Crowd Density Heat Map Generation via Perspective Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems struggle to accurately represent crowd density in wide regions, especially when using aerial vehicles, as the size of individuals becomes too small in images taken from high altitudes, making it difficult to extract crowd density information effectively.
Innovation Solution
An image processing apparatus and method that includes an information obtainer, a density map generator, a perspective map generator, and a heat map generator, which split images into patch images, estimate crowd density classes, and generate a heat map displaying crowd density information on a map based on location and distance information, allowing for the representation of crowd density across a wide region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If images are taken from high altitudes using aerial vehicles to cover wide regions, then the coverage area is improved, but the measurement precision of crowd density deteriorates due to small individual sizes in the images
Solution Approach 1:
The patent divides the wide-region image into multiple patch images representing different areas. Each patch image is processed separately by the classifier to estimate crowd density, allowing for more accurate local measurements while maintaining wide area coverage through aggregation of multiple patches
Solution Approach 2:
The patent introduces a height dimension by obtaining altitude information from GPS data and using it to generate a perspective map. This perspective map provides distance information that is used to correct and normalize crowd density measurements across different regions, compensating for the reduced measurement precision at high altitudes
2Ease of operation
If patch images are resized to have the same pixel size to facilitate classification, then the ease of operation is improved, but the manufacturing precision of the density map deteriorates due to loss of spatial information
Solution Approach 1:
The patent introduces a perspective map as an intermediary that contains distance information for each region. This perspective map is used to correct the crowd density estimates after classification, compensating for the spatial information loss that occurred during resizing by applying distance-based normalization
Data Source
AI summary
There are provided an apparatus and method for generating a heat map that displays crowd density with visual properties such as colors. The image processing apparatus includes at least one processor to implement: an information obtainer configured to obtain an image by photographing a region of interest and location information indicating a location from which the image is photographed; a density map generator configured to generate a crowd density map of the image based on crowd density information estimated from the image; a perspective map generator configured to generate a perspective map that provides distance information indicating a real-life distance between two points of the image based on the location information; and a heat map generator configured to generate a heat map that displays the crowd density information on a map of the region of interest based on the distance information.