Stereoscopic Object Detection via Disparity Field Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face inefficiencies in detecting the presence of objects on target surfaces, such as trays or conveyor belts, leading to disruptions in automated processes due to unexpected occupied receptacles, which can indicate upstream process failures.
Innovation Solution
A machine vision system utilizing stereoscopic imaging with two cameras capturing overlapping views of a target surface, generating a disparity field to compare pixel values and determine the presence of objects by identifying differences in corresponding image elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional single-camera imaging systems are used to detect objects on target surfaces, then the system complexity is low, but the measurement precision and reliability of object presence detection is insufficient
Solution Approach 1:
The patent transitions from single-view 2D imaging to multi-view 3D stereoscopic imaging. By introducing a second camera positioned at a different viewpoint, the system gains depth information and spatial context that enables more accurate object presence detection. The disparity field generated from comparing the two views provides additional dimensional data that improves measurement precision without requiring complex processing of a single complex view.
Solution Approach 2:
The imaging system is segmented into multiple independent camera units, each capturing a view of the target surface. This segmentation allows each camera to be simple and standardized while the combined system achieves enhanced detection capability. The disparity field is then computed by segmenting and comparing corresponding regions across the two camera views, maintaining system modularity while improving precision.
2Reliability
If stereoscopic imaging with multiple cameras is implemented, then the measurement precision and reliability of object presence detection is improved, but the device complexity increases
Solution Approach 1:
The system performs preliminary calibration to establish the geometric relationship between multiple cameras and the target surface before actual detection. This preliminary setup creates a disparity field that maps corresponding points across different camera views, enabling reliable object presence detection during operation. The calibration process prepares the system in advance, reducing the complexity of real-time processing while maintaining high reliability.
Solution Approach 2:
Instead of using a single complex camera system, the patent uses multiple simpler camera units that capture redundant views of the target surface. This copying approach improves reliability through redundancy - if one camera detects an object, the other cameras provide verification. The disparity field acts as a virtual copy that stores the spatial relationship information, allowing reliable detection without requiring complex hardware in each individual camera.
3Measurement precision
If detailed image comparison processing is performed to detect objects, then the measurement precision is improved, but the processing time and productivity are reduced
Solution Approach 1:
The patent extracts and utilizes only the essential information needed for object presence detection - the disparity field that captures depth and spatial relationships. By focusing comparison only on corresponding regions identified through the disparity field rather than processing every pixel in detail, the system achieves high measurement precision while reducing processing time. The disparity field serves as a compressed representation that retains critical detection information.
Solution Approach 2:
The system performs partial comparison by focusing on specific regions where objects are likely to be detected, rather than exhaustively analyzing the entire image. The disparity field guides the comparison to relevant areas, allowing the system to achieve sufficient precision without the excessive processing time that would result from analyzing every possible region in detail. This selective approach balances precision and productivity.
Data Source
AI summary
Described are methods, systems, apparatus, and computer program products for determining the presence of an object on a target surface. A machine vision system includes a first image capture device configured to image a first portion of a target surface from a first viewpoint and a second image capture device configured to image a second portion of the target surface from a second viewpoint. The machine vision system is configured to acquire a first image from the first image capture device, a second image from the second image capture device, rectify the first image and second image, retrieve a disparity field, generate difference data by comparing, based on the mappings of the disparity field, image elements in the first rectified image and a second image elements in the second rectified image; and determine whether the difference data is indicative of an object on the target surface.