Dept. of Electr. Eng. Nat. Tsing Hua Univ., Hsinchu.
IEEE Trans Med Imaging. 1992;11(2):141-52. doi: 10.1109/42.141636.
The classification of ultrasonic liver images is studied, making use of the spatial gray-level dependence matrices, the Fourier power spectrum, the gray-level difference statistics, and the Laws texture energy measures. Features of these types are used to classify three sets of ultrasonic liver images-normal liver, hepatoma, and cirrhosis (30 samples each). The Bayes classifier and the Hotelling trace criterion are employed to evaluate the performance of these features. From the viewpoint of speed and accuracy of classification, it is found that these features do not perform well enough. Hence, a new texture feature set (multiresolution fractal features) based on multiple resolution imagery and the fractional Brownian motion model is proposed to detect diffuse liver diseases quickly and accurately. Fractal dimensions estimated at various resolutions of the image are gathered to form the feature vector. Texture information contained in the proposed feature vector is discussed. A real-time implementation of the algorithm produces about 90% correct classification for the three sets of ultrasonic liver images.
研究了超声肝脏图像的分类，利用空间灰度依赖矩阵、傅里叶功率谱、灰度差分统计和 Laws 纹理能量测度。使用这些类型的特征来对三组超声肝脏图像（正常肝脏、肝癌和肝硬化（每组 30 个样本）进行分类。采用贝叶斯分类器和 Hotelling 迹准则来评估这些特征的性能。从分类的速度和准确性的角度来看，发现这些特征的性能不够好。因此，基于多分辨率图像和分数布朗运动模型，提出了一种新的纹理特征集（多分辨率分形特征），以便快速准确地检测弥漫性肝脏疾病。收集图像的各个分辨率下的分形维数来形成特征向量。讨论了所提出的特征向量中包含的纹理信息。算法的实时实现对三组超声肝脏图像的分类正确率约为 90%。
