Quick identification method for binary image connected domain marker
A connected domain labeling and binary image technology, applied in the field of image processing, can solve the problems of unsuitable parallelization, low efficiency, and lack of real-time processing of large images, and achieve the effect of improving overall performance, improving scanning efficiency, and simple method.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0038] Such as figure 1 As shown, a binary image connected domain label fast recognition method, including the following steps:
[0039] S1. Initialization of CPU and GPU:
[0040] The server environment is used to establish a virtual graphics workstation cluster. The virtual graphics workstation is logically combined to form a dynamic virtual machine. The server environment includes a server cluster and a main control server, that is, the server environment includes multiple CUDA-based GPU graphics cards installed on the server cluster. The GPU graphics cards are connected through the SLI interface (Scalable Link Interface, upgradeable connection interface), and the block size of the CUDA architecture platform is 256 or 512; the server cluster is used as a computing node, and the main control server is responsible for scheduling and controlling all tasks, and is responsible for Manage the computing of each node on the cluster and integrate storage resources;
[0041] S2. ...
Embodiment 2
[0054] The difference from Embodiment 1 is that the server environment includes a server computer and a master server.
[0055] S1. The initialization of CPU and GPU is specifically:
[0056] Use the server environment to establish a cluster of virtual graphics workstations. The virtual graphics workstations are logically combined to form a dynamic virtual machine. The server environment includes a server computer and a master server, that is, the server environment includes multiple CUDA-based GPU graphics cards installed on the server computer. Each GPU graphics card is connected through the SLI interface, and the block size of the CUDA architecture platform is 256 or 512; the server computer is used as a computing node, and the main control server is responsible for scheduling and controlling all tasks, and is responsible for managing the calculation of each node and storage resource integration;
[0057] The technical solution of the present invention is used in image re...
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, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com