Occupancy Detection via 3D Digital Model and Binary Image Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current solutions for detecting the occupancy state of locations in monitored zones, such as parking spaces, are costly, require extensive sensor deployment, and often necessitate electrical wiring or destructive integration, while intelligent video solutions require high camera placement for a bird's eye view.

Innovation Solution

A method and device using an image sensor to obtain images, create a three-dimensional digital model of the monitored area, transform the images into binary format, count pixels associated with occupancy states, and compare these counts to predetermined thresholds to determine occupancy, allowing for efficient detection without sensors per location and adaptable to various configurations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If optical, magnetic or ultrasonic sensors are positioned at each parking space to detect occupancy, then detection reliability is improved, but device complexity and cost increase

Engineering Contradiction:
Improveoccupancy detection reliabilityVSAvoidsensor deployment complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple sensor functions into a single image processing system. Instead of deploying separate optical, magnetic, or ultrasonic sensors at each parking space, the invention uses a single image sensor combined with image processing algorithms to detect occupancy status, thereby reducing device complexity while maintaining detection capability

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a digital representation (binary image) of the physical parking space occupancy state. By capturing images and transforming them into binary representations where pixels indicate occupied or free spaces, the system copies the physical state into a processable format without requiring physical sensors at each location

Inventive Principle:
Principle #26Copying

2Measurement precision

If sensors are integrated into the asphalt of parking spaces, then detection precision is improved, but ease of manufacture deteriorates due to destructive integration

Engineering Contradiction:
Improveoccupancy detection precisionVSAvoidsensor installation ease
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent extracts the detection function from the parking space infrastructure itself. Instead of embedding sensors into the asphalt, the system uses an external image sensor that captures images of the parking spaces. This extraction eliminates the need for destructive integration while maintaining the ability to detect occupancy status through image analysis

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If cameras are positioned very high for a bird's eye view to count parking spaces, then productivity is improved, but device complexity increases

Engineering Contradiction:
Improveparking space counting efficiencyVSAvoidcamera installation complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent transforms the two-dimensional image data into a three-dimensional understanding of parking space occupancy. By using a digital model with depth information and processing images from various angles, the system achieves accurate counting and occupancy detection without requiring cameras to be positioned at very high altitudes, thus reducing installation complexity while maintaining productivity

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3038014B1Device and method for detecting an occupancy state of a location
Publication Date: 2018.02.07 ORANGE SA
  • EP3038014B1 patent drawingFigure 1
  • EP3038014B1 patent drawingFigure 2~4
  • EP3038014B1 patent drawingFigure 5A~5B

AI summary

The application process comprises, for at least one location situated in a monitored area (Z,SURF) by an image sensor (5): - a step of obtaining an image of the monitored area (Z) acquired by the image sensor (5); - a step of obtaining a three-dimensional digital model adapted to the image and representing delimited locations in the monitored area; - a step of transforming the image into a binary image comprising pixels of a first color associated with a first occupancy state and pixels of a second color associated with a second occupancy state; - a step of counting on the binary image, for at least one location represented by the digital model, a number of pixels belonging to at least a part of a volume associated with this location by the digital model and having a color selected from the first and second colors;and - a step of detecting an occupancy state of said at least one location by comparing the number of pixels counted for that location with a predetermined threshold.