Textural Features for Image Classification
- R. Haralick, K. Shanmugam, I. Dinstein
- MathematicsIEEE Transactions on Systems, Man and Cybernetics
- 1 November 1973
These results indicate that the easily computable textural features based on gray-tone spatial dependancies probably have a general applicability for a wide variety of image-classification applications.
An adaptive filter for smoothing noisy radar images
- V. Frost, J. Stiles, K. Shanmugam, J. Holtzman, S.A. Smith
- Computer Science, MathematicsProceedings of the IEEE
An algorithm for smoothing noisy radar images is presented, which is easily implemented in the spatial domain and is computationally very efficient and shown that the filter preserves edges.
Computer Classification of Reservoir Sandstones
A procedure is developed to extract numerical features which characterize the pore structure of reservoir rocks. The procedure is based on a set of descriptors which give a statistical description of…
Combined spectral and spatial processing of ERTS imagery data
Land use classification using texture information in ERTS-A MSS imagery
The author has identified the following significant results. Preliminary digital analysis of ERTS-1 MSS imagery reveals that the textural features of the imagery are very useful for land use…
Comparative Study of a Discrete Linear Basis for Image Data Compression
A new discrete linear transform for image compression which is used in conjunction with differential pulse-code modulation on spatially adjacent transformed subimage samples and finds that for low compression rates, the Karhunen-Loeve outperforms both the Hadamard and the discrete linear basis method.
Condensed anisotropic diffusion for speckle reducton and enhancement in ultrasonography
Experiments indicate better speckle reduction and effective preservation of edges and local details in second-order diffusion-based methods.
A Computationally Simple Procedure for Imagery Data Compression by the Karhunen-Loeve Method
It is shown that the eigenvalues and eigenvectors of the N × N bisymmetric covariance matrix can be obtained from the eigenevalues and Eigenvector of two N/2 × N/ 2 submatrices.
Crop classification using multidate/multifrequency radar data
Both C- and L-band radar data acquired over a test site near Colby, Kansas during the summer of 1978 were used to identify three types of vegetation cover and bare soil. The effects of frequency,…
The Recognition of Extended Targets: SAR Images for Level And Hilly Terrain
- J. Stiles, V. Frost, J. Holtzman, K. Shanmugam
- Environmental ScienceIEEE Transactions on Geoscience and Remote…
- 1 April 1982