Textural Features for Image Database Retrieval（纹理特征的图像数据库检索）.pdf

Textural Features for Image Database Retrieval Selim Aksoy and Robert M. Haralick Intelligent Systems Laboratory Department of Electrical Engineering University of Washington Seattle, WA 98195-2500 faksoy,haralickg@isl.ee.washington.edu Abstract in finding similarities between images in a database. In this paper, we discuss two textural feature extraction methods to This paper presents two feature extraction methods and represent images for content-based retrieval. In the first one, two decision methods to retrieve images having some sec- texture is defined as being specified by the statistical distri- tion in them that is like the user input image. The features bution of the spatial dependencies of gray level properties. used are variances of gray level co-occurrences and line- Variances of gray level co-occurrence matrices are used to angle-ratio statistics constituted by a 2-D histogram of an- extract this information. This is consistent with [6], where gles between two intersecting lines and ratio of mean gray Haralick et al. used co-occurrence matrices to classify sand- levels inside and outside the regions spanned by those an- stone photomicrographs, panchromatic aerial photographs, gles. and ERTS multispectral satellite images. Comparative stud- The decision method involves associating with any pair ies [10, 4] showed that gray level spatial dependencies are of images either the class “relevant” or “i