Image character superposition method based on FPGA

Abstract

The invention discloses an image character superposing method based on an FPGA. The method comprises the following steps: sequentially traverse the starting points of single character areas from a preset pixel position of an image; according to a preset first mapping table, determining whether characters need to be superposed in the character area to which the starting point belongs, and accordingto the preset first mapping table, obtaining character contents needing to be superposed in the character area; acquiring character content actually superposed in the character area according to thecharacter content attribute; searching a second mapping table according to the actually superimposed character content, and obtaining memory address data of superimposed character bitmap information;traversing and reading each piece of pixel data in the superposed character bitmap information according to the memory address data, and writing the pixel data into image pixels. According to the method, real-time and efficient character information superposition of the image can be realized only by using few resources in the FPGA, and the superposition position, the superposition content and thesuperposition character display can flexibly adapt to the character superposition requirement.

Description

technical field [0001] The invention relates to the technical field of FPGA video processing, in particular to an FPGA-based image character overlay method. Background technique [0002] With the wide application of video technology, video character superposition technology has also obtained unprecedented development. Video character overlay is to display character information at a specific position of the video image and provide auxiliary prompt information. [0003] The conventional method of superimposing character information is to use software, based on ARM or CPU, to cache a complete frame of image in the access memory space and then superimpose characters. It takes tens of milliseconds or even longer to process a picture, which improves the processing efficiency. Low, cannot be applied to video character superimposition at higher frame rates, and requires very large access memory space resources to cache complete images; [0004] For the improvement of conventional ...

AI Technical Summary

Problems solved by technology

[0003] The conventional method of superimposing character information is to use software, based on ARM or CPU, to cache a complete frame of image in the access memory space and then superimpose characters. It takes tens of milliseconds or even longer to process a picture, which improves the processing efficiency. Low, cannot be applied to video character superimposition at higher frame rates, and requires very large access memory space resources to cache complete images;

[0004] For the improvement of conventional overlay character method, adopt FPGA to carry out image character overlay, for example adopt the register memory of FPGA to store the superimposed character, each character allocates independent register memory, causes to take up a large amount of FPGA memory resources, and overlay position, overlay content and The superimposed character display method cannot flexibly adapt to different superimposed requirements

Images