摘要
准确有效的清晰度评价函数是采用数字图像处理技术实现自动调焦的基础。采用Matlab软件的图像处理工具比较空域和频域上的图像清晰度评价算法;提出一种基于离散小波变换的快速算法(FDWT)的图像清晰度算子,采用Matlab软件对该算法进行仿真,实验结果表明该算法具有较高的可靠性和聚焦灵敏度。同时还与其它已有方法进行了对比,包含基于图像边缘检测的方法:Robert算子、Sobel算子、Prewitt算子、Laplace算子;频域方法:离散余弦变换(DCT)等。结果表明FDWT算法不仅计算速度快,而且模拟出的MTF曲线平滑,具有较好的效果。
Accurate and efficient image evaluation function is the foundation for auto-focusing system based on digital image processing. Based on comparison of image evaluation algorithm on time domain and frequency domain with Matlab Image-processing Toolbox, a focusing evaluation algorithm based on Fast Discrete Wavelet Transform （FDWT） is proposed, and the simulation with Matlab show that the proposed algorithm is quite reliable and sensitive. The comparison of this algorithm with other focusing evaluation algorithms, such as Edge Detection Algorithm （including Robert, Sobel, Prewitt and Laplace） and Discrete Cosine Transform （DCT） is also done, and it shows that FDWT algorithm is high in computing speed, and could evaluate the digital image accurately and effectively.
出处
《通信技术》
2011年第12期7-9,共3页
Communications Technology
