论文部分内容阅读
选取高频分量作为清晰度判据是自动聚焦的常用手段。在分析了镜头系统造成图像模糊的原理的基础上,对基于DCT(discretecosinetransform)变换的清晰度判据进行了讨论,把全局DCT算法的改进问题转化成窗口尺度大小选取的问题,进而分析了窗口的大小对于结果的影响。对于实际的自动聚焦系统,提出了窗口尺度选取的原则。该清晰度判据应用在作者开发的高清晰度图像采集系统中。实验结果证明,该文提出的判据可以用于实际聚焦系统。
The selection of high-frequency components as a criterion of sharpness is a common means of auto-focusing. Based on the analysis of the principle of image blur caused by the lens system, the criterion of sharpness based on discrete cosine transform (DCT) is discussed. The problem of improving the global DCT algorithm is transformed into the selection of the window size, and then the window The effect of size on the result. For the actual auto-focus system, the principle of window scale selection is proposed. The sharpness criterion is applied in the high definition image acquisition system developed by the author. Experimental results show that the criterion proposed in this paper can be used in practical focusing system.