Textural descriptors

The Shape Analyzer is a scientific instrument which is used to analyze the shape, size, and texture of objects. The item for analysis may be the object itself, its photograph, optical or electron micrograph, etc. The shape-size analysis is achieved by converting the profile to a set of shape and size descriptors which are complete, unequivocal and invariant. The texture analysis converts the full image of the object into a set of textural descriptors which are unequivocal, complete and invariant. These shape, size and texture descriptors can be used in research, in quality control, in process control and specifications. Additional details are given in Table 1 and the Appendix. 

In order to do this the points of the image have to be incorporated into the analysis in the form of invariant descriptors. The interpretation of the textural morphology features depends upon the physics of the situation (for example, the gray level may be related to altitude or in another case to the chemical analysis from point to point. The interpretation also depends upon the mathematical relationships between the various morphological descriptors used. There are three major types of morphological texture descriptors. These are statistical terms, symmetry operations, and Invariant texture morphology descriptors (ITMD s). 

Indications of Physical Meanings of Shape and Textural Descriptors... 

It has to be noted that the direct analysis of the texture descriptors is generally difficult the descriptors help to compare one structure to another, in... 

Keywords— Multi-modality medical images, feature analysis, shape, color and texture descriptor. 

Figure 1,2 and 3 shows the accuracy value of precision and recal for texture, shape and color descriptor. It shows that CT, XR, US, PET and PX are suitable to be classified using texture descriptor as depicted in Fig. 1. As for moment shape descriptor, the result almost similar with texture but the value of accuracy is higher as shown in Fig. 2. It explains that shape descriptor using local level has better presentation compare to texture descriptor in analysing multi-modality medical images. Finally the color descriptor in Fig.3 shows that GX, PX, NM and PET have higher value. This is due to both modalities have used more colors compare to other modalities which concentrate only on grey-scale image.

Compared to most white and red wines, Vin Santo wines, and especially the slightly sweet and sweet styles, are characterized primarily by their flavor and taste, rather than their aroma. Accordingly, in addition to sweetness and acidity, the most used descriptors to evaluate Vin Santo in relation to its perception in the mouth are alcoholicity (warm sensation), texture, viscosity, and overall taste persistence. Among the flavor descriptors, those relating to caramelization (like flavors of honey, milk-honey candy, molasses, caramel) are the most used, as these are more suitable to describe the different Vin Santo. It is estimated that these descriptors... 

Methods of objective measurement of cereal foam structures are reviewed, including image analysis, confocal microscopy and x-ray tomography. The analysis of foam structures and their relationship with mechanical and rheological properties is described, and also the relationships between these structures and sensory descriptors such as crispness, crunchiness and texture. The size, shape and anisotropy of bubbles and their cell walls in foams are seen as critical in determining their fracture properties and sensory perception of crispness. Techniques for measuring crispness using acoustic emission and force-deformation profiles are discussed. 

Swyngedau s equation has proven to be a good model descriptor of food powder agglomerate compression (Yan and Barbosa-Canovas, 1997, 2000). Yan and Barbosa-Canovas (1997) converted the Swyngedau model into Equation 17, which allows for the direct use of force-deformation readings recorded from a TA-TX2 texture analyzer. 

Here, we will outline the scheme they have proposed. In [52] the wavelet transform was used both to analyse the image prior to segmentation, enabling feature selection, as well as to provide spatial frequency-based descriptors as features for segmenting textures (see Fig. 27). Smooth and textured images can easily be distinguished from each other by examining their wavelet transforms. 

The subsequent step consists in training assessors to rate the perceived intensity of some descriptors of foods on an evaluation scale. Panelists learn to quantify their perception, initially by ranking series of single odor or taste or texture stimuli with respect to the intensity of a particular characteristic, then quantifying the perception on the evaluation scale. The test samples used are model systems for instance, a single taste or flavor compound in water or other neutral media or solid or semisolid materials differentiated in their texture properties or samples obtained by spiking product samples with a flavor... 

---