Optoelectronic Sensor Detection Using Coarse-Fine Object Linking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D sensors in safety technology face challenges in reliably detecting objects of a minimum size due to image errors like specular reflections and loss of pattern structures, leading to false shutdowns and reduced system availability.
Innovation Solution
An optoelectronic sensor evaluates image data with both fine and coarse detection capabilities, ignoring isolated finely detected objects and combining them with nearby coarsely detected objects to form a single object, thereby enhancing detection capability while maintaining high system availability by intelligently eliminating small erroneous detections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a finer detection capability is used to detect smaller objects, then detection precision is improved, but false detections increase due to image errors like holes in depth maps
Solution Approach 1:
The patent segments the detection process into two distinct stages: coarse detection and fine detection. The coarse detection stage identifies larger objects to establish a baseline, while the fine detection stage detects smaller objects with higher precision. This segmentation allows the system to apply different detection thresholds and processing methods appropriate to each detection level, thereby improving overall detection capability while maintaining system reliability by filtering false detections through the hierarchical approach.
2Reliability
If holes in depth maps are treated as objects to ensure safety, then detection reliability is improved, but false shutdowns increase and system availability decreases
Solution Approach 1:
The patent merges the results of coarse detection and fine detection to form a comprehensive object detection outcome. By combining the broader context from coarse detection with the detailed information from fine detection, the system can distinguish between actual small objects and false detections like depth map holes. This merging approach maintains safety by detecting all potential objects while reducing false shutdowns through cross-validation between the two detection levels.
3Productivity
If a coarse detection capability is used to maintain high system availability, then false shutdowns are reduced, but detection precision decreases and safety margins increase
Solution Approach 1:
The patent implements a dynamic detection system that adapts its processing based on the detection stage. The system dynamically switches between coarse and fine detection modes, adjusting the level of processing and threshold criteria according to the current operational context. This dynamic approach allows the system to maintain high availability during normal operation while enabling high-precision detection when needed, optimizing the balance between productivity and measurement precision.
Data Source
Figure 1
Figure 2~3d
Figure 4~5
AI summary
An optoelectronic sensor (10) for reliably detecting objects (32, 34) of a minimum size in a monitored area (12) with an image sensor (16a-b) for capturing image data of the monitored area (12) and with an evaluation unit (24) specified, which is designed to detect objects (32, 34) in the image data with a detection capability. The evaluation unit (24) is also designed to recognize in the image data both objects (34, 36) that are finely detected with a fine detection capability and objects (32) that are roughly detected with a coarse detection capability that is coarser than the fine detection capability and to ignore those finely detected objects (36) which are not in a vicinity (38) of a coarsely detected object (32).