Load Carrier Identification via Feature Vector Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying load carriers in industrial environments, such as warehouses, are inefficient as they require direct targeting with barcode readers, limiting the ability to identify carriers from a distance or with varying orientations.
Innovation Solution
A device equipped with an image capture system and neural networks that processes images to determine feature vectors and identities of load carriers, allowing for identification using optical codes like barcodes or matrix codes, even from a distance and with varying poses, and autonomously determining movement routes for industrial trucks or robots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If barcode readers are used to capture identification codes, then load carriers can be identified, but the devices must be aimed specifically at the identification code from a short distance away
Solution Approach 1:
The patent replaces the mechanical aiming and positioning system of traditional barcode readers with an optical image capture device (camera) that can identify load carriers from a distance. The system uses image processing algorithms to automatically detect and recognize identification codes without requiring precise manual aiming, thus substituting mechanical positioning with optical detection and computational processing.
Solution Approach 2:
The patent introduces an image processing system as an intermediary between the image capture device and the identification code recognition. This intermediary processes the captured images to extract and recognize identification codes, bridging the gap between distant optical capture and precise code identification, thereby eliminating the need for direct targeting.
2Measurement precision
If barcode readers require direct targeting from short distance, then identification can be achieved, but the ability to identify carriers from a distance or with varying orientations is limited
Solution Approach 1:
The patent transitions from the limited one-dimensional line-of-sight requirement of traditional barcode readers to a two-dimensional image capture approach. The image capture device can detect identification codes from various distances and angles by capturing two-dimensional images, which are then processed to extract code information regardless of the load carrier's orientation or distance from the device.
Solution Approach 2:
The patent changes the operational parameters of the identification system by using image capture devices that can operate at various distances and angles rather than requiring fixed short-distance positioning. The system processes images with varying parameters (distance, angle, orientation) to maintain identification accuracy across different operational conditions.
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
The invention relates to a device (110, 150) for identifying a charge carrier (120) from a plurality of charge carriers in an industrial environment (100), in particular a warehouse. Each charge carrier (120) is provided with an optical identification code (121) which assigns an identity to the respective charge carrier (120). The device (110, 150) comprises a communication interface (113, 153) configured to receive an image (201) of the charge carrier (120) from an image acquisition device (117). The device (110, 150) further comprises a processor (111, 151) configured to determine a feature vector of the charge carrier (120) based on the image of the charge carrier (120). The device (110, 150) further comprises a memory (115, 155) which is configured to provide a reference feature vector for each of the plurality of charge carriers.The processor (111, 151) is further equipped to determine the identity of the charge carrier (120) on the basis of a comparison between the feature vector and the multitude of reference feature vectors.