Laser Projection Safety Using Depth View Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projector safety solutions, such as face recognition and sensor-based systems, are prone to false alarms and increased costs, and are not robust enough to distinguish between real and projected human faces, or are interfered with by other devices, posing risks from directed laser beams, especially in handheld pico-projectors.
Innovation Solution
A method and subsystem that uses stereoscopic cameras to compute a depth view and apply body feature recognition algorithms to safely project images by detecting human presence and adjusting the projector accordingly, preventing laser beams from being directed at humans, without significantly increasing manufacturing costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If face recognition algorithms are used to detect human presence, then safety monitoring is enabled, but false alarms occur when projected images contain human faces
Solution Approach 1:
The patent introduces depth information as a new dimension to distinguish real human faces from projected images. By computing a depth view from stereoscopic camera images and comparing it with the captured image, the system can identify objects at different depths, thereby eliminating false alarms caused by projected human faces while maintaining accurate safety monitoring.
2Reliability
If sensor-based safety systems are implemented, then laser beam safety is improved, but additional costly components and interference from other devices are introduced
Solution Approach 1:
The patent makes the existing camera serve multiple functions: it is used both for capturing the projected image content and for detecting human presence in the projection area. By computing depth information from the same camera images, the system eliminates the need for separate sensors, reducing component cost and complexity while maintaining laser beam safety.
3Measurement precision
If stereoscopic cameras and depth computation are used, then accurate human detection is achieved, but device complexity increases
Solution Approach 1:
The patent creates a depth view as a computational copy of the scene information extracted from stereoscopic images. This depth map is then overlaid with the captured image to identify objects at different depths. By using this copied depth information rather than direct sensor measurements, the system achieves accurate human detection while leveraging existing camera hardware.
Data Source
AI summary
A projection subsystem includes a projector for projecting an image on a projection surface and a safety feature for tuning the projector when presence of a human is detected in front of the projection surface. This safety feature includes: —an apparatus for computing a depth view corresponding to at least a portion of the projection surface, —a camera for acquiring a captured image, —a computing circuit for detecting an object between the projector and the projection surface from this depth view and determining that the object is a human from the captured image, and a control circuit for tuning the projector accordingly.