Trailer Object Detection Using 3D Point Cloud Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack an automated solution to detect and prevent packages or objects from being left in trailers during unloading, due to inconsistent 3D data from signal noise and the presence of trailer walls, floors, and ceilings, which incurs high costs for shipping companies.
Innovation Solution
A trailer monitoring unit (TMU) captures 3D images, segments them to remove data points representing walls, floors, and ceilings, and analyzes remaining points to determine the presence or absence of objects by segmenting the data into bins and comparing values to a threshold, providing communication for object presence or absence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If 3D imaging is used to detect objects in trailers, then automated detection capability is improved, but data consistency deteriorates due to signal noise and trailer structure interference
Solution Approach 1:
The patent segments the 3D point cloud data into distinct regions representing trailer walls, floors, ceilings, and potential objects. By dividing the complex scene into manageable segments, the system can process and analyze each region separately, improving both automation and measurement precision despite the challenging environment.
Solution Approach 2:
The patent extracts and removes points corresponding to trailer structural elements (walls, floors, ceilings) from the 3D data set. This extraction process isolates the relevant object detection data from the overwhelming background noise of the trailer structure, thereby maintaining data consistency while enabling automated detection.
2Measurement precision
If traditional manual checking methods are used, then detection accuracy is maintained, but productivity deteriorates due to lack of automated detection
Solution Approach 1:
The patent replaces manual visual inspection with an automated 3D imaging and point cloud analysis system. This substitution maintains detection accuracy through sophisticated algorithms while dramatically improving productivity by enabling continuous, rapid scanning without human intervention during the unloading process.
Solution Approach 2:
The patent creates a digital 3D copy of the trailer interior environment through point cloud data. This virtual representation allows automated analysis and object detection without requiring physical manual inspection, thereby maintaining accuracy while enhancing unloading efficiency.
Data Source
AI summary
A system and method for detecting a presence or absence of objects in a trailer are described. A 3D depth-camera is oriented to capture a 3D image. The 3D image includes a plurality of 3D point data defining a portion of a wall, floor, and top of a trailer. The plurality of 3D point data is then analyzed to determine a first, second, and third sub-plurality of points, associated with the portion of the wall, floor, and top, respectively. The first, second, and third sub-pluralities are then removed from the plurality of points to obtain a modified plurality of points, representing a modified 3D image. The modified 3D image is then segmented into a plurality of bins, and the plurality of bins are analyzed to determine one or more points-bin values. A communication is then provided based on whether any of the points-bin values exceeds a threshold value.