Image Co-processor Segmentation for High-Speed Manufacturing Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing/inspection systems lack sufficient processing power to handle the demands of next-generation manufacturing systems, leading to increased processing times due to limited processor capabilities and the inability to efficiently offload and accelerate image processing functions at a fine-grained level.
Innovation Solution
A multi-core processor system is introduced, allowing for distributed image processing via a network-connected system with a management processor, general-purpose processor, special-purpose accelerators, and storage units, enabling the reuse of existing infrastructure and offloading image processing workloads to multi-core processors for accelerated image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If a single general-purpose processor is used for image processing, then the system is simple to manage, but the processing power is insufficient to handle next-generation manufacturing inspection demands
Solution Approach 1:
The system segments image processing functions from general-purpose control functions by introducing a dedicated image co-processor. The co-processor contains multiple specialized image processing units that can be independently configured and executed, allowing complex image processing tasks to be handled separately from the main control system, thereby increasing processing power without proportionally increasing overall system complexity.
Solution Approach 2:
An image co-processor is introduced as an intermediary between the general-purpose control processor and the inspection system. This co-processor acts as a mediator that receives image data, executes specialized image processing algorithms, and returns results to the control system, effectively bridging the gap between simple system management and high processing power requirements.
2Productivity
If image processing functions are offloaded to another general-purpose system, then system load is reduced, but no actual image processing acceleration is achieved
Solution Approach 1:
The image co-processor implements local quality by providing specialized image processing units with dedicated hardware accelerators for specific image processing algorithms. Each processing unit is optimized for particular tasks such as edge detection, noise filtering, or pattern recognition, ensuring high processing speed and reliable processing capability for those specific functions rather than using a general-purpose approach.
Solution Approach 2:
The system changes parameters by transitioning from general-purpose software-based image processing to specialized hardware-based processing with configurable processing pipelines. The co-processor allows dynamic configuration of processing parameters and algorithms, enabling optimized processing speed and reliable results through hardware-level parameter adjustments rather than software execution.
3Loss of time
If image processing is performed within a single integrated system, then data flow is simplified, but processing time increases dramatically with larger inspection areas
Solution Approach 1:
The system segments the inspection area and distributes image processing across multiple independent processing units within the co-processor. Each processing unit can handle a portion of the inspection area simultaneously, reducing overall processing time for large inspection areas. The segmented architecture allows parallel processing while maintaining simplified data flow through dedicated data paths.
Solution Approach 2:
The patent transitions from a single-dimension sequential processing model to a multi-dimensional parallel processing architecture. The co-processor implements multiple processing units that operate concurrently on different image data streams, adding a temporal dimension of parallelism to the system. This dimensional change enables processing time reduction without requiring proportionally increased system architecture complexity.
Data Source
AI summary
The present invention relates to machine vision computing environments, and more specifically relates to a system and method for selectively accelerating the execution of image processing applications using a multi-core processor system. To this extent, a multi-core processor system is generally defined as one that is multi-platform, and potentially distributed via a network or other connection. The invention provides a machine vision system and method for executing image processing applications referred to herein as an image co-processor that comprises (among other things) a plurality of multi-core processors (MCPs) that work to process multiple images in an accelerated fashion.