2.2. Image preprocessing
The purpose of image preprocessing was to segment the region
of interest (ROI) from the peripheral and intermuscular fat as well
as the connective tissue and surrounding muscles. Image preprocessing
was conducted on marbling standards and hyperspectral
images of pork samples to obtain the ROI for marbling detection.
All operations in this section were performed using MATLAB
7.3.0 (The MathWorks, Inc., MA, USA).
The contour of marbling standards, referring to the outer
boundary of meat, was obtained by using a thresholding technique
and an edge detection algorithm. The thresholding technique (Jain,
1989) transforms a gray-level image (the green channel of marbling
standards) to a binary image (i.e. black and white image).
The grey-level of 200 was used as threshold (the range of gray-levels
varies from 0 to 255). Pixels whose values are higher than the
threshold are closed (set to 0), otherwise are open (set to 1). The
obtained binary images of the marbling standards were used to extract
the contour of marbling samples on these standards by
employing the Sobel edge detector (Jain, 1989).