we refer to as the nonutilitarian choice. The results indicate that most participants made the utilitarian choice but that this choice was moderated in important ways by perceived risk to the driver and risk to the pedestrians. As a second contribution, we demonstrate the value of formulating AV moral dilemmas in a game-theoretic framework that considers the possible influence of others’ behavior. In the fourth experiment, we show that participants were more (less) likely to make the utilitarian choice, the more utilitarian (nonutilitarian) other drivers behaved; furthermore, unlike the game-theoretic prediction that decision-makers inevitably converge to nonutilitarianism, we found significant evidence of utilitarianism. We discuss theoretical implications for our understanding of human decision-making in moral dilemmas and practical guidelines for the design of autonomous machines that solve these dilemmas while, at the same time, being likely to be adopted in practice.

This dataset captures mobility and network parameters from urban vehicular ad hoc networks (VANETs) with a mix of normal and emergency traffic.

The data includes records at 100 ms intervals from 1500 vehicles (5% emergency) under varying traffic densities. Features cover channel utilization, buffer occupancy, neighbor counts, transmission power, and relay efficiency, along with target variables for packet scheduling and rate selection.

Introduction: Traffic congestion remains a significant challenge in urban environments, and optimizing traffic signals plays a crucial role in easing traffic flow. This dataset is designed to aid researchers and developers working on intelligent traffic management systems. It provides comprehensive data collected from three different sources, each offering unique insights into vehicle detection and traffic patterns. Dataset's collection strategy: Kaggle Data Collection 1.Source: Curated datasets from Kaggle, including well-known vehicle detection collections. 2.Content: Contains images and labels of vehicles such as cars, buses, and bikes. 3.Purpose: Provides standardized data for baseline testing and model comparisons. 4.Format: Images in JPEG format with associated YOLO compatible label files (.txt). Link: https://www.kaggle.com/datasets/tubasiddiqui/toy-cars-annotated-on-yolo-format

Custom Data Collection 1.Source: Synthetic and toy vehicle images created in controlled conditions. 2.Content: Features miniature models of cars, buses, and motorcycles. 3.Purpose: Ensures a controlled environment for initial model training and testing, simulating various lighting and angle conditions. 4.Format: JPEG images with YOLO annotation files. Real-Environment Data (Skardu City) 1.Source: Collected from various locations in Skardu city. 2.Content: Real-world images capturing vehicles in diverse scenarios, including intersections, narrow streets, and busy roads. 3.Purpose: Provides data reflecting real traffic conditions, environmental variations, and vehicle diversity, crucial for training robust models. 4.Format: High-resolution JPEG images with detailed annotation files.

Potential Applications Traffic Signal Optimization: Train machine learning models to adjust traffic signals dynamically based on real-time vehicle detection. Autonomous Vehicle Navigation: Use real-world data to enhance the perception systems of self-driving cars. Traffic Flow Analysis: Analyze congestion patterns and develop predictive models for traffic management. Smart City Initiatives: Develop solutions to improve urban mobility and reduce traffic-related issues. How to Use the Dataset 1.Download: interested user can download the dataset from this platform. 2.Training: Use the YOLO compatible images and labels to train object detection models. 3.Testing and Validation: Validate your models on real-world data to assess performance under varying conditions. Acknowledgments We thank the team involved in data collection across Skardu city and the community contributions from Kaggle. This dataset aims to facilitate advancements in smart traffic systems and support innovative solutions for traffic management. Contribute Feedback and contributions are welcome! Let's collaborate to improve and expand this dataset for future research and practical applications.