Abstract
The image processing methods of the current big data volume are based on parallel computing to improve the processing speed. There are two ways of mainstream parallel image processing, one is to use DSP or FPGA parallel processing, and the other is to use GPU-based CUDA parallel distributed system. DSP or FPGA parallel image processing mode can realize the complex operation, the fast and low power consumption, but the development is difficult, the developer needs to be familiar with the hardware and software knowledge, write algorithms for different hardware structure, program portability is very poor and the development cycle is long. In GPU-based CUDA parallel processing system, the GPU is responsible for performing highly threaded image parallel processing tasks, the CPU is responsible for logical image processing and serial computing, and the CPU and GPU work together. GPU as a coprocessor, low power consumption, large memory and transmission capacity, currently fully support C and C language, easy to develop and because of the hardware structure fixed algorithm portability is high. The GPU parallel processing technology, with its unique multi-threaded architecture acceleration model, plays an important role in improving the speed of mask defect detection and processing.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
