【作者】 刘辉；

【导师】 李国丽；

【作者基本信息】 合肥工业大学 ， 电工理论与新技术， 2006， 硕士

【摘要】 随着数字化医学图像的广泛应用，X线摄影、CT、MRI等医学图像的数量日益增加，医学图像压缩技术的重要性和必要性越来越突出。医学图像压缩技术在图像存档与传输系统中非常重要，图像数据压缩既可以提高图像传输速率，缩短传输时间，还可以减少图像存档需要的存储空间。 然而医学图像不同于一般图像，它们有其自身的特点，对医学图像进行数据压缩在保证图像使用质量的前提下，将医学图像的位图信息转变成另外一种能使数据量缩减的表达形式，其思路在于减少图像数据中的冗余信息，避免表达同一信息的相关数据重复存储。 小波神经网络是将小波变换和神经网络相结合的一种新方法，虽然它的理论还不完善，但已在许多领域被广泛应用，显示了良好的应用性能。本文提出一种用于图像压缩的小波神经网络算法，在医学图像压缩中的应用进行了研究，实验结果表明该算法是具有较好的图像压缩效果。更多还原

【Abstract】 With the extensive application in clinical, the amount of digital medical image, such as X-ray, CT, MRI, etc. increases rapidly day after day, the importance and necessity of medical image compression technical becomes more and more significant. Medical image compression is very important in image saving and transmission system, it can not only improve the image transmission rate ,shorten the transmission time, but also reduce the storage in the image saving.However, medical images are different from common images, they have their own characteristics. Medical image compression technical changes medical image bitmap information into another expression form which can reduce data amount. The purpose of image compression is to reduce the redundant information and avoid saving the same information repeatedly.Wavelet network is a kind of new artificial algorithm which combine the wavelet transformation with the neural network. This algorithm has been applied extensively to many fields although its theory is not perfect yet. This thesis proposes a kind of Wavelet neural network algorithm which has been used in the medical image compression, the experiment result shows that the method has a good compression effect.更多还原