QUANTITATIVE ANALYSIS OF CUT SURFACES OF BRASSICAS BY MEANS OF FRACTAL TEXTURE FOURIER IMAGE ANALYSIS
Appearance and texture changes related to tissue deterioration are commonly considered as measures of freshness and quality decline in fresh-cut research and industry. The cutting operation is one of the most limiting steps in the production of fresh-cut vegetable products. A smooth cut surface is desirable to avoid major damage to vegetable cells and reduce enzymatic browning, dehydration, discolorations or microbiological spoilage. A new approach to evaluate the cuts quality is presented in this work. It is based on the quantification of irregular colour patterns that emerge from the surfaces after slicing. Stems of broccoli (Brassica oleracea L. var. italic) and cauliflower (Brassica oleracea L. var. botrytis) were cut using different methods (mechanical and manual) and different sharpness of the blades cutting edge. Cut surfaces were captured by a computer vision system using a calibrated flatbed scanner and transformed into the CIELab colour space using a quadratic transformation function. To classify the different cuts two methods are proposed: the use of mean L values (traditional method) and as a newer approach, the utilisation of fractal dimension value (FD). Latter uses Fourier fractal texture image analysis to describe the complexity of lightness intensity distribution over the surface. Results show that fractal dimension values and their evolution in time provide the right index to differentiate between cutting techniques.
Mir-Bel, J., Barriobero , J. and López, R. (2012). QUANTITATIVE ANALYSIS OF CUT SURFACES OF BRASSICAS BY MEANS OF FRACTAL TEXTURE FOURIER IMAGE ANALYSIS. Acta Hortic. 934, 167-172
Brassica oleracea, fractal surface, fresh cut, cut quality