Abstract: Among other things, a texture of an image is characterized by deriving lacunarity measures from the image based on a wavelet analysis.

Abstract: A light image is conveyed from a biological tissue through a flexible optical system to an image receiver, where it is converted to a form which may be entered into a computer. The computer segments the image by generating a segmentation mask defining the boundary of a region of interest in at least one spectral band, estimates at least one rotationally and translationally invariant statistical measure of coefficient distributions of the multiscale wavelet maxima representations of the digital images in at least one spectral band, characterizes the condition of the tissue based on the estimated values, and outputs the characterization of the condition of the tissue.

Abstract: Systems and methods for the multispectral imaging of skin tissue enables automatic characterization of the condition of a region of interest of the skin, based on direct digital imaging of the region of interest or the digitization of color photographic slides of the region of interest, illuminated by appropriately filtered light. Preferably, a digital image at a low spectral band is automatically segmented and that segmented mask is used to segment the other images by a digital processor. Parameters are estimated through wavelet maxima representations. Parameters related to the texture, asymmetry, blotchiness and border irregularities are also automatically estimated. The region of interest is automatically characterized by the digital processor, based on those parameters. The region of interest may include a skin lesion, in which case the present invention enables the characterization of the lesion as malignant or benign. The region of interest may also include wounds or burns.