APPLICATIONS OF DIGITAL IMAGE ANALYSIS (DIA) TO FOOD-QUALITY ASSESSMENT: AN OVERVIEW

Recently, the role of digital image analysis has been grown widely in different technological fields such as, space research, communications, remote sensation, medicine and in analysis, processing and quality assessment of foods.
The term image refers to a two-dimensional light-intensity function, denoted by f (x, y), when the value or amplitude of f at spatial coordinates (x, y) gives the intensity (brightness) of the image at that point. We may consider a digital image as a matrix whose row and column indices indentify a point in the image and the corresponding matrix element value identifies the gray level at the point. The elements of such a digital image array are called image elements, picture elements, pixels, or pels with the last two names being commonly used as abbreviations of “pictures elements”. An expansion in image analysis applications is occurring within the agriculture and food industries with the result that image analysis can be used for the characterization of food products. It is noteworthy that images are often studied for detect-ing or enhancing geometrical structures.
Image analysis can be used in many aspects of food industry, analysis and quality assurance. For in-stance, image analysis can be used to discriminate cereal grains and classify cereal kernels according to their physical dimensions. Meanwhile, colour analysis of individual wheat grains might facilitate the identification of grains in the wheat-grading context. Moreover, by selecting the near IR wavelengths of excitation and emission, images obtained can be applied to discriminate starch, gluten and bran which present the three major components of wheat grain. The study of colour or intensity of the points (pixels) in an image can be a way to obtain chemical information, such as fat and lean contents in meat and meat products. In case of minced meat, the fat can be differentiated from lean using UV light. Furthermore, digital image analysis was developed to measure the size and spatial distribution of the satellite microbial colonies as a function of distance from the primary colony.
Bar coding represents an important application of image analysis. Bar coding is a form of artificial identifier. It is a machine readable code consisting of a pattern of black and white bars and space defined ratios which represent alphanumeric character. A sensor scans the bar code symbol and converts the vis-ual image into an electrical signal.