【摘要】 该文提出了一种在基于离散余弦变换(DCT,Discrete Cosine Transform)压缩域进行纹理图像分类的方法。此方法主要利用了DCT域能量具有方向性的特点,直接在游程长编码(RLE,Run LengthEncoding)形成的码流中根据这种方向性来提取、组织特征矢量,对纹理图像进行分类。此方法只需对压缩图像进行简单的Huffman解码,要处理的数据量较少,处理速度快。实验结果表明,这种分类方法具有较高的准确性。更多还原

【Abstract】 A novel method of texture-image classification in DCT compressed-domain is proposed. The textural feature; of energy-orientation is extracted in DCT transfonned-domain and this textural feature can be used to classify texture-images in RLE bit-stream directly. The experimental results demonstrate that texture-images can be classified at higher correct, classification rate under random noise.更多还原