Application of pixel intensity, fractal dimension and skeleton parameters for detection of adulteration of cow ghee with vegetable fat derived from image analysis.

Abstract

Abstract

In Indian subcontinent clarified butter fat (ghee) is most widely used milk product due to its enjoyable taste, delicious aroma, high nutritional value and unique flavour. Because of its premium nature and high demand ghee adulteration has always been a serious problem. In this paper pixel intensity, fractal dimension, and some skeleton parameters derived from image analysis after applying the developed protocol to the images acquired with flat bed scanner was used for detection of adulteration of cow ghee with vegetable fat. ImageJ open license free software was used to process the images.

Pixel intensity, fractal dimension and skeleton parameters namely junctions, branches and end point voxels were measured for pure cow ghee and compared with adulterated ghee of 5%, 10%, 15% and 20% vegetable fat. Pixel intensity, fractal dimension and skeleton parameters  of images acquired byflat bed scanner showed an increasing trend with the level of adulteration for all seven different trials; it differed significantly over different adulteration levels (P<0.05). The fitted regression equation showed a linear trend with coefficient of determination of >0.9 indicated the potential of these parameters for detecting the adulteration.

These parameters can be employed for routine laboratory gheeadulteration detection with advantages of convenience, rapidity, accuracy, large sample size handling with repeatability by critically following the developed protocol.