【摘要】 离散小波变换(DWT)虽然广泛用于图像处理,但DWT存在两个缺点:其一,缺乏平移不变性,这意味着信号的微小平移将导致各尺度上的小波系数的能量分布有较大变化;其二,缺乏方向敏感性,可分离的二维小波变换只有三个方向的高频信息即水平、垂直和对角。利用对偶树复小波变换(DT-CWT)进行图像纹理分类,可以克服上述离散小波变换的不足,该方法具有逼近的平移不变性和更多的方向选择性,有利于特征的跟踪、定位和保留。本文采用对偶树复小波变换和BP神经网络相结合的方法对织物纹理进行分类,实验表明,分类率可达98%。更多还原

【Abstract】 Although the discrete wavelet transform (DWT) is a powerful image processing tool,and it has two disadvantages that undermine its usage in many applications. First,DWT is shift sensitive because input-signal shifts generate unpredictable changes in DWT coefficients. Second,DWT suffers from poor directionality because the discrete wavelet transform coefficients reveal only three spatial orientations. In order to overcome the shortcoming of the commonly-used image processing methods,the fabric texture classification method based on dual tree complex wavelet transform (DT-CWT) was proposed. Compared with the traditional discrete wavelet transform,the dual tree complex wavelet transform has the properties of approximate shift invariance and more directionality. These properties are good for tracing,locating and preserving image features. In this study,DT-CWT and BP neural network together were used to classify the fabric texture. Experimental result shows that the classification rate can attain 98%.更多还原