Hardware implementation method of related parallel computation of groups in image recognition
A parallel computing and hardware implementation technology, applied in computer parts, computing, character and pattern recognition, etc., can solve problems such as time-consuming, and achieve the effect of improving computing speed, reducing requirements, and improving parallelism.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0023] Hereinafter, the present invention will be described in detail with reference to the accompanying drawings and the embodiments.
[0024] The purpose of using hardware to implement group-related calculations is to speed up, and the method of speeding up calculations is parallel computing.
[0025] The present invention uses all template windows in the group to perform calculations with one window of the real-time graph at the same time to achieve speed increase, so if n windows are used for calculation at the same time, the speed is increased by n times.
[0026] One condition for using n arithmetic units to perform calculations at the same time is to have n channels to read out the template data. When n is large, the hardware is difficult to implement. For example, n=49, the data width is 16, according to the current technology level, this is impossible to achieve. This is the difficulty of group-related parallel computing.
[0027] However, from figure 2 It can be seen that...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap