Programmable Pixel Data Routing for Flexible Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems face inflexibility in hardware implementations, leading to poor performance and high power consumption due to a limited range of filters and operators, while software-based solutions are inefficient with repeated instruction calls.
Innovation Solution
A hybrid pixel data processing apparatus combining a hardware accelerator with a software accelerator, featuring a main controller, central memory buffer, and programmable engine, allowing flexible sequential processing and efficient data routing through a data path unit with matrix Arithmetic Logic Units and multiplexing buses, enabling a range of image processing functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a pure hardware implementation is used, then execution speed is improved, but adaptability deteriorates
Solution Approach 1:
The system dynamically switches between hardware accelerator and software processor based on the type of processing required. The hardware accelerator handles fixed, performance-critical operations while the software processor handles variable, flexibility-required operations, creating a dynamic adaptive system that optimizes both speed and versatility.
Solution Approach 2:
The processing system is segmented into two distinct components: a hardware accelerator for fixed operations and a software processor for flexible operations. This segmentation allows each component to specialize in its strengths, with the hardware providing speed for specific tasks and the software providing adaptability for varying requirements.
2Adaptability or versatility
If a software-based vision accelerator is used, then adaptability is improved, but execution speed deteriorates
Solution Approach 1:
The system dynamically routes processing tasks to either the software processor or hardware accelerator based on the operation type. This dynamic task distribution ensures that flexibility-required operations use software while speed-critical operations use hardware, optimizing both adaptability and execution speed simultaneously.
Solution Approach 2:
The system merges software processing and hardware acceleration into a unified vision processing system. By combining the flexibility of software with the speed of hardware in a single integrated architecture, the system achieves both adaptability and high execution speed that neither approach could achieve alone.
3Adaptability or versatility
If a software-based vision accelerator is used, then adaptability is improved, but power consumption deteriorates
Solution Approach 1:
The system dynamically selects the processing path based on operation requirements, using the energy-efficient hardware accelerator for fixed operations and the flexible software processor only when necessary. This dynamic selection minimizes overall power consumption while maintaining required adaptability.
Solution Approach 2:
The system uses the hardware accelerator to perform copies of frequently executed fixed operations, eliminating the need to repeatedly execute software instruction sequences for the same operations. This reduces both power consumption and execution time for repetitive tasks while maintaining software flexibility for varying operations.
Data Source
AI summary
A pixel data processing apparatus comprises a data path unit comprising a hardware module dedicated to performing, when in use, predetermined functionality in relation to image data. The apparatus also comprises a data store for storing image data and a programmable engine. The programmable engine is arranged to route, when in use, data associated with the image data through the data path unit in a predetermined manner.