One Hour of Commuting, Half an Hour of Traffic Jam, AI Saves the Day
The highway congestion video dataset released by City University of Hong Kong conducts real-time congestion status assessments on highways to help transportation departments relieve traffic pressure and prevent accidents in a timely manner.
Transportation is an important link that has undergone the most obvious changes in the urbanization process.
According to statistics in recent years, my country's total expressway mileage is 142,600 kilometers, ranking first in the world, far ahead of the United States, which ranks second.
In terms of infrastructure, the rapid construction of highway and railway networks has provided more and more convenient options for long-distance and inter-city transportation. In terms of people's livelihood, map navigation and shared travel have brought convenience to the people and expanded their living radius.
The commute is long and the municipal government is also anxious
With economic development and the increase in car ownership, the pressure on public transportation is also increasing year by year.During rush hours in the morning and evening in large cities, serious congestion often occurs, with "a distance of two kilometers taking half an hour to drive."
According to the "Monitoring Report on Commute Time Consumption in Major Cities across the Country" (hereinafter referred to as the "Monitoring Report") released by the China Academy of Urban Planning and Design, in 36 major cities across the country, more than 10 million people are suffering from extreme commutes of more than 60 minutes.In Beijing, Shanghai, Guangzhou and Shenzhen, the average daily traffic congestion time accounts for more than 40% of the total commuting time.
According to data released by China's Ministry of Transport, traffic congestion causes economic losses of 5%-8% of GDP, or about 250 billion yuan per year.
In such a complex road network, it is particularly important to implement all-weather, all-round and three-dimensional monitoring of vehicles.
As long as we monitor and predict traffic flow, and promptly conduct congestion relief, accident prediction and location, we can effectively alleviate congestion pressure.
Instead of getting stuck in traffic, study the congestion dataset
Friends who are interested in working in intelligent transportation can learn about the data sets we recommend today.
Highway Traffic Videos Highway congestion video dataset,It was released by City University of Hong Kong and has a total size of 88 M after decompression. The dataset contains 254 road video clips of the same road section, divided into training set and test set.
The dataset contains 254 road video clips and description files of the same road section.
The description file contains key information such as file name, date, timestamp, driving direction, weather, start frame, total number of frames, congestion rating, notes, etc.
Highway Traffic Videos
Highway Congestion Video Dataset
Publishing Agency:City University of Hong Kong
Quantity included:254 video clips & description files
Data format:avi & mat & txt Data size:60 MB
address:https://orion.hyper.ai/datasets/9342
Among them, weather is divided into: overcast (overcast), clear (clear), rain (rainy);
Congestion ratings are divided into: light (mild congestion), medium (moderate congestion) and heavy (heavy congestion)
In addition to analyzing highway congestion, the data set can also be used in the application of traffic video analysis (Smart Traffic), by identifying vehicle turns, lane information, vehicle queue length, lane space occupancy, license plates, vehicle models, etc., to effectively intervene in actual scenarios such as real-time analysis of road conditions, traffic light control, and smart parking.
The smarter you are, the smoother your journey.
