【作者】 蒋凤林；

【导师】 卢迪；

【作者基本信息】 哈尔滨理工大学 ， 信号与信息处理， 2008， 硕士

【摘要】 自动调焦技术是计算机视觉、成像系统和各种精密仪器中的关键技术之一。随着数字成像器件的实用化,近十年来人们对自动调焦技术的研究集中在数字图像处理的方法上,并取得了大量成果。基于图像技术的自动调焦方法采用了与传统调焦技术完全不同的方式进行调焦,传统的调焦方法是通过传感器检测检点或者测量距离的方式实现的,而基于图像技术的调焦方法直接根据图像分析出图像的质量,从而获得当前的成像状态,然后完成调焦操作。图像质量分析是该调焦方法中关键的技术。本文首先对数字成像系统的自动调焦的原理和研究现状进行了简单的介绍,在此基础上,详细介绍了图像质量评价的过去与现状,然后,本文从两种途径详细论述了图像质量分析方法的实现:(1)基于灰度子图的图像质量评价,提出了一种基于灰度区域搜索、对比的评价方法。将图像视为由具有不同灰度的区域组成的画面,寻找原始图像中每个灰度子图的失真,比较每个灰度子图在原始图像与退化图像中的变化,进而评价整幅图像的质量的退化。(2)基于小波分析的图像质量评价,以小波分析的方法对图像进行多分辨率分解,对各小波分量做统计处理,分析了图像质量变化体现在小波能量谱上的变化。通过分析小波系数的变化,可以确定图像质量的优劣。调焦的实现对于对焦深度法是计算每次成像的调焦函数值,结合一维搜索方法、Fibonacci搜索法、函数逼近法或爬山法,不断逼近正焦位置;对于离焦深度法是根据评价出的图像质量和已知的系统参数,确切地改变焦距调整量。在自动调焦理论的基础上提出了一种采用DSP(Digital SignalProcessor)+FPGA(Field Programmable Gate Array)的自动调焦系统的架构,并分析了系统的总体性能。更多还原

【Abstract】 Auto-focus is one of the key problems in computer vision, imaging systems and various precision instruments. In the last ten years, research on auto-focusing techniques has been focused on methods related to digital image processing on account of the maturing of digital imaging devices, and a lot of achievements have been accomplished. The auto-focus means based on image technology is different from those traditional auto-focus means entirely. The traditional ones must depend on some special assistant equipment. They use these equipments to measure distance or find focus and realize auto-focus. However, the auto-focus based on image technology is completed by image quality analysis. It can judge the imaging state from the current image quality, and then the system can adjust focus distance correctly supervised by imaging state. The image quality analysis is the key technology.First, the dissertation briefs the principles and research status of auto-focus in digital imaging systems. Then, the dissertation discussed the past and present state of image quality evaluation. Following that, two image quality analysis methods are presented in detail in the thesis:(1) One is the assessment of image quality through gray scale area. It describes a new method which evaluates quality of image through searching and contrasting gray scale area. In this method, image was treated as being composed of many different gray scale areas. Distortion of each gray scale area was computed. Then the image quality was assessed and the simulation results are presented. And it indicates that this algorithm works effectively.(2) The other one is the wavelet based image quality evaluation. It analyzes image from different frequency resolutions through wavelet theory, and analyzes the change of Wavelet Coefficients through statistics analysis, and then assesses the image quality.The realization of auto-focus, for the DFF (depth from focus), is similar with the 1-demension search, Fibonacci search, function approaching or MCS (mountain climbing servo), calculates the auto-focus function value, compares it with the previous one and decides the next position; for the DFD (depth from defocus), moves a special distance relative to the image quality level and system parameters directly. Depending on above auto-focus theories, the thesis puts forward an auto-focus system architecture and analyzes its performance, which is constructed by high speed DSP (Digital Signal Processor) and FPGA (Field Programmable Gate Array).更多还原