Laser Point Detection Without Image Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting a laser point on a projection screen suffer from poor real-time performance due to the need for image fusion across multiple cameras, which slows down the process of acquiring the location of the laser point.
Innovation Solution
A method and apparatus that allow for direct laser point detection on individual image frames captured by cameras, without the need for initial image fusion, by determining the presence and location of the laser point within specific detection areas or using downsampling to improve processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If image fusion is performed on overlapping areas captured by multiple cameras before laser point detection, then the coverage area of the screen is increased, but the real-time performance of laser point location acquisition deteriorates
Solution Approach 1:
The patent divides the screen into multiple screen areas, with each camera responsible for capturing images in a specific screen area. Laser point detection is performed independently on each camera's image frame, avoiding the need for image fusion across the entire screen. This segmentation approach maintains comprehensive screen coverage while significantly reducing processing time by eliminating the image fusion step.
2Area of stationary object
If image fusion is performed on overlapping areas captured by multiple cameras, then the complete screen coverage is achieved, but the processing complexity and time consumption increase
Solution Approach 1:
The patent segments the screen into multiple independent detection areas corresponding to each camera's field of view. Each camera independently detects laser points in its own image frame without requiring fusion with other cameras' images. This eliminates the complex image fusion process while maintaining complete screen coverage through coordinated detection across multiple cameras.
Solution Approach 2:
The patent extracts and removes the image fusion step from the traditional processing pipeline. By detecting laser points independently in each camera's image frame and directly obtaining location information, the complex image fusion operation is eliminated entirely, reducing processing complexity while maintaining functionality.
3Measurement precision
If laser point detection is performed on the entire frame after image fusion, then accurate location detection is achieved, but the processing time becomes extremely long
Solution Approach 1:
The patent divides the detection process into multiple parallel segments, with each camera independently detecting laser points in its own image frame. This segmentation allows simultaneous processing of multiple image frames rather than sequential processing of a single fused image, significantly reducing detection time while maintaining location accuracy through direct detection in each camera's high-resolution image.
Data Source
AI summary
A method for detecting a location of a laser point on a screen includes acquiring a first image frame captured by a first camera among N cameras; detecting whether a first laser point exists in the first image frame; and when the first laser point exists in the first image frame, and no laser point exists in an image frame captured by another camera except the first camera among the N cameras, using the first laser point as the laser point on the screen, and acquiring information about a location of the first laser point on the screen. In the present application, images captured by all cameras do not need to be fused first; instead, laser point detection is performed directly on a captured image frame. This greatly improves efficiency of the laser point detection and improves detection accuracy.