由于遥感影像本身固有的特性和外界环境成像条件的不同，使得有的图像存在对比度差、图像边缘模糊、噪声较大等缺点。在土地资源利用调查项目中，为了有效地提高内、外业的工作效率，保证成果质量，需要对遥感图像中的噪声或者视觉效果不好的图像进行图像增强处理，达到去除噪声、图像平滑、增大对比度，改善视觉效果的目的。 本文结合土地利用调查的项目，以甘肃省某县的遥感图像为对象，在分析图像空间域变换和频率域变换算法的基础上，通过编写程序实现了遥感图像在空间域与频率域几种常用增强算法的研究、应用及对比。本文主要研究内容如下： (1)空间域图像处理算法的研究及应用 应用点运算、几何运算对图像实现增强处理。 (2)图像频率域变换算法的介绍 傅里叶变换理论是遥感图像频域滤波增强算法与理论的基础，把图像从空间域变换到频率域，有利于解决在空间域中无法处理的问题。 (3)图像增强算法的应用 分别研究了空间域中值滤波图像增强算法和频率域图像增强算法，对空间域中值滤波图像增强算法进行了实例分析。 (4)频率域图像增强算法的研究及应用 通过程序提取图像的频谱信息，获得有效的频率范围，以此为根据设置截止频率参数，并选取不同的滤波函数，对图像进行低通滤波和高通滤波分析，实现遥感图像在频率域中的图像增强算法及应用。介绍了改进后的同态滤波在频率域中的滤波处理，并提出了利用同态滤波器函数中的参数来作为高通滤波和低通滤波的判别标准。对比分析改进后的同态滤波器与基本的频率域滤波器图像增强后的效果，认为改进后的同态滤波器是一种比较理想的频率域滤波器，对提高土地调查中的工作效率和质量具有重要的意义。 通过频域滤波增强算法处理的遥感影像在实际中被多次检核，频率域图像处理的方法具有重要的理论参考和实际应用价值。所以，通过理论分析和实验验证得出，高空间 分辨率遥感图像频域滤波增强算法可以为进一步的地物识别、判断以及矢量化奠定良好的基础，且算法简单，也有利于实时实现，符合现代数字图像处理的趋势。

As the inherent characteristics of remote sensing and imaging conditions of different external environment, making some images with poor contrast, fuzzy edges and big noise disadvantage.In the survey of land use projects, in order to improve the inside and outside job effectivel and ensure quality resultsy, need to enhancement image processing for remote sensing image's noise or poor visual images to remove noise,image smoothing , increase contrast and improve the visual effects purposes. In this paper combine with the investigation of land use projects, use county in Gansu Province as the object of remote sensing images, in the base of analysis the image spatial domain transform and frequency domain transform algorithm, through programming achieved remote sensing image in spatial domain and frequency domain of several common enhancement algorithm, application,and comparison. This mainly includes the following: (1)Spatial domain image processing algorithms and its application Application point operations, geometry operations to achieve enhanced image processing. (2)Image frequency domain transform algorithm description Fourier Transform theory is the basis of frequency filter enhancement algorithm and theory for remote sensing image, transform the image from space domain into the frequency domain, is conducive to resolving the problem can not be processed in space domain. (3)Application of image enhancement algorithm Discussed the spatial domain median filtering's image enhancement algorithm and frequency domain's image enhancement algorithm,and analyse spatial domain median filtering's image enhancement algorithm. (4)Frequency domain image enhancement algorithms and its application Through the program extract the frequency information of image,access to effective frequency range,this basis set the cutoff frequency parameters,and select a different filter function, to analyse the image in low-pass filtering and high-pass filtering,realize the remote sensing images in the frequency domain's image enhancement algorithm and its application.Introduced improved homomorphic filtering in the frequency domain filtering,and recommendations on using filter function with the parameters of state to a high-pass filter and low-pass filter's criteria.Compared with the improved homomorphic filtering and the basic frequency domain filtering after the effect of image enhancement,consider improved homomorphic filtering as an ideal frequency domain filtering,to improve the work of land survey efficiency and quality is of great significance. By frequency domain filtering processing of remote sensing image enhancement algorithm was repeated in the actual check, the frequency domain image enhancement method has important theoretical and practical value for reference. Therefore, through the theoretical analysis and experimental verification obtained, high spatial resolution remote sensing image enhancement algorithm in frequency domain filtering can make a good function in the further to identify the surface features, judge and vector, the algorithm is simple and is also beneficial to realize,in accordance with the trend of modern digital image processing.