Tile-Based Image Decoding for Multi-Camera Resource Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing systems for capturing and decoding indicia, such as barcodes and QR codes, are inefficient in utilizing processing resources when multiple cameras are used, leading to wasted processing power and high installation costs due to the need for separate processors for each camera, with only one processor typically being utilized for decoding after an indicia is identified.
Innovation Solution
A decoding system that divides captured images into tiles, performs a transform on each tile to derive metadata, and uses a processor to identify clusters of tiles that may contain indicia, requesting a region of interest (ROI) for further analysis and decoding, allowing for more efficient use of processing resources across multiple cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple stationary cameras are used to simultaneously capture images of multiple surfaces, then the coverage and capture capability are improved, but the processing cost and system complexity increase due to requiring separate processors for each camera
Solution Approach 1:
The patent consolidates processing resources by having multiple cameras share common processing units. Instead of dedicating one processor per camera, the system uses a pool of processors that can be dynamically allocated to handle images from multiple cameras, thereby reducing overall system complexity and cost while maintaining multi-camera coverage capability.
Solution Approach 2:
The processing units are designed to be universal and multi-functional, capable of handling images from any of the multiple cameras. This allows the same processing hardware to serve multiple purposes and multiple cameras, eliminating the need for dedicated processors for each camera and reducing system complexity.
2Power
If separate processors are dedicated to each camera, then the processing capability per camera is improved, but the resource utilization efficiency deteriorates because only one processor is typically utilized for decoding after indicia identification
Solution Approach 1:
The system implements dynamic resource allocation where processing units can be dynamically assigned to different cameras based on real-time needs. When one camera captures an indicia requiring decoding, other processors remain available to handle images from other cameras, creating a dynamic balancing act that optimizes both processing capability and resource utilization efficiency.
Solution Approach 2:
The patent ensures continuous useful action by maintaining a pool of ready processors that can immediately begin processing images from other cameras while one processor is occupied with decoding. This eliminates idle time and ensures that processing resources are continuously utilized across multiple cameras, improving overall resource utilization efficiency.
3Productivity
If multiple processors are used to search through images, then the search speed is improved, but the decoding efficiency deteriorates because only one processor can decode the indicia at a time
Solution Approach 1:
The patent segments the processing workflow into distinct phases: image search/identification and indicia decoding. Multiple processors can simultaneously perform the search phase across different camera images, while a dedicated decoding processor handles the decoding phase. This segmentation allows parallel processing during search without compromising decoding efficiency.
Solution Approach 2:
The system performs preliminary actions by having multiple processors simultaneously search through images from multiple cameras to identify indicia before the decoding phase begins. This preliminary parallel search accelerates the identification process, and once indicia are identified, they are passed to a decoding processor that can then focus exclusively on the decoding task without time loss.
Data Source
AI summary
A decoding device includes decoding storage to store metadata of tiles of a captured image of an object, wherein each tile includes a 2D array of pixels, and the metadata is derived by a subsampling circuit from the captured image via transform and describes at least one characteristic of each tile. The decoding device also includes a processor to compare a metadata value of each tile to a first threshold to identify tiles that includes a portion of an indicia; upon identification of a cluster of adjacent tiles that each meet the first threshold and that are numerous enough to meet a second threshold, request a ROI that includes the cluster from a subsampling circuit; analyze the ROI to determine whether it contains the indicia; and decode and transmit data encoded within the indicia to a server.