A video image analysis system was designed to quantitatively measure the amount of specks and the color of peanut butter samples for grade purposes. The video image of a peanut butter surface 5.8 cm by 5.8 cm was captured and converted into 384 by 384 picture elements (pixels). The intensity of each of the 147,456 pixels was classified into one of 256 shades of gray from zero for black to 255 for white. The percent of total pixels that represented specks was defined as the speck index. The aveage shade of gray of all 147,456 pixels was defined as the color index. The speck and color indices were computed for 52 peanut butter samples that had been graded by experienced Agricultural Marketing Service (AMS) inspectors. The speck and color indices were both in good agreement with the AMS speck and color classifications assigned to the samples by the AMS inspectors.
1Paper No. 11148 of the Journal Series of the North Carolina Agricultural Research Service, Raleigh, NC 27695–7601. The use of trade names in this publication does not imply endorsement by the North Carolina Agricultural Research Service nor the United States Department of Agriculture of the products named, nor criticism of similar ones not mentioned.