基于FPGA的图像坏点修正算法的硬件实现

Hardware Implementation of Missing Image Pixel Correction Algorithm Based on FPGA

随着图像技术的不断发展,图像质量的要求将越来越高。目前的图像传感器技术虽然有了很大的提高,但是由于处理过程中各个环节的非理想化等原因,图像传感器总会存在着一些坏点,影响图像质量。现场可编程门阵列(FPGA)是在专用集成电路(ASIC)的基础上发展起来的一种可动态编程的器件,与其他中小规模的集成电路相比,其优点主要在于它有很强的灵活性,电路可配置,设计周期短。文中阐述了一种适合于硬件实现的坏点修正算法,以Altera公司的Cyclone EP1C3T144C8型的FPGA芯片为平台构建图像坏点修正实时处理模块,并成功实现了对640×480×10像素的灰度图像的坏点修正处理,得到了预期的理想效果。

With the rapid development in image technology, the quality of image is more and more critical. Though image sensor technology improves rapidly, there remain some missing pixels due to processing imperfections, which influences the quality of image. Field Programmed Gate Array （FPGA） is a dynamic programmed device, developed on the basis of Application Specific Integrated Circuit （ASIC）. It has the advantages over other small and middle scale integrated circuits of flexibility, configurable circuit and short design cycle. This paper describes a missing pixel correction algorithm which is easy to realize in hardware. A missing pixel correction real-time process module is designed for image sensor on the platform based on Ahera corporation＇s Cyclone EP1C3T144C8 FPGA, and the missing pixel correction process for 640×480 × 10 pixels gray images is successfully implemented with expected desirable results.