基于脊波变换的医学图像增强研究
发布时间：2018-12-16 06:13
【摘要】：随着科学技术的发展,特别是医疗技术和图像成像技术的发展,获取各类医学图像不再困难。医学图像在现代临床医学中占有相当重要的地位,临床医生可以利用医学图像对病情有效地进行诊断和分析。但由于受到数据采集设备、和噪声等因素的影响,医学图像的质量会受到影响。为了更好地协助临床医生的工作,为其提供更清晰的医学图像,医学图像增强处理成为医学处理中的一个重要问题。本文围绕医学图像增强的方法展开分析和研究,结合图像稀疏表示和多小波变换理论,提出了一种基于脊波变换的医学图像增强算法。本文主要工作如下:首先研究了医学图像的空间域增强方法。空间域法是直接对原图像的灰度进行处理,主要有灰度变换、直方图均衡化,以及空间域滤波等增强方法。然后研究了小波变换理论和脊波变换理论,并在此基础上,提出了一种基于脊波变换的医学图像增强方法。该算法主要过程为:首先对图像进行有限Radon变换,然后再进行一维多小波变换,接着对多小波变换系数进行增强处理,并在此基础上对医学图像进行重构处理。最后,对重构后的图像进行灰度变换,以得到增强效果更好的图像。实验结果表明,使用该算法对医学图像进行增强处理,能够获取良好的增强效果。
[Abstract]:With the development of science and technology, especially medical technology and image imaging technology, it is no longer difficult to obtain all kinds of medical images. Medical images play an important role in modern clinical medicine. Clinicians can use medical images to diagnose and analyze the disease effectively. However, the quality of medical images will be affected by data acquisition equipment, noise and other factors. In order to better assist clinicians in their work and provide them with clearer medical images, medical image enhancement has become an important issue in medical processing. In this paper, a medical image enhancement algorithm based on ridgelet transform is proposed by analyzing and researching the methods of medical image enhancement, combining the sparse representation of image and the theory of multi-wavelet transform. The main work of this paper is as follows: firstly, the spatial domain enhancement method of medical image is studied. The spatial domain method is used to deal with the gray level of the original image directly, including gray level transformation, histogram equalization, spatial filtering and other enhancement methods. Then, wavelet transform theory and ridgelet transform theory are studied, and a medical image enhancement method based on ridgelet transform is proposed. The main process of the algorithm is as follows: firstly, the image is transformed by finite Radon, then by one dimensional multiwavelet transform, then the coefficients of multiwavelet transform are enhanced, and then the medical image is reconstructed. Finally, the reconstructed image is transformed into gray level to get a better enhancement effect. The experimental results show that the algorithm can achieve good enhancement effect.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41
，
本文编号：2381922
[Abstract]:With the development of science and technology, especially medical technology and image imaging technology, it is no longer difficult to obtain all kinds of medical images. Medical images play an important role in modern clinical medicine. Clinicians can use medical images to diagnose and analyze the disease effectively. However, the quality of medical images will be affected by data acquisition equipment, noise and other factors. In order to better assist clinicians in their work and provide them with clearer medical images, medical image enhancement has become an important issue in medical processing. In this paper, a medical image enhancement algorithm based on ridgelet transform is proposed by analyzing and researching the methods of medical image enhancement, combining the sparse representation of image and the theory of multi-wavelet transform. The main work of this paper is as follows: firstly, the spatial domain enhancement method of medical image is studied. The spatial domain method is used to deal with the gray level of the original image directly, including gray level transformation, histogram equalization, spatial filtering and other enhancement methods. Then, wavelet transform theory and ridgelet transform theory are studied, and a medical image enhancement method based on ridgelet transform is proposed. The main process of the algorithm is as follows: firstly, the image is transformed by finite Radon, then by one dimensional multiwavelet transform, then the coefficients of multiwavelet transform are enhanced, and then the medical image is reconstructed. Finally, the reconstructed image is transformed into gray level to get a better enhancement effect. The experimental results show that the algorithm can achieve good enhancement effect.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41
，
本文编号：2381922
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2381922.html
