数字图像自动聚焦技术研究及系统实现
|
摘要
从上个世纪中叶至今,光电成像系统的应用发展迅速,从简单地用于完成对静态图像的记录,逐渐发展到用于对目标对象进行分析,并广泛应用于工业、农业、医学、军事及科学研究等领域。在一个成像系统中,光学镜头的焦距对准是排在所有其它功能之前首先需要解决的基本问题,它的最终效果将直接决定最后成像的图像质量及后续图像应用的有效性。自动聚焦技术的发展与成像系统的发展有着紧密的联系。自动聚焦方式的发展集中体现在成像系统的自动聚焦技术研究上以及对焦机构设计上的不断创新。而对焦机构集中体现了成像系统光学、机械、电子一体化的程度,并直接决定了它的性能。
自动聚焦已成为各种成像系统的重要功能,与国外相比,国内在这方面的工作做得还比较少,同时图像处理技术的发展使得自动聚焦趋于数字化和智能化。提出采用基于图像技术的自动聚焦方法并研发性能优良的自动聚焦系统具有重要的现实意义,它不仅对推动光机电一体化技术进步,发展现代光学仪器产业具有重要的意义,同时如果能把自动聚焦模块与系统集成,必定能带动我国数字成像系统产业的发展,使我国在国内、国际市场上具有更强的竞争力。本论文研究课题来源于浙江省网上重大招标项目“基于数字技术的图像自动调焦系统”(编号:021105778,KYZ011103002)和国家自然科学基金项目“基于神经网络识别及控制的图像自动调焦方法和系统的研究”(编号:60672063)。
基于图像技术的自动聚焦方法采用了与传统聚焦技术完全不同的方式进行对焦,传统的聚焦方法是通过传感器检测焦点或测量距离的方式实现的,而基于图像技术的聚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,然后完成聚焦操作。自动聚焦装置是光电成像系统的一个重要的部分。本文首先结合自动聚焦领域的发展和现状,对目前为止所出现的各类自动聚焦方法分别进行了论述,同时分析和比较了这些自动聚焦方法的优缺点,并认为自动聚焦方法的未来发展中,图像处理方法是最具发展前景的。之后,对自动聚焦技术的发展趋势进行了论述,并认为高度集成化、智能化、低功耗和高速处理将是今后自动聚焦技术的显著特征。
从基本原理来说,自动聚焦可以分成两大类:一类是基于镜头与被摄目标之间距离测量的测距方法,另一类是基于聚焦屏上成像清晰的聚焦检测方法。近年来还有一种就是基于图像处理的自动聚焦方法。本文对三类自动聚焦方式进行了论述,并着重讨论了数字图像自动对焦系统原理、基本组成模块及对焦评价函数应具有的性质。
数字图像预处理技术是基于图像处理的自动聚焦方法的基础和前提条件,本文讨论了图像的数字化及其特性,结合CMOS图像传感器成像对光线敏感的特点,提出了灰度值线性变换和灰度直方图均衡两种灰度校正技术;针对不同的噪声,从加性和乘性两个角度介绍了不同的实现方法,对于加性噪声可采用邻域平均法和中值滤波的方法,通过实验比较,邻域平均法会削弱图像的边缘并对椒盐噪声的滤除效果不好,中值滤波在滤除噪声的同时很好地保存了图像边缘。而对于乘性噪声可采用同态滤波的方法,先把噪声与信号分离,然后滤波,最后恢复信号,它的滤波效果由滤波器决定。
图像质量分析评价是自动聚焦方法中的关键技术,本文从三种途经详细论述了图像质量评价方法的实现。
(1)基于对比度的图像质量评价方法从图像的时域、频域及信息熵三个角度,建立能表示图像对比度的一些调焦函数,并对这些评价函数做了详细的比较,最后确定出时域的Brenner函数和绝对方差函数具有更好的综合性能。在此基础上,提出基于灰度对比度变化率和基于自相关函数对图像作清晰度评价的两种改进算法,仿真结果表明这两种方法简单实用,兼顾空间域函数的优点并且克服了空间域评价函数对噪声干扰敏感和对对焦区域选择要求高的缺点,具有更好的实时响应性能。
(2)基于功率谱的客观图像质量评价方法假设场景的功率谱具有不变特性,引入了人类视觉系统,加入了维纳噪声滤波器,对图像质量进行评价得到一个确定的IQM数值。该数值与人的视觉评价具有很高的相关性。IQM方法可以对图像直接进行质量评价。log(IQM)值与人的视觉评价值基本上可以满足一种线性关系,因而它的评价结果可以代表人的视觉评价,作为其他应用的参考,也可以很直接地应用到调焦控制。调焦应用中,根据IQM值可以明确地控制焦距调整量,与对比度方法相比减少了搜索过程,因而在精度要求不是很高的情况下,该方法具有更快的响应能力。
(3)基于小波与神经网络的图像质量评价方法利用小波分析对图像进行多分辨率分解,分析其细节信息并采用统计的方法提取图像特征,再利用人工神经网络对图像特征进行质量模式识别,得到图像的质量等级。该方法基于生物视觉机理,在分析人眼调焦和成像系统调焦之间关系的基础上,利用神经网络的非线性特性,在对焦评判规则中包含了人的主观因素,以改善成像系统的调焦效果。实验表明,该方法能达到满意的识别率。
调焦实现过程中对图像进行分块并选择合适的处理方法,能使实际处理的数据仅为其中的一小部分,从而可以大大减少计算量。本文详细介绍了相对复杂的对比度法调焦的工作过程,通过对成像过程中可能发生的一些异常现象进行分析总结,提出了一种阈值比较的方法,克服了搜索过程中的干扰波动。同时还简要说明了功率谱法和小波与神经网络法的调焦操作过程。对比度法调焦是一个多次调整、逼近的过程,因而它可以达到很高的精度,但每次调整之后都要重新计算,因而它会比较费时;而功率谱法和小波与神经网络法调焦基本上是一次调焦,虽然它的精度要差一些,但它的响应速度更快。
在自动聚焦技术研究的基础上,提出了基于FPGA的自动调焦系统实现方案,分别对组成系统的光学成像模块、图像采集模块、图像处理模块、控制模块和驱动模块的特性和要求进行分析,同时对自动调焦系统实现方案进行了具体的设计并确定了器件类型。对系统实现过程中涉及到的几个关键问题进行了分析研究,包括系统各模块时间延时特性、FFT算法实现、I~2C接口实现、图像信号采集的芯片配置、VGA接口实现及FPGA器件的配置等。最后对系统实现中出现的各类问题及出现原因进行了详细说明与分析,提出了改进后的系统结构,为以后系统的改进指明了方向。
From the middle period of the last century to now, the applications of photoelectricity imaging systems have been extended, which were limited to perform the function of recording static images at the early time and now come into use to analyze a target object. And nowadays photoelectricity imaging systems are applied to many fields, such as industry, agriculture, medicine, military affairs and laboratory research etc. In a photoelectricity imaging system, precisely focusing is a primary problem to be solved, which greatly affects the image quality in back end and the efficiency of coming performance with image. There is a close relationship between the development of the auto-focusing technology and the development of cameras. And the development of the auto-focusing technology is represented by the innovations in the design of auto-focusing mechanism inside a camera, which embodies the level of integration of optics, mechanics and electronics, and directly affects the performance of a camera.
Auto-focus is an important function in all kinds of imaging systems. It has widely application in many fields. In our country, the research and development about auto-focus is poor, but we are facing kinds of application requirement. The image processing is a rapidly developing technology. It promotes the development of the auto-focus technology greatly. And numeralization and intelligentization has being a tendence of the auto-focus technology. For these conditions, providing the auto-focus means based on image processing technology and developing the auto-focusing systems with excellent performance will have important significance. It not only can push the advancement of the integrated technology of photoelectricity and mechanics, develop the industry of contemporary optical instruments, but also can progress the industry of digital imaging systems if the auto-focusing modules are integrated into the imaging systems. Finally it makes our imaging systems to have stronger competition in domestic and international markets. The research subject comes from the Zhejiang provincial grand item of science & technology "The auto-focusing image system based digital technology" (No. 021105778, KYZ011103002) and the national nature science foundation item of China "Study on the auto-focusing method and system based on neural network identification and control" (No. 60672063).
The auto-focusing method based on image technology is different from the traditional auto-focusing method entirely. The traditional ones must depend on some special assistant units. They use these units to measure distance or find focus and realize auto-focus. However, the auto-focus based on image technology is completed by analyzing the image quality directly, judging the imaging state from the current image quality, and adjusting correctly the focusing distance supervised by the imaging state. The auto-focusing device is an important part in a photoelectricity imaging system. The paper dissertates all kinds of auto-focusing methods considering the process and actuality of the auto-focusing technology, and then analyzes and compares their characteristics. It comes that the auto-focusing method based on image technology has developmental potential. Then the paper describes a developing trend of the auto-focusing technology and thinks that high integration, intelligence, low power consumption and high processing speed are the coming remarkable character of the auto-focusing technology.
From the essential principle, the auto-focus can be divided into two classes. The one is the method based on measuring the distance between lens and an object. The other is the method based on measuring the focusing condition on a focusing screen. In the recent years, the auto-focusing method based on image treatment appeared. The paper dissertates three kings of auto-focusing methods and emphatically discusses the essential principle and the compositive modules of auto-focusing systems and the due property of focusing evaluation functions.
The preprocess technology of digital image is a basis and prerequisite of the auto-focusing method based on image treatment. The paper discusses the image numeralization and its character, presents the two greyness-emendation technology based on the linear transform and the histogram equilibrium of image greyness-value considering the ray sensitivity of CMOS image sensors, introduces the different realization methods against positive noise and multiple noise. As to positive noise, the mean filter of neighborhoods and median filter can be used. According to the result of the experiments, the former will weaken the image edge and will not be good for reducing noise of spiced salts. The latter has a good effect to reduce positive noise and to conserve the image edge. As for the reduction of multiple noise, the homomorphic filter can be chosen, which separates signal and noise, filters noise, and then restores original signal. The filter's effect is determined by its design.
The image quality analysis and evaluation is the key technology. In the thesis, three image quality analysis methods are presented detailedly.
(1) The first one is the image quality evaluation based on image contrast. It evaluates the image definition at temporal field, frequency field and information entropy. By comparing their performance in the number of calculation and the time consumption of calculation, we find out the two better functions that are respectively the Brenner function and the absolute variance function. On the foundation, the paper presents the two improved algorithms based on the greyness-contrast variation and the autocorrelation function. The simulation results indicate that the two methods are simple and practical, not only have the merits of special evaluation functions but also overcome the demerits of strong sensitivity to noise and high demand for focusing area of special evaluation functions.
(2) The next one is the image quality evaluation based on image power spectrum. It supposes nature scenes have the same power spectrum, introduces a human vision system and includes a Winner de-noise filter. It evaluates a picture and gets an IQM (Image Quality Measure) value. The IQM value is highly relative to the estimation result through human vision system. The IQM method can evaluate image quality directly. The log(IQM) value and the human vision evaluation can satisfy an approximate linear relation. So its evaluation results can represent the human vision evaluation and be used for the references of other applications and the direct focusing control. The IQM value can control the adjustive focusing distance definitely in focusing applications. It decreases searching process compared with the image contrast method and has faster response when the demand for precision is not strict.
(3) The last one is the image quality evaluation based on wavelet analysis and neural network. It analyzes image from different frequency resolutions through wavelet theory, and abstracts the image characteristics from the detail information through statistics analysis, then uses an artificial neural network for quality pattern recognition in order to get image quality grades. The method is based on the biology vision mechanism. On the foundation analyzing the focusing relation of human's eyes and imaging systems, it includes human subjective factors into focusing evaluation rules to improve the focusing effect of imaging systems by the nonlinear character of neural network. The experiment results indicate that the recognition rate of this method is good.
An auto-focusing system carries on partitioning and selecting of image in a focusing process to reduce the data quantity to be process, so the calculation decreases greatly. The paper introduces detailedly the focusing process of the complicated image contrast method, presents the method comparing thresholds to overcome disturbing influence in the course of focusing search by analyzing and summarizing the exceptional phenomena which may happen in the imaging process, and shows briefly the focusing courses of the power spectrum method and the wavelet analysis & neural network method. The focusing realization for the contrast method is an approaching course of repetitious regulations, so it can reach to a high precision. But it takes time because of calculating over again after every regulation. As for the power spectrum method or the wavelet analysis & neural network method, they move a special distance relative to the image quality level directly. They have more rapid response although their precision is a little lower.
Depending on the above research of the auto-focusing technology, the paper presents the realization scheme of an auto-focusing system based on FPGA (Field Programmable Gate Array), analyzes the characteristics and demands for the optics imaging module, the image collection module, the image processing module, the control module and the driving module, designs and confirms the concrete realization scheme and systemic parts, and studies some key problems of the system realization. Those problems include the time-delay character of all modules, the realization of FFT algorithms, the interface realization of I~2C, the chip configuration of image collection, the interface realization of VGA and the chip configuration of FPGA. Finally, the paper analyzes detailedly the problems and reasons of the system realization, presents the improved system structure, and designates the direction of improving systems.
自动聚焦已成为各种成像系统的重要功能,与国外相比,国内在这方面的工作做得还比较少,同时图像处理技术的发展使得自动聚焦趋于数字化和智能化。提出采用基于图像技术的自动聚焦方法并研发性能优良的自动聚焦系统具有重要的现实意义,它不仅对推动光机电一体化技术进步,发展现代光学仪器产业具有重要的意义,同时如果能把自动聚焦模块与系统集成,必定能带动我国数字成像系统产业的发展,使我国在国内、国际市场上具有更强的竞争力。本论文研究课题来源于浙江省网上重大招标项目“基于数字技术的图像自动调焦系统”(编号:021105778,KYZ011103002)和国家自然科学基金项目“基于神经网络识别及控制的图像自动调焦方法和系统的研究”(编号:60672063)。
基于图像技术的自动聚焦方法采用了与传统聚焦技术完全不同的方式进行对焦,传统的聚焦方法是通过传感器检测焦点或测量距离的方式实现的,而基于图像技术的聚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,然后完成聚焦操作。自动聚焦装置是光电成像系统的一个重要的部分。本文首先结合自动聚焦领域的发展和现状,对目前为止所出现的各类自动聚焦方法分别进行了论述,同时分析和比较了这些自动聚焦方法的优缺点,并认为自动聚焦方法的未来发展中,图像处理方法是最具发展前景的。之后,对自动聚焦技术的发展趋势进行了论述,并认为高度集成化、智能化、低功耗和高速处理将是今后自动聚焦技术的显著特征。
从基本原理来说,自动聚焦可以分成两大类:一类是基于镜头与被摄目标之间距离测量的测距方法,另一类是基于聚焦屏上成像清晰的聚焦检测方法。近年来还有一种就是基于图像处理的自动聚焦方法。本文对三类自动聚焦方式进行了论述,并着重讨论了数字图像自动对焦系统原理、基本组成模块及对焦评价函数应具有的性质。
数字图像预处理技术是基于图像处理的自动聚焦方法的基础和前提条件,本文讨论了图像的数字化及其特性,结合CMOS图像传感器成像对光线敏感的特点,提出了灰度值线性变换和灰度直方图均衡两种灰度校正技术;针对不同的噪声,从加性和乘性两个角度介绍了不同的实现方法,对于加性噪声可采用邻域平均法和中值滤波的方法,通过实验比较,邻域平均法会削弱图像的边缘并对椒盐噪声的滤除效果不好,中值滤波在滤除噪声的同时很好地保存了图像边缘。而对于乘性噪声可采用同态滤波的方法,先把噪声与信号分离,然后滤波,最后恢复信号,它的滤波效果由滤波器决定。
图像质量分析评价是自动聚焦方法中的关键技术,本文从三种途经详细论述了图像质量评价方法的实现。
(1)基于对比度的图像质量评价方法从图像的时域、频域及信息熵三个角度,建立能表示图像对比度的一些调焦函数,并对这些评价函数做了详细的比较,最后确定出时域的Brenner函数和绝对方差函数具有更好的综合性能。在此基础上,提出基于灰度对比度变化率和基于自相关函数对图像作清晰度评价的两种改进算法,仿真结果表明这两种方法简单实用,兼顾空间域函数的优点并且克服了空间域评价函数对噪声干扰敏感和对对焦区域选择要求高的缺点,具有更好的实时响应性能。
(2)基于功率谱的客观图像质量评价方法假设场景的功率谱具有不变特性,引入了人类视觉系统,加入了维纳噪声滤波器,对图像质量进行评价得到一个确定的IQM数值。该数值与人的视觉评价具有很高的相关性。IQM方法可以对图像直接进行质量评价。log(IQM)值与人的视觉评价值基本上可以满足一种线性关系,因而它的评价结果可以代表人的视觉评价,作为其他应用的参考,也可以很直接地应用到调焦控制。调焦应用中,根据IQM值可以明确地控制焦距调整量,与对比度方法相比减少了搜索过程,因而在精度要求不是很高的情况下,该方法具有更快的响应能力。
(3)基于小波与神经网络的图像质量评价方法利用小波分析对图像进行多分辨率分解,分析其细节信息并采用统计的方法提取图像特征,再利用人工神经网络对图像特征进行质量模式识别,得到图像的质量等级。该方法基于生物视觉机理,在分析人眼调焦和成像系统调焦之间关系的基础上,利用神经网络的非线性特性,在对焦评判规则中包含了人的主观因素,以改善成像系统的调焦效果。实验表明,该方法能达到满意的识别率。
调焦实现过程中对图像进行分块并选择合适的处理方法,能使实际处理的数据仅为其中的一小部分,从而可以大大减少计算量。本文详细介绍了相对复杂的对比度法调焦的工作过程,通过对成像过程中可能发生的一些异常现象进行分析总结,提出了一种阈值比较的方法,克服了搜索过程中的干扰波动。同时还简要说明了功率谱法和小波与神经网络法的调焦操作过程。对比度法调焦是一个多次调整、逼近的过程,因而它可以达到很高的精度,但每次调整之后都要重新计算,因而它会比较费时;而功率谱法和小波与神经网络法调焦基本上是一次调焦,虽然它的精度要差一些,但它的响应速度更快。
在自动聚焦技术研究的基础上,提出了基于FPGA的自动调焦系统实现方案,分别对组成系统的光学成像模块、图像采集模块、图像处理模块、控制模块和驱动模块的特性和要求进行分析,同时对自动调焦系统实现方案进行了具体的设计并确定了器件类型。对系统实现过程中涉及到的几个关键问题进行了分析研究,包括系统各模块时间延时特性、FFT算法实现、I~2C接口实现、图像信号采集的芯片配置、VGA接口实现及FPGA器件的配置等。最后对系统实现中出现的各类问题及出现原因进行了详细说明与分析,提出了改进后的系统结构,为以后系统的改进指明了方向。
From the middle period of the last century to now, the applications of photoelectricity imaging systems have been extended, which were limited to perform the function of recording static images at the early time and now come into use to analyze a target object. And nowadays photoelectricity imaging systems are applied to many fields, such as industry, agriculture, medicine, military affairs and laboratory research etc. In a photoelectricity imaging system, precisely focusing is a primary problem to be solved, which greatly affects the image quality in back end and the efficiency of coming performance with image. There is a close relationship between the development of the auto-focusing technology and the development of cameras. And the development of the auto-focusing technology is represented by the innovations in the design of auto-focusing mechanism inside a camera, which embodies the level of integration of optics, mechanics and electronics, and directly affects the performance of a camera.
Auto-focus is an important function in all kinds of imaging systems. It has widely application in many fields. In our country, the research and development about auto-focus is poor, but we are facing kinds of application requirement. The image processing is a rapidly developing technology. It promotes the development of the auto-focus technology greatly. And numeralization and intelligentization has being a tendence of the auto-focus technology. For these conditions, providing the auto-focus means based on image processing technology and developing the auto-focusing systems with excellent performance will have important significance. It not only can push the advancement of the integrated technology of photoelectricity and mechanics, develop the industry of contemporary optical instruments, but also can progress the industry of digital imaging systems if the auto-focusing modules are integrated into the imaging systems. Finally it makes our imaging systems to have stronger competition in domestic and international markets. The research subject comes from the Zhejiang provincial grand item of science & technology "The auto-focusing image system based digital technology" (No. 021105778, KYZ011103002) and the national nature science foundation item of China "Study on the auto-focusing method and system based on neural network identification and control" (No. 60672063).
The auto-focusing method based on image technology is different from the traditional auto-focusing method entirely. The traditional ones must depend on some special assistant units. They use these units to measure distance or find focus and realize auto-focus. However, the auto-focus based on image technology is completed by analyzing the image quality directly, judging the imaging state from the current image quality, and adjusting correctly the focusing distance supervised by the imaging state. The auto-focusing device is an important part in a photoelectricity imaging system. The paper dissertates all kinds of auto-focusing methods considering the process and actuality of the auto-focusing technology, and then analyzes and compares their characteristics. It comes that the auto-focusing method based on image technology has developmental potential. Then the paper describes a developing trend of the auto-focusing technology and thinks that high integration, intelligence, low power consumption and high processing speed are the coming remarkable character of the auto-focusing technology.
From the essential principle, the auto-focus can be divided into two classes. The one is the method based on measuring the distance between lens and an object. The other is the method based on measuring the focusing condition on a focusing screen. In the recent years, the auto-focusing method based on image treatment appeared. The paper dissertates three kings of auto-focusing methods and emphatically discusses the essential principle and the compositive modules of auto-focusing systems and the due property of focusing evaluation functions.
The preprocess technology of digital image is a basis and prerequisite of the auto-focusing method based on image treatment. The paper discusses the image numeralization and its character, presents the two greyness-emendation technology based on the linear transform and the histogram equilibrium of image greyness-value considering the ray sensitivity of CMOS image sensors, introduces the different realization methods against positive noise and multiple noise. As to positive noise, the mean filter of neighborhoods and median filter can be used. According to the result of the experiments, the former will weaken the image edge and will not be good for reducing noise of spiced salts. The latter has a good effect to reduce positive noise and to conserve the image edge. As for the reduction of multiple noise, the homomorphic filter can be chosen, which separates signal and noise, filters noise, and then restores original signal. The filter's effect is determined by its design.
The image quality analysis and evaluation is the key technology. In the thesis, three image quality analysis methods are presented detailedly.
(1) The first one is the image quality evaluation based on image contrast. It evaluates the image definition at temporal field, frequency field and information entropy. By comparing their performance in the number of calculation and the time consumption of calculation, we find out the two better functions that are respectively the Brenner function and the absolute variance function. On the foundation, the paper presents the two improved algorithms based on the greyness-contrast variation and the autocorrelation function. The simulation results indicate that the two methods are simple and practical, not only have the merits of special evaluation functions but also overcome the demerits of strong sensitivity to noise and high demand for focusing area of special evaluation functions.
(2) The next one is the image quality evaluation based on image power spectrum. It supposes nature scenes have the same power spectrum, introduces a human vision system and includes a Winner de-noise filter. It evaluates a picture and gets an IQM (Image Quality Measure) value. The IQM value is highly relative to the estimation result through human vision system. The IQM method can evaluate image quality directly. The log(IQM) value and the human vision evaluation can satisfy an approximate linear relation. So its evaluation results can represent the human vision evaluation and be used for the references of other applications and the direct focusing control. The IQM value can control the adjustive focusing distance definitely in focusing applications. It decreases searching process compared with the image contrast method and has faster response when the demand for precision is not strict.
(3) The last one is the image quality evaluation based on wavelet analysis and neural network. It analyzes image from different frequency resolutions through wavelet theory, and abstracts the image characteristics from the detail information through statistics analysis, then uses an artificial neural network for quality pattern recognition in order to get image quality grades. The method is based on the biology vision mechanism. On the foundation analyzing the focusing relation of human's eyes and imaging systems, it includes human subjective factors into focusing evaluation rules to improve the focusing effect of imaging systems by the nonlinear character of neural network. The experiment results indicate that the recognition rate of this method is good.
An auto-focusing system carries on partitioning and selecting of image in a focusing process to reduce the data quantity to be process, so the calculation decreases greatly. The paper introduces detailedly the focusing process of the complicated image contrast method, presents the method comparing thresholds to overcome disturbing influence in the course of focusing search by analyzing and summarizing the exceptional phenomena which may happen in the imaging process, and shows briefly the focusing courses of the power spectrum method and the wavelet analysis & neural network method. The focusing realization for the contrast method is an approaching course of repetitious regulations, so it can reach to a high precision. But it takes time because of calculating over again after every regulation. As for the power spectrum method or the wavelet analysis & neural network method, they move a special distance relative to the image quality level directly. They have more rapid response although their precision is a little lower.
Depending on the above research of the auto-focusing technology, the paper presents the realization scheme of an auto-focusing system based on FPGA (Field Programmable Gate Array), analyzes the characteristics and demands for the optics imaging module, the image collection module, the image processing module, the control module and the driving module, designs and confirms the concrete realization scheme and systemic parts, and studies some key problems of the system realization. Those problems include the time-delay character of all modules, the realization of FFT algorithms, the interface realization of I~2C, the chip configuration of image collection, the interface realization of VGA and the chip configuration of FPGA. Finally, the paper analyzes detailedly the problems and reasons of the system realization, presents the improved system structure, and designates the direction of improving systems.
引文
[1]吴世法.近代成像技术与图像处理.北京:国防工业出版社,1997.
[2]沈庭芳,方子文编著.数字图像处理及模式识别.北京:北京理工大学出版社,1998.
[3]夏德深,傅德盛。现代图像处理技术与应用.南京:东南大学出版社.1997.
[4]Kenneth.R.Castleman著,朱志刚等译.数字图像处理.北京:电子工业出版社,1998.
[5]王润生编著.图像理解.长沙:国防科技大学出版社,1995.
[6]黄其昆.自动调焦二十年.大众摄影,2001,(5):52-55.
[7]董太和.相机的MF/AF机构及其精度:相机机构专题讲座(1).照相机,2000,(2).
[8]许盛,李见为.高级显微镜系统智能化自动调焦的研究及实现.光学仪器,2000,22(5):22-26.
[9]王建民,浦昭邦,尹继学,晏磊.图像式显微自动调焦瞄准系统的研究.光学技术,2000,26(5):410-417.
[10]许盛,李见为.基于智能化自动调焦的高级显微镜系统.光电工程,2000,27(1):33-36.
[11]一川.中国应开发AF相机.照相机,2000,(2):8-10.
[12]徐之海,李奇.现代成像系统.北京:国防工业出版社,2001.
[13]白磊.CCD显微测量系统自动调焦技术的研究.哈尔滨工业大学硕士学位论文,2002.
[14]H.Harms and H.M.Aus.Comparison of Digital Focus Criteria for a TV Microscope Systeml.Cytometry,1984,(5):236-243.
[15]Mason DC,Green DK.Automated Forusing of a Computer Controlled Microscopel.IEEE Trans Biomed Eng BME,1975,22:137-150.
[16]黄剑琪.基于频谱分析的数字对焦技术研究.浙江大学硕士学位论文,2001.
[17]M.Subbarao,T.S.Choi,and A.Nikzad.Focusing Techniques.Optical Engineering,1993,32(11):2824-2836.
[18]E.Krotkov.Focusing.International Journal of Computer Vision,1987,(1):223-237.
[19]Yalin Xiong and Steven A.Shafer.Depth from Focusing and Defocusing.Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1992.
[20]M.Subbarao,T.Choi,and A.Nikzad.Focusing Techniques.J.Optical Eng.,1993,32(11):2824-2836.
[21]S.Jutamulia,T.Asakura,R.D.Bahuguna,P.C.De.Guzman.Autofocusing Based on Power-Spectra Analysis.Appl.Optics,1994,33(26):6210-6212.
[22]Murali Subbarao,Jenn-Kwei Tyan.Selecting the Optimal Focus Measure for Autofocusing and Depth-From-Focus.IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(8):861-870.
[23]曹茂永,孙农亮,郁道银.基于灰度梯度的数字图像评价函数.光电工程,2003,30(4):69-72.
[24]李奇,冯华君,徐之海.数字图像清晰度评价函数研究.光子学报,2002,31(6):376-378.
[25]M.Subbarao,J.K.Tyan.The Optimal Focus Measure for Passive Autofocusing and Depth-from-Focus.Proceeding of SPIE Conference in Video Metrics,1995.
[26]Biberman V.and Konstantinos K.Image Quality In Perception of Displayed Information.New York:Plenum Press,1997.
[27]赵荣椿,赵忠明.数字图像导论.西安:西北工业大学出版社,1996.
[28]杨枝灵,王开等.数字图像获取、处理及实践应用.北京:人民邮电出版社,2003.
[29]王新成.高级图像处理技术.北京:中国科学技术出版社,2001.
[30]Chen Guojin,Liu Bin,Gong Ye,Xu Guohua,Shi Huli.Research on Image Auto-focus System Based on Neural Network.Proceedings of Fifth International Symposium on Instrumentation and Control Technology,2003:461-465.
[31]Chen Guojin,Zhu Miaofen,Guan Liming,Liu Bin.Implementing the Auto-focusing System Based on the Digital Image Process Technology.Proceedings of IEE Irish Signals and Systems Conference,2005:92-95.
[32]郁道银,谈恒英.工程光学.北京:机械工业出版社,1999.
[33]夏良正.数字图像处理.南京:东南大学出版社,1999.
[34]李智勇,沈振康.一种单镜头运动测距方法.电子学报,1990,18(5):70-74.
[35]Yoshida Y,Shinohara S,Ikeda H,Tada K,Yoshida H,Nishide K,Sum M.Control of Lens Position in Auto-focus Cameras Using Semiconductor Laser Range Finder.Circuits and Systems,1991,(1):395-398.
[36]Getzlaff S,Schreiter J,Graupner A,Schuffny R.A System-on-chip Realization of a CMOS Image Sensor with Programmable Analog Image Preprocessing.The 2001 IEEE International Symposium on Circuits and Systems,2001(4):486-489.
[37]Sang Ku Kim,Sang Rae Park.Simultaneous Out-of-focus Blur Estimation and Restoration for a Digital Auto-focusing System.IEEE Transaction on Consumer Electronics,1998,44(3):1071-1075.
[38]Sang Ku Kim,Joon Ki Paik.Out-of-focus Blur Estimation and Restoration for Digital Auto-focusing System.Electronics Letters,1998,34(12):1217-1219.
[39]Chung Nam Cho,etc.Digital Auto-focusing of Multiple Objects Based on Image Restoration.Proceedings of SPIE-THE International Society for Optical Engineering,2000,20(23):1642-1651.
[40]Dick Merrill.The Next-generation Digital Camera.Optics & Photonies News,1992,(1):26-49.
[41]侯雨石,何玉青,陈永飞,张忠廉.数码相机图像传感器技术.光学技术2003,29(1):59-62.
[42]D.Marr and E.Hildreth.Theory of Edge Detection.London:Proc.R.Soc.,1980.
[43]F.Bergholm.Edge Focusing.IEEE Trans.Patt.Anal.Mach.Intell.,1987,9(6):726-741.
[44]Canny J.A Computational Approach to Edge Detection.IEEE Trans.PAMI, 1986,8(6):679-698.
[45]程佩青.数字信号处理教程.北京:清华大学出版社,1998.
[46]杨长生编著.图像与声音压缩技术.杭州:浙江大学出版社,2000.
[47]ISO/IEC JTC1/SC29 WG1.JPEG 2000 Part Ⅰ Final Committee Draft Version 1.0.2000.
[48]曹茂永,孙农亮.离焦模糊图像清晰度评价函数的研究.仪器仪表学报,2001,22(3):426-430.
[49]Eli Peli.Contrast Sensitivity Function and Image Discrimination.J.Opt.Soc.Am.A,2000,18(2):283-293.
[50]Brenner.J.F,Dew,B.S.Horton,J.B.Neirath.An Automated Microscope for Cytologic Research.J.Histochem.Cytochem.,1971,24:100-111.
[51]Jan-Mark Geusebroek,Frans Cornelissen,Arnold W.M.Smeulders and Hugo Geerts.Robust Autofocusing in Microscopy.Cytometry,2000,36(1):1-9.
[52]黄剑琪,冯华君,徐之海,戴顺妹,李奇,边美娟.边缘特征增强算法和小波分析在精确聚焦中的应用.光子学报,2000,29(10):932-936.
[53]白立芬,于水,薛实福,李庆祥.基于图像处理的自动调焦方法中清晰度判断函数的研究.仪器仪表学报,1999,20(4):85-89.
[54]郑南宁.计算机视觉与模式识别.北京:国防工业出版社,1998.
[55]吴振峰,左洪福,邱根良.光学显微镜自动调焦的技术研究.光学仪器,2000,22(4):9-12.
[56]Vollath D.Automatic Focusing by Correlative Methods.J.Microsc.,1987,147:279-288.
[57]Xiao Tang,Pierre L'Hostis,and Yu Xiao.An Auto-focusing Method in a Microscopic Test Bed for Optical Discs.Journal of Research of the National Institute of Standards and Technology,2000,105(4):565-569.
[58]D.G.Marimont and Y.Rubner.A Probabilistic Framework for Edge Detection and Scale Selection.The Sixth International Conference on Computer Vision,1998,4(7):207-214.
[59]Kang-Sun Choi,Jun-Suk Lee,and Sun-Jae Ko.New Autofocusing Technique Using the Frequency Selective Weighted Median Filter for Video Cameras.IEEE Transactions on Consumer Electronics,1999,45.
[60]张晓森.基于概率的字符识别系统的若干问题研究.西安电子科技大学硕士学位论文.
[61]邢彦辰.基于局部熵差的图象匹配方法在图象实时监测中应用研究.哈尔滨工业大学硕士学位论文,2003.
[62]东秀阁.基于信息熵的图像分割方法研究及其在DSP中的快速实现.天津大学硕士学位论文.
[63]A.Santos,C.Ortizde Solorzano,J.J.Vaquero,J.M.Pena,N.Malpica and F.Del Pozo.Evaluation of Autofocus Functions in Molecular Autogenetic.Journal of Microscopy,1997,188(3):264-272.
[64]Firestone,L.Cook,K.Culp,K.Talsania,N.Preston.Comparison of Autofocus Methods for Automated Microscopy.Cytometry,1991,12:195-206.
[65]Vollath D.The Influence of the Scene Parameters and Noise on the Behaviour of Automatic Focusing Algorithms.J.Microsc.,1988,151:133-146.
[66]N.B.Nill and B.H.Bouzas.Objective Image Quality Measures Derived From Digital Image Power Spectra.Optical Engineering,1992,31(4):813-825.
[67]K.Falconer.Fractal Geometry - Mathematical Foundations and Applications.New York:Wiley,1990.
[68]盛骤,谢式千,潘承毅编.概率论与数理统计.北京:高等教育出版社,2000.
[69]D.J.Granrath.The Role of Human Visual Models in Image Processing.Proc.IEEE,1981,69(5):552-561.
[70]N.B.Nill.A Visual Model Weighted Cosine Transform for Image Compression and Quality Assessment.IEEE Trans.Commun,1985,33(6):551-557.
[71]J.J.DePalma and E.M.Lowry.Sine Wave Response of the Visual System,(Ⅱ).Sine Wave and Square Wave Contrast Sensitivity.J.Opt.Soc.Am,1962,52(3):328-335.
[72]J.L.Mannos and D.J.Sakrison.The Effects of a Visual Fidelity Criterion on the Encoding of Images.IEEE Trans.Inform.Theory,1974,20(4):525-536.
[73]K.N.Ngan,K.S.Leong,and H.Singh.Adaptive Cosine Transform Coding of Images in Perceptual Domain.IEEE Trans.Acoust.Speech Signal Process,1989,37(11):1743-1750.
[74]张兆礼,赵春晖,梅晓丹著.现代图像处理技术及Matlab实现.北京:人民邮电出版社,2001.
[75]寿天德.视觉信息处理的脑机制.上海:上海科技教育出版社,1997.
[76]杨雄里.视觉的神经机制.上海:上海科学技术出版社,1996.
[77]徐广第.眼科屈光学.北京:军事医学科学出版社,2001.
[78]刘健勤,盛津芳,魏敏.面向智能体的视觉神经信息处理.北京:科学出版社,2000.
[79]程正兴著.小波分析算法与应用.西安:西安交通大学出版社,1998.
[80]王建平,盛军,朱程辉.基于小波分析的视频图像字符特征提取方法研究.微电子与计算机,2002,(5):51-56.
[81]杨胜波,于春海.小波分析在生物医学信号/图像处理中的应用.仪器仪表学报,2002,23(3):179-181.
[82]孟军,魏同立,吴金,常昌远.数字图像离散小波变换的原理与硬件实现分析.东南大学学报,2002,32(6):1-6.
[83]韦志辉,程军.基于小波变换的一种新的图像质量评价方法.南京理工大学学报,1998,22(6):552-555.
[84]甘斌,张雄伟,甘仲民.基于小波变换的多尺度图像边缘处理.数字电视与数字视频,2001,(8):20-22.
[85]黄应清,季向琦,张智诠.图像小波系数的分布规律及其对图像质量的影响.红外与激光工程,2000,29(3):15-18.
[86]张新明,沈兰荪.基于小波和统计特性的自适应图像增强.信号处理,2001,17(3):227-231.
[87]E.A.Fadhel and P.Bhattacharyya.Application of a Steerable Wavelet Transform Using Neural Network for Signature Verification.Pattern Analysis &Applications,1999,(2):184-195.
[88]B.J.W.Fleming,D.Yu,R.G.Harrison,and D.Jubb.Wavelet-based Detection of Coherent Structures and Self-affnity in Financial Data.J.Eur.Phys.,2001,(20): 543-546.
[89]Z.Peng,Y.He,Q.Lu and F.Chu.Feature Extraction of the Rub-impact Rotor System by Means of Wavelet Analysis.Journal of Sound and Vibration,2003,259(4):1000-1010.
[90]M.Antonopoulos-Domis,T.Tambouratzis.System Identification,A Transient via Wavelet Multiresolution Analysis.Ann.Nucl.Energy,1998,25(6):465-480.
[91]I.Urriza,J.I.Artigas,L.A.Barragan,J.I.Garcia and D.Navarro.VLSI Implementation of Discrete Wavelet Transform for Lossless Compression of Medical Images.Real-Time Imaging,2001,(7):203-217.
[92]边肇祺,张学工等编著.模式识别.北京:清华大学出版社,2000.
[93]Zhengxin Hou,Wenguan Wang and Jinghui Chu.A Kind of 2D Isotropic Wavelet and Its Application in Image Primitive Detection.Journal of Circuits Systems and Computers,2002,11(2):123-135.
[94]黄德双编著.神经网络模式识别系统理论.北京:电子工业出版社,1996.
[95]焦李成编著.神经网络的应用与实现.西安:西安电子科技大学出版社,1996.
[96]Abhijit S.Pandya,Robert B.Macy著,徐勇,荆涛等译.神经网络模式识别及应用.北京:电子工业出版社,1999.
[97]杨建刚编著.人工神经网络实用教程.杭州:浙江大学出版社,2002.
[98]L.Bruzzone,F.Roli and S.B.Serpico.Structure Neural Networks for Signal Classification.Signal Processing,1998,64:271-290.
[99]Steven K.Rogers,John M.Colombi,Curtis E.Martin,etc.Neural Networks for Automatic Target Recognition.Neural Networks,1995,8(7):1153-1184.
[100]Krzysztof J.Cios,Inbo Shin.Image Recognition Neural Network:IRNN.Neurocomputing,1995,(7):159-185.
[101]A.Ravichandran,B.Yegnanarayana.Studies on Object Recognition From Degraded Images Using Neural Networks.Neural Networks,1995,8(3):481-488.
[102]J.Antonio Catalan,J.S.Jin and T.Gedeon.Reducing the Dimensions of Texture Features for Image Retrieval Using Multilayer Neural Networks.Pattern Analysis & Applications,1999,(2):196-203.
[103]D.de Ridder,R.P.W.Duin,RW.Verbeek and L.J.van Vliet.The Applicability of Neural Networks to Non-linear Image Processing.Pattern Analysis &Applications,1999,(2):111-128.
[104]A.Datta,S.Pal and S.Chakraborti.Image Thinning by Neural Networks.Neural Comput & Applications,2002,(11):122-128.
[105]S.Singh,A.Amin.Neural Network Recognition of Hand-printed Characters.Neural Comput & Applications,1999,(8):67-76.
[106]Mandana Ebadian Dehkordi,Nasser Sherkat,Tony Allen.Handwriting Style Classification.IJDAR,2003,(6):55-74.
[107]E.A.Faclhel and P.Bhattacharyya.Application of a Steerable Wavelet Transform Using Neural Network for Signature Verification.Pattern Analysis &Applications,1999,(2):184-195.
[108]罗发龙,李衍达著.神经网络信号处理.北京:电子工业出版社,1993.
[109]王志宏,徐建生.BP神经网络结构设计的频谱取值模型研究.武汉化工学院学报,2001,23(9):63-65.
[110]邓先礼编著.最优化技术.重庆:重庆大学出版社,1998.
[111]粟塔山等编著.最优化计算原理与算法程序设计.长沙:国防科技大学出版社,2001.
[112]E.Hughlett and P.Kaiser.An Autofocus Technique for Imaging Microscopy.ICASSP-92,1992,(3):93-96.
[113]Ooi K.Izumi,K.Nozaki,M.Takeda.An Advanced Auto-Focus System for Video Camera Using Quasi Condition Reasoning.IEEE International Conference on Consumer Electronics,1990:348-349.
[114]羊国光,宋菲君.高等物理光学.北京:中国科技大学出版社,1991.
[115]向阳,向才新.部分相干光学成像系统的光学传递函数和像质评价标准.光学学报,2002,22(11).
[116]http://www.33tt.com/htmldata/2004_03/1/29/article_1_1.htm.
[117]http://www.oecr.com/bbs/cgi-bin/topic.cgi?forum=3&topic=39.
[118]http://www.cgan.net/book/books/print/scanner/7.2.htm.
[119]http://www.dophoto.com/new_page__66.htm.
[120]http://www.ehongnet.com/knowledg/mtftran.htm.
[121]徐之海,李奇.现代成像系统.北京:国防工业出版社,2001.
[122]http://www.netdzb.com/pav/hotn ews_asp/ShowMessage.asp?MsgID=1699.
[123]http://bbs.dpsalon.com/cgi-bin/topic.cgi?fo rum=10&topic=39.
[124]Xilinx Incorporation.Spartan-IIE 1.8V FPGA Family:Complete Data Sheet,2004.
[125]王诚,薛小刚,钟信潮.FPGA/CPLD设计工具Xilinx ISE5.x使用详解.北京:人民邮电出版社,2003.
[126]潘松,黄继业.EDA技术实用教程.北京:科学出版社,2002.
[127]http://www.edacn.net/cgi-bin/topic.cgi? forum=9&topic=13.
[128]谭建成.电机控制专用集成电路.北京:机械工业出版社,1999.
[129]杜坤梅,李铁才.电机控制技术.哈尔滨:哈尔滨工业大学出版社.
[130]任治锦.电机与电气控制.北京:机械工业出版社,2002.
[131]黄立培.电动机控制.北京:清华大学出版社,2003.
[132]胡广书.数字信号处理--理论、算法与实现.北京:清华大学出版社,1997.
[133]贝斯[著],刘凌,胡永生[译].数字信号处理的FPGA实现.北京:清华大学出版社,2003.
[134]戴明桢.数字信号处理的硬件实现.北京:航空工业出版社,1998.
[135]辛春艳.VHDL硬件描述语言.北京:国防工业出版社,2002.
[136]徐志军,徐光辉.CPLD/FPGA的开发与应用.北京:电子工业出版社,2002.
[137]http://www.mtxindia.com/Anl.htm.
[138]Philips Semiconductors Incorporation.SAA7114 PAL/NTSC/SECAM Video Decoder,2004.
[139]http://www17.cds.ne.jp/～stray/ccir656.html#Contents.
[140]http://yekaionline.wx-e.com/fordesigner/analogtvfundamentall.htm.
[141]广州周立功单片机发展有限公司.I~2C总线规范.
[142]Stephanie Tapp.Xilinx配置快速入门指南,2003.
[143]陈贺新.非线性滤波器与数字图像处理.北京:国防工业出版社,1997.
[144]瞿蓬,林喜荣.一种基于图像处理的自动调焦系统.电子技术应用,2002,(10):33-35.
[145]胡炳梁,曹剑中,熊仁生等.变焦距镜头组的自适应调焦的实现.光子学报,2003,32(8):1004-1006.
[146]郑春阳,李胜利,徐敏.采用步进电机的CCD自动聚焦控制系统.微特电机,2004,(2):29-30.
[147]陶洪久,饶俊飞,周祖德.单幅图像的超分辨率重建方法.武汉理工大学学报,2004,28(6):943-946.
[148]黄桂灶,徐向东,陆祖康等.单片机处理信号的自动聚焦系统.光电工程,2002,29(1):42-44.
[149]周礼,王章野,金剑秋等.基于HVS的小波图像融合新算法.中国图象图形学报,2004,9(9):1088-1094.
[150]李振华,敬忠良,孙韶媛等.基于不可分离小波框架变换的多聚焦图像融合算法.上海交通大学学报,2005,39(4):557-560.
[151]那彦,殷廷瑞,杨万海.基于成像机理的小波包变换多聚焦图像融合.应用光学,2004,25(2):51-54.
[152]张新曼,韩九强.基于对比度视觉模型的多聚焦融合图像的精确重构.小型微型计算机系统,2004,25(11):1995-1997.
[153]任四刚,李见为,谢利利.基于灰度差分法的自动调焦技术.光电工程,2003,30(2):53-55.
[154]韩华,王洪剑,彭思龙.基于局部结构相似性的单幅图像超分辨率算法.计算机辅助设计与图形学学报,2005,17(5):941-947.
[155]朱倩,陈金华,李峰.基于图像分析的数码相机对焦技术研究.江苏大学学报,2003,24(4):61-64.
[156]赵天昀,徐振强,李正文.基于小波变换的多聚焦图像融合方法.计算机应用,2004,24:157-159.
[157]汪雪林,文伟,彭思龙.基于小波域局部高斯模型的图像超分辨率.中国图象图形学报,2004,9(8):941-946.
[158]白立芬,徐毓娴,于水,李庆祥.基于图像处理的显微镜自动调焦方法研究.仪器仪表学报,1999,20(6):612-614.
[159]李国新,王国宇,王汝霖等.基于自动聚焦算法的多聚焦图像融合.计算机应用研究,2005,(3):166-168.
[160]王卫卫,水鹏朗,宋国乡.小波域多聚焦图像融合算法.系统工程与电子技术,2004,20(5):668-671.
[161]梁音,柳建,田金文等.一种改进的图像超分辨率模型与数字图像分辨率的研究.计算机工程与应用,2004,(26):59-61.
[162]黄卉,檀结庆.一种新的基于清晰度的多聚焦图像融合规则.计算机工程与应用,2004,(14):51-52.
[163]郑春阳,李永新,李胜利.用于CCD图像采集的自动聚焦控制系统设计.测控技术,2004,23(5):51-53.
[164]田宜彬,徐之海,冯华君.用于数码相机的瞳孔控制自动对焦技术.光学技 术,2003,29(1):53-55.
[165]王任华,沈忙作.自动对焦算法研究.光电工程,2000,27(4):11-13.
[166]Chen Guojin,Zhang Kesong,Gong Ye,Xu Guohua,Shi Huli.Research on Image Definition Function Using in Real Time Auto-focusing System.Proceedings of Fifth International Symposium on Instrumentation and Control Technology,2003:466-470.
[167]Guojin Chen,Miaofen Zhu,Liming Guan,Kesong Zhe,Guohua Xu,Huli Shi.Study on Contrast Evaluation Function of CMOS Digital Camera.Proceedings of the 2005 IEEE/ASME International Conference on Advanced Intelligent Mechatronics,2005:379-383.
[168]裘晓光,陈国金.基于数字图像边缘信息的自动对焦算法研究.杭州电子科技大学学报,2004,24(4):91-94.
[169]杨文鹏,陈国金.基于OMAP的数字调焦控制系统的实现.机电工程,2004,21(11):25-28.
[2]沈庭芳,方子文编著.数字图像处理及模式识别.北京:北京理工大学出版社,1998.
[3]夏德深,傅德盛。现代图像处理技术与应用.南京:东南大学出版社.1997.
[4]Kenneth.R.Castleman著,朱志刚等译.数字图像处理.北京:电子工业出版社,1998.
[5]王润生编著.图像理解.长沙:国防科技大学出版社,1995.
[6]黄其昆.自动调焦二十年.大众摄影,2001,(5):52-55.
[7]董太和.相机的MF/AF机构及其精度:相机机构专题讲座(1).照相机,2000,(2).
[8]许盛,李见为.高级显微镜系统智能化自动调焦的研究及实现.光学仪器,2000,22(5):22-26.
[9]王建民,浦昭邦,尹继学,晏磊.图像式显微自动调焦瞄准系统的研究.光学技术,2000,26(5):410-417.
[10]许盛,李见为.基于智能化自动调焦的高级显微镜系统.光电工程,2000,27(1):33-36.
[11]一川.中国应开发AF相机.照相机,2000,(2):8-10.
[12]徐之海,李奇.现代成像系统.北京:国防工业出版社,2001.
[13]白磊.CCD显微测量系统自动调焦技术的研究.哈尔滨工业大学硕士学位论文,2002.
[14]H.Harms and H.M.Aus.Comparison of Digital Focus Criteria for a TV Microscope Systeml.Cytometry,1984,(5):236-243.
[15]Mason DC,Green DK.Automated Forusing of a Computer Controlled Microscopel.IEEE Trans Biomed Eng BME,1975,22:137-150.
[16]黄剑琪.基于频谱分析的数字对焦技术研究.浙江大学硕士学位论文,2001.
[17]M.Subbarao,T.S.Choi,and A.Nikzad.Focusing Techniques.Optical Engineering,1993,32(11):2824-2836.
[18]E.Krotkov.Focusing.International Journal of Computer Vision,1987,(1):223-237.
[19]Yalin Xiong and Steven A.Shafer.Depth from Focusing and Defocusing.Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1992.
[20]M.Subbarao,T.Choi,and A.Nikzad.Focusing Techniques.J.Optical Eng.,1993,32(11):2824-2836.
[21]S.Jutamulia,T.Asakura,R.D.Bahuguna,P.C.De.Guzman.Autofocusing Based on Power-Spectra Analysis.Appl.Optics,1994,33(26):6210-6212.
[22]Murali Subbarao,Jenn-Kwei Tyan.Selecting the Optimal Focus Measure for Autofocusing and Depth-From-Focus.IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(8):861-870.
[23]曹茂永,孙农亮,郁道银.基于灰度梯度的数字图像评价函数.光电工程,2003,30(4):69-72.
[24]李奇,冯华君,徐之海.数字图像清晰度评价函数研究.光子学报,2002,31(6):376-378.
[25]M.Subbarao,J.K.Tyan.The Optimal Focus Measure for Passive Autofocusing and Depth-from-Focus.Proceeding of SPIE Conference in Video Metrics,1995.
[26]Biberman V.and Konstantinos K.Image Quality In Perception of Displayed Information.New York:Plenum Press,1997.
[27]赵荣椿,赵忠明.数字图像导论.西安:西北工业大学出版社,1996.
[28]杨枝灵,王开等.数字图像获取、处理及实践应用.北京:人民邮电出版社,2003.
[29]王新成.高级图像处理技术.北京:中国科学技术出版社,2001.
[30]Chen Guojin,Liu Bin,Gong Ye,Xu Guohua,Shi Huli.Research on Image Auto-focus System Based on Neural Network.Proceedings of Fifth International Symposium on Instrumentation and Control Technology,2003:461-465.
[31]Chen Guojin,Zhu Miaofen,Guan Liming,Liu Bin.Implementing the Auto-focusing System Based on the Digital Image Process Technology.Proceedings of IEE Irish Signals and Systems Conference,2005:92-95.
[32]郁道银,谈恒英.工程光学.北京:机械工业出版社,1999.
[33]夏良正.数字图像处理.南京:东南大学出版社,1999.
[34]李智勇,沈振康.一种单镜头运动测距方法.电子学报,1990,18(5):70-74.
[35]Yoshida Y,Shinohara S,Ikeda H,Tada K,Yoshida H,Nishide K,Sum M.Control of Lens Position in Auto-focus Cameras Using Semiconductor Laser Range Finder.Circuits and Systems,1991,(1):395-398.
[36]Getzlaff S,Schreiter J,Graupner A,Schuffny R.A System-on-chip Realization of a CMOS Image Sensor with Programmable Analog Image Preprocessing.The 2001 IEEE International Symposium on Circuits and Systems,2001(4):486-489.
[37]Sang Ku Kim,Sang Rae Park.Simultaneous Out-of-focus Blur Estimation and Restoration for a Digital Auto-focusing System.IEEE Transaction on Consumer Electronics,1998,44(3):1071-1075.
[38]Sang Ku Kim,Joon Ki Paik.Out-of-focus Blur Estimation and Restoration for Digital Auto-focusing System.Electronics Letters,1998,34(12):1217-1219.
[39]Chung Nam Cho,etc.Digital Auto-focusing of Multiple Objects Based on Image Restoration.Proceedings of SPIE-THE International Society for Optical Engineering,2000,20(23):1642-1651.
[40]Dick Merrill.The Next-generation Digital Camera.Optics & Photonies News,1992,(1):26-49.
[41]侯雨石,何玉青,陈永飞,张忠廉.数码相机图像传感器技术.光学技术2003,29(1):59-62.
[42]D.Marr and E.Hildreth.Theory of Edge Detection.London:Proc.R.Soc.,1980.
[43]F.Bergholm.Edge Focusing.IEEE Trans.Patt.Anal.Mach.Intell.,1987,9(6):726-741.
[44]Canny J.A Computational Approach to Edge Detection.IEEE Trans.PAMI, 1986,8(6):679-698.
[45]程佩青.数字信号处理教程.北京:清华大学出版社,1998.
[46]杨长生编著.图像与声音压缩技术.杭州:浙江大学出版社,2000.
[47]ISO/IEC JTC1/SC29 WG1.JPEG 2000 Part Ⅰ Final Committee Draft Version 1.0.2000.
[48]曹茂永,孙农亮.离焦模糊图像清晰度评价函数的研究.仪器仪表学报,2001,22(3):426-430.
[49]Eli Peli.Contrast Sensitivity Function and Image Discrimination.J.Opt.Soc.Am.A,2000,18(2):283-293.
[50]Brenner.J.F,Dew,B.S.Horton,J.B.Neirath.An Automated Microscope for Cytologic Research.J.Histochem.Cytochem.,1971,24:100-111.
[51]Jan-Mark Geusebroek,Frans Cornelissen,Arnold W.M.Smeulders and Hugo Geerts.Robust Autofocusing in Microscopy.Cytometry,2000,36(1):1-9.
[52]黄剑琪,冯华君,徐之海,戴顺妹,李奇,边美娟.边缘特征增强算法和小波分析在精确聚焦中的应用.光子学报,2000,29(10):932-936.
[53]白立芬,于水,薛实福,李庆祥.基于图像处理的自动调焦方法中清晰度判断函数的研究.仪器仪表学报,1999,20(4):85-89.
[54]郑南宁.计算机视觉与模式识别.北京:国防工业出版社,1998.
[55]吴振峰,左洪福,邱根良.光学显微镜自动调焦的技术研究.光学仪器,2000,22(4):9-12.
[56]Vollath D.Automatic Focusing by Correlative Methods.J.Microsc.,1987,147:279-288.
[57]Xiao Tang,Pierre L'Hostis,and Yu Xiao.An Auto-focusing Method in a Microscopic Test Bed for Optical Discs.Journal of Research of the National Institute of Standards and Technology,2000,105(4):565-569.
[58]D.G.Marimont and Y.Rubner.A Probabilistic Framework for Edge Detection and Scale Selection.The Sixth International Conference on Computer Vision,1998,4(7):207-214.
[59]Kang-Sun Choi,Jun-Suk Lee,and Sun-Jae Ko.New Autofocusing Technique Using the Frequency Selective Weighted Median Filter for Video Cameras.IEEE Transactions on Consumer Electronics,1999,45.
[60]张晓森.基于概率的字符识别系统的若干问题研究.西安电子科技大学硕士学位论文.
[61]邢彦辰.基于局部熵差的图象匹配方法在图象实时监测中应用研究.哈尔滨工业大学硕士学位论文,2003.
[62]东秀阁.基于信息熵的图像分割方法研究及其在DSP中的快速实现.天津大学硕士学位论文.
[63]A.Santos,C.Ortizde Solorzano,J.J.Vaquero,J.M.Pena,N.Malpica and F.Del Pozo.Evaluation of Autofocus Functions in Molecular Autogenetic.Journal of Microscopy,1997,188(3):264-272.
[64]Firestone,L.Cook,K.Culp,K.Talsania,N.Preston.Comparison of Autofocus Methods for Automated Microscopy.Cytometry,1991,12:195-206.
[65]Vollath D.The Influence of the Scene Parameters and Noise on the Behaviour of Automatic Focusing Algorithms.J.Microsc.,1988,151:133-146.
[66]N.B.Nill and B.H.Bouzas.Objective Image Quality Measures Derived From Digital Image Power Spectra.Optical Engineering,1992,31(4):813-825.
[67]K.Falconer.Fractal Geometry - Mathematical Foundations and Applications.New York:Wiley,1990.
[68]盛骤,谢式千,潘承毅编.概率论与数理统计.北京:高等教育出版社,2000.
[69]D.J.Granrath.The Role of Human Visual Models in Image Processing.Proc.IEEE,1981,69(5):552-561.
[70]N.B.Nill.A Visual Model Weighted Cosine Transform for Image Compression and Quality Assessment.IEEE Trans.Commun,1985,33(6):551-557.
[71]J.J.DePalma and E.M.Lowry.Sine Wave Response of the Visual System,(Ⅱ).Sine Wave and Square Wave Contrast Sensitivity.J.Opt.Soc.Am,1962,52(3):328-335.
[72]J.L.Mannos and D.J.Sakrison.The Effects of a Visual Fidelity Criterion on the Encoding of Images.IEEE Trans.Inform.Theory,1974,20(4):525-536.
[73]K.N.Ngan,K.S.Leong,and H.Singh.Adaptive Cosine Transform Coding of Images in Perceptual Domain.IEEE Trans.Acoust.Speech Signal Process,1989,37(11):1743-1750.
[74]张兆礼,赵春晖,梅晓丹著.现代图像处理技术及Matlab实现.北京:人民邮电出版社,2001.
[75]寿天德.视觉信息处理的脑机制.上海:上海科技教育出版社,1997.
[76]杨雄里.视觉的神经机制.上海:上海科学技术出版社,1996.
[77]徐广第.眼科屈光学.北京:军事医学科学出版社,2001.
[78]刘健勤,盛津芳,魏敏.面向智能体的视觉神经信息处理.北京:科学出版社,2000.
[79]程正兴著.小波分析算法与应用.西安:西安交通大学出版社,1998.
[80]王建平,盛军,朱程辉.基于小波分析的视频图像字符特征提取方法研究.微电子与计算机,2002,(5):51-56.
[81]杨胜波,于春海.小波分析在生物医学信号/图像处理中的应用.仪器仪表学报,2002,23(3):179-181.
[82]孟军,魏同立,吴金,常昌远.数字图像离散小波变换的原理与硬件实现分析.东南大学学报,2002,32(6):1-6.
[83]韦志辉,程军.基于小波变换的一种新的图像质量评价方法.南京理工大学学报,1998,22(6):552-555.
[84]甘斌,张雄伟,甘仲民.基于小波变换的多尺度图像边缘处理.数字电视与数字视频,2001,(8):20-22.
[85]黄应清,季向琦,张智诠.图像小波系数的分布规律及其对图像质量的影响.红外与激光工程,2000,29(3):15-18.
[86]张新明,沈兰荪.基于小波和统计特性的自适应图像增强.信号处理,2001,17(3):227-231.
[87]E.A.Fadhel and P.Bhattacharyya.Application of a Steerable Wavelet Transform Using Neural Network for Signature Verification.Pattern Analysis &Applications,1999,(2):184-195.
[88]B.J.W.Fleming,D.Yu,R.G.Harrison,and D.Jubb.Wavelet-based Detection of Coherent Structures and Self-affnity in Financial Data.J.Eur.Phys.,2001,(20): 543-546.
[89]Z.Peng,Y.He,Q.Lu and F.Chu.Feature Extraction of the Rub-impact Rotor System by Means of Wavelet Analysis.Journal of Sound and Vibration,2003,259(4):1000-1010.
[90]M.Antonopoulos-Domis,T.Tambouratzis.System Identification,A Transient via Wavelet Multiresolution Analysis.Ann.Nucl.Energy,1998,25(6):465-480.
[91]I.Urriza,J.I.Artigas,L.A.Barragan,J.I.Garcia and D.Navarro.VLSI Implementation of Discrete Wavelet Transform for Lossless Compression of Medical Images.Real-Time Imaging,2001,(7):203-217.
[92]边肇祺,张学工等编著.模式识别.北京:清华大学出版社,2000.
[93]Zhengxin Hou,Wenguan Wang and Jinghui Chu.A Kind of 2D Isotropic Wavelet and Its Application in Image Primitive Detection.Journal of Circuits Systems and Computers,2002,11(2):123-135.
[94]黄德双编著.神经网络模式识别系统理论.北京:电子工业出版社,1996.
[95]焦李成编著.神经网络的应用与实现.西安:西安电子科技大学出版社,1996.
[96]Abhijit S.Pandya,Robert B.Macy著,徐勇,荆涛等译.神经网络模式识别及应用.北京:电子工业出版社,1999.
[97]杨建刚编著.人工神经网络实用教程.杭州:浙江大学出版社,2002.
[98]L.Bruzzone,F.Roli and S.B.Serpico.Structure Neural Networks for Signal Classification.Signal Processing,1998,64:271-290.
[99]Steven K.Rogers,John M.Colombi,Curtis E.Martin,etc.Neural Networks for Automatic Target Recognition.Neural Networks,1995,8(7):1153-1184.
[100]Krzysztof J.Cios,Inbo Shin.Image Recognition Neural Network:IRNN.Neurocomputing,1995,(7):159-185.
[101]A.Ravichandran,B.Yegnanarayana.Studies on Object Recognition From Degraded Images Using Neural Networks.Neural Networks,1995,8(3):481-488.
[102]J.Antonio Catalan,J.S.Jin and T.Gedeon.Reducing the Dimensions of Texture Features for Image Retrieval Using Multilayer Neural Networks.Pattern Analysis & Applications,1999,(2):196-203.
[103]D.de Ridder,R.P.W.Duin,RW.Verbeek and L.J.van Vliet.The Applicability of Neural Networks to Non-linear Image Processing.Pattern Analysis &Applications,1999,(2):111-128.
[104]A.Datta,S.Pal and S.Chakraborti.Image Thinning by Neural Networks.Neural Comput & Applications,2002,(11):122-128.
[105]S.Singh,A.Amin.Neural Network Recognition of Hand-printed Characters.Neural Comput & Applications,1999,(8):67-76.
[106]Mandana Ebadian Dehkordi,Nasser Sherkat,Tony Allen.Handwriting Style Classification.IJDAR,2003,(6):55-74.
[107]E.A.Faclhel and P.Bhattacharyya.Application of a Steerable Wavelet Transform Using Neural Network for Signature Verification.Pattern Analysis &Applications,1999,(2):184-195.
[108]罗发龙,李衍达著.神经网络信号处理.北京:电子工业出版社,1993.
[109]王志宏,徐建生.BP神经网络结构设计的频谱取值模型研究.武汉化工学院学报,2001,23(9):63-65.
[110]邓先礼编著.最优化技术.重庆:重庆大学出版社,1998.
[111]粟塔山等编著.最优化计算原理与算法程序设计.长沙:国防科技大学出版社,2001.
[112]E.Hughlett and P.Kaiser.An Autofocus Technique for Imaging Microscopy.ICASSP-92,1992,(3):93-96.
[113]Ooi K.Izumi,K.Nozaki,M.Takeda.An Advanced Auto-Focus System for Video Camera Using Quasi Condition Reasoning.IEEE International Conference on Consumer Electronics,1990:348-349.
[114]羊国光,宋菲君.高等物理光学.北京:中国科技大学出版社,1991.
[115]向阳,向才新.部分相干光学成像系统的光学传递函数和像质评价标准.光学学报,2002,22(11).
[116]http://www.33tt.com/htmldata/2004_03/1/29/article_1_1.htm.
[117]http://www.oecr.com/bbs/cgi-bin/topic.cgi?forum=3&topic=39.
[118]http://www.cgan.net/book/books/print/scanner/7.2.htm.
[119]http://www.dophoto.com/new_page__66.htm.
[120]http://www.ehongnet.com/knowledg/mtftran.htm.
[121]徐之海,李奇.现代成像系统.北京:国防工业出版社,2001.
[122]http://www.netdzb.com/pav/hotn ews_asp/ShowMessage.asp?MsgID=1699.
[123]http://bbs.dpsalon.com/cgi-bin/topic.cgi?fo rum=10&topic=39.
[124]Xilinx Incorporation.Spartan-IIE 1.8V FPGA Family:Complete Data Sheet,2004.
[125]王诚,薛小刚,钟信潮.FPGA/CPLD设计工具Xilinx ISE5.x使用详解.北京:人民邮电出版社,2003.
[126]潘松,黄继业.EDA技术实用教程.北京:科学出版社,2002.
[127]http://www.edacn.net/cgi-bin/topic.cgi? forum=9&topic=13.
[128]谭建成.电机控制专用集成电路.北京:机械工业出版社,1999.
[129]杜坤梅,李铁才.电机控制技术.哈尔滨:哈尔滨工业大学出版社.
[130]任治锦.电机与电气控制.北京:机械工业出版社,2002.
[131]黄立培.电动机控制.北京:清华大学出版社,2003.
[132]胡广书.数字信号处理--理论、算法与实现.北京:清华大学出版社,1997.
[133]贝斯[著],刘凌,胡永生[译].数字信号处理的FPGA实现.北京:清华大学出版社,2003.
[134]戴明桢.数字信号处理的硬件实现.北京:航空工业出版社,1998.
[135]辛春艳.VHDL硬件描述语言.北京:国防工业出版社,2002.
[136]徐志军,徐光辉.CPLD/FPGA的开发与应用.北京:电子工业出版社,2002.
[137]http://www.mtxindia.com/Anl.htm.
[138]Philips Semiconductors Incorporation.SAA7114 PAL/NTSC/SECAM Video Decoder,2004.
[139]http://www17.cds.ne.jp/～stray/ccir656.html#Contents.
[140]http://yekaionline.wx-e.com/fordesigner/analogtvfundamentall.htm.
[141]广州周立功单片机发展有限公司.I~2C总线规范.
[142]Stephanie Tapp.Xilinx配置快速入门指南,2003.
[143]陈贺新.非线性滤波器与数字图像处理.北京:国防工业出版社,1997.
[144]瞿蓬,林喜荣.一种基于图像处理的自动调焦系统.电子技术应用,2002,(10):33-35.
[145]胡炳梁,曹剑中,熊仁生等.变焦距镜头组的自适应调焦的实现.光子学报,2003,32(8):1004-1006.
[146]郑春阳,李胜利,徐敏.采用步进电机的CCD自动聚焦控制系统.微特电机,2004,(2):29-30.
[147]陶洪久,饶俊飞,周祖德.单幅图像的超分辨率重建方法.武汉理工大学学报,2004,28(6):943-946.
[148]黄桂灶,徐向东,陆祖康等.单片机处理信号的自动聚焦系统.光电工程,2002,29(1):42-44.
[149]周礼,王章野,金剑秋等.基于HVS的小波图像融合新算法.中国图象图形学报,2004,9(9):1088-1094.
[150]李振华,敬忠良,孙韶媛等.基于不可分离小波框架变换的多聚焦图像融合算法.上海交通大学学报,2005,39(4):557-560.
[151]那彦,殷廷瑞,杨万海.基于成像机理的小波包变换多聚焦图像融合.应用光学,2004,25(2):51-54.
[152]张新曼,韩九强.基于对比度视觉模型的多聚焦融合图像的精确重构.小型微型计算机系统,2004,25(11):1995-1997.
[153]任四刚,李见为,谢利利.基于灰度差分法的自动调焦技术.光电工程,2003,30(2):53-55.
[154]韩华,王洪剑,彭思龙.基于局部结构相似性的单幅图像超分辨率算法.计算机辅助设计与图形学学报,2005,17(5):941-947.
[155]朱倩,陈金华,李峰.基于图像分析的数码相机对焦技术研究.江苏大学学报,2003,24(4):61-64.
[156]赵天昀,徐振强,李正文.基于小波变换的多聚焦图像融合方法.计算机应用,2004,24:157-159.
[157]汪雪林,文伟,彭思龙.基于小波域局部高斯模型的图像超分辨率.中国图象图形学报,2004,9(8):941-946.
[158]白立芬,徐毓娴,于水,李庆祥.基于图像处理的显微镜自动调焦方法研究.仪器仪表学报,1999,20(6):612-614.
[159]李国新,王国宇,王汝霖等.基于自动聚焦算法的多聚焦图像融合.计算机应用研究,2005,(3):166-168.
[160]王卫卫,水鹏朗,宋国乡.小波域多聚焦图像融合算法.系统工程与电子技术,2004,20(5):668-671.
[161]梁音,柳建,田金文等.一种改进的图像超分辨率模型与数字图像分辨率的研究.计算机工程与应用,2004,(26):59-61.
[162]黄卉,檀结庆.一种新的基于清晰度的多聚焦图像融合规则.计算机工程与应用,2004,(14):51-52.
[163]郑春阳,李永新,李胜利.用于CCD图像采集的自动聚焦控制系统设计.测控技术,2004,23(5):51-53.
[164]田宜彬,徐之海,冯华君.用于数码相机的瞳孔控制自动对焦技术.光学技 术,2003,29(1):53-55.
[165]王任华,沈忙作.自动对焦算法研究.光电工程,2000,27(4):11-13.
[166]Chen Guojin,Zhang Kesong,Gong Ye,Xu Guohua,Shi Huli.Research on Image Definition Function Using in Real Time Auto-focusing System.Proceedings of Fifth International Symposium on Instrumentation and Control Technology,2003:466-470.
[167]Guojin Chen,Miaofen Zhu,Liming Guan,Kesong Zhe,Guohua Xu,Huli Shi.Study on Contrast Evaluation Function of CMOS Digital Camera.Proceedings of the 2005 IEEE/ASME International Conference on Advanced Intelligent Mechatronics,2005:379-383.
[168]裘晓光,陈国金.基于数字图像边缘信息的自动对焦算法研究.杭州电子科技大学学报,2004,24(4):91-94.
[169]杨文鹏,陈国金.基于OMAP的数字调焦控制系统的实现.机电工程,2004,21(11):25-28.