数字信号处理器(DSP)专门针对信号处理而设计, 以其独特的特点在信号处理、图形图像处理、通信、交通、航空航天等领域得到越来越广泛的应用。在图形图像处理领域中, 图像去雾一直是研究热点, 开发时多使用Matlab和C语言仿真, 将其直接移植到DSP中运行效率通常较低, 难以做到实时处理。针对该问题, 提出一种快速有效的图像去雾算法, 优化算法过程, 获得更好的处理结果; 利用TI公司的多核DSP和开发工具, 对C代码进行实时优化和并行处理, 取得很好的优化效果, 能够对低分辨率进行图像去雾的实时处理。

The Digital Signal Processor (DSP) is specially designed for signal processing.In recent years,the DSP has found wide application in signal processing,image processing,communication,traffic and aerospace etc.As a practical part of image processing,the image haze removal has been researched for decades.During the research and development of haze removal algorithm,the Matlab and C programming language were usually used,and were supported by DSP.Yet,the efficiency was too low by simply running the source code.To solve this problem,a novel algorithm is proposed for optimizing the algorithm and improving the processing effect.The multi-core DSP and development tool of TI Company are used for real-time optimization of C code and parallel processing.A fine optimization effect is obtained,which shows that the method can be used for processing hazy images with low resolution ratio in realtime.