Elevator Car Object Detection Using 3D Reflection Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Elevator sensors face challenges in accurately detecting objects due to reflective surfaces causing reflections, which obstruct clear line of sight and hinder precise object analysis, especially when large objects block smaller ones.
Innovation Solution
A method and apparatus that utilize sensor data from an elevator car with reflective surfaces to separate reflection objects from real objects based on three-dimensional locations and elevator dimensions, forming combined objects for accurate analysis of interior and exterior spaces, including passenger count and type, door status, and lobby objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are arranged in the elevator car to detect objects, then object detection capability is provided, but reflective surfaces cause reflections that reduce detection accuracy
Solution Approach 1:
The patent transitions from two-dimensional image data to three-dimensional spatial information by calculating real-world coordinates of detected objects. This dimensional transformation allows the system to distinguish between actual objects and their reflections by comparing their spatial positions relative to the elevator car boundaries, thereby resolving the ambiguity caused by reflective surfaces.
Solution Approach 2:
The patent introduces an intermediary processing layer that calculates real-world coordinates as an intermediate representation between raw sensor data and final object identification. This intermediary coordinate system serves as a reference frame to differentiate real objects from reflections, mediating the harmful effect of surface reflections on detection accuracy.
2Area of stationary object
If a camera provides wide angle view of the elevator car, then coverage area is increased, but line of sight to objects may be blocked by large objects
Solution Approach 1:
By calculating three-dimensional real-world coordinates from two-dimensional image data, the system recovers depth information that was lost in the projection process. This allows determination of whether objects are occluding others by comparing their spatial positions and depths, compensating for the line-of-sight limitations of wide-angle camera views.
3Measurement precision
If multiple sensors are used to improve detection accuracy, then object detection precision increases, but device complexity increases
Solution Approach 1:
The patent replaces the need for multiple physical sensors with a computational approach that processes data from existing sensors. By using coordinate transformation algorithms and spatial reasoning, the system achieves enhanced detection precision without proportionally increasing hardware complexity, substituting mechanical/sensor complexity with information processing complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances object detection accuracy by distinguishing reflections from real objects, enabling precise determination of passenger numbers, types, door status, and lobby information, improving elevator system functionality.
Implementation Method 1
one or more of the surfaces inside the elevator car may be reflective, thus causing reflections in the data recorded by the sensors
Data Source
AI summary
According to an aspect, there is provided a solution in which sensor data relating to both real objects sensed by at least one sensor associated with an elevator car and reflection objects of the real objects may be used in determining at least one characteristic associated with an interior space and/or exterior space of the elevator car.