Traffic Flow Estimation Using Camera Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in estimating traffic flow across all lanes due to limited vehicle data collection, as not all automobiles can upload data to cloud environments, and data collected may be unique to each manufacturer, leading to incomplete traffic flow estimation.
Innovation Solution
A traffic flow estimation apparatus that identifies the lane of a second moving object by comparing images, determining its position change over time, and using regression models to estimate traffic flow based on data from a first moving object, allowing for interpolation and accurate estimation of traffic flow across a wider range even when data from adjacent lanes is not available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vehicle data is collected from multiple connected cars to estimate traffic flow, then traffic flow estimation accuracy is improved, but data collection coverage deteriorates because not all automobiles can upload data
Solution Approach 1:
The patent uses image data captured by a camera as an intermediary to indirectly observe and estimate the traffic flow of vehicles in adjacent lanes. Instead of directly collecting data from all vehicles, the system captures images of vehicles in target lanes and uses image processing to estimate traffic flow, thereby bridging the gap between limited direct data collection and comprehensive traffic flow estimation across multiple lanes.
2Loss of information
If vehicle data from all lanes is collected to estimate traffic flow in each lane, then estimation completeness is improved, but system complexity deteriorates due to data management requirements
Solution Approach 1:
The patent extracts only the necessary information from image data - specifically, the traffic flow characteristics of vehicles in target lanes. Instead of collecting and managing all vehicle data from every lane, the system selectively extracts relevant visual information from captured images, processes it to determine traffic flow, and uses this extracted information to estimate conditions in adjacent lanes, thereby reducing data management complexity while maintaining estimation completeness.
3Measurement precision
If traffic flow is estimated for each lane using direct vehicle data, then estimation accuracy is improved, but data availability deteriorates when vehicles in adjacent lanes cannot upload data
Solution Approach 1:
The patent creates a visual copy of the traffic scene by capturing images of vehicles in target lanes. This image copy contains sufficient information to estimate traffic flow characteristics without requiring direct data transmission from vehicles in adjacent lanes. The system processes this visual copy to extract traffic flow information, effectively copying the necessary data from the visual domain rather than relying on electronic data transmission from all vehicles.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is a traffic flow estimation device (1) that estimates a traffic flow of a target area. The traffic flow estimation device (1) includes a storage unit (30) storing lane information of a target area whose traffic flow is to be estimated, and acquires a plurality of images and position information and speed information, the plurality of images being captured at a plurality of different timings by a first moving object (6) that is in motion in the target area and including a second moving object (7) around the first moving object, the position information and speed information being of the first moving object whose images are captured at the timings; identifies a lane in which the first moving object is in motion; calculates, based on the images, a change over time of positions of the second moving object in the images; estimates a lane in which the second moving object is in motion based on a positional relationship of the second moving object with respect to the first moving object detected in the images; and estimates a traffic flow for each lane in the target area based on the speed information of the first moving object, the information indicative of a change over time of positions of the second moving object, and the lanes in which the first moving object and the second moving object are in motion.