Search

The aim of this study was to investigate the possibility of computer vision system (CVS) application in fat content estimation for different types of fermented sausages. Four different types of local fermented sausages with different fat contents were studied: Njeguška, Kulen, Pirotska and tea sausage. Results obtained for CVS-estimated fat content were compared to the results of traditional chemical analysis. Relative errors of fat content estimation in Njeguška, Kulen, Pirotska and tea sausage were 1.47%, 0.46%, 20.84% and 11.19%, respectively. Results of t-test showed a significant (p<0.01) difference between mean fat contents obtained by CVS and chemical analysis in the case of Pirotska sausage. On the other hand, there was no significant (p<0.01) difference between mean fat contents obtained by the two methods for the rest of the analysed sausages. The results indicate CVS has potential for application in the analysis of fat content of fermented sausages. [ABSTRACT FROM AUTHOR]

Purpose: To develop a method to estimate the mean fractional volume of fat (vfat) within a region of interest (ROI) of a tissue sample for wide-angle x-ray scatter (WAXS) applications. A scatter signal from the ROI was obtained and use of f in a WAXS fat subtraction model provided a way to estimate the differential linear scattering coefficient μs of the remaining fatless tissue. Methods: The efficacy of the method was tested using animal tissue from a local butcher shop. Formalin fixed samples, 5 mm in diameter 4 mm thick, were prepared. The two main tissue types were fat and meat (fibrous). Pure as well as composite samples consisting of a mixture of the two tissue types were analyzed. For the latter samples, Vfat for the tissue columns of interest were extracted from corresponding pixels in CCD digital x-ray images using a calibration curve. The means f were then calculated for use in a WAXS fat subtraction model. For the WAXS measurements, the samples were interrogated with a 2.7 mm diameter 50 kV beam and the 6° scattered photons were detected with a CdTe detector subtending a solid angle of 7.75 × 10-5 sr. Using the scatter spectrum, an estimate of the incident spectrum, and a scatter model, μs was determined for the tissue in the ROI. For the composite samples, a WAXS fat subtraction model was used to estimate the ßs of the fibrous tissue in the ROI. This signal was compared to ß s of fibrous tissue obtained using a pure fibrous sample. Results: For chicken and beef composites, f = 0.33 ± 0.05 and 0.32 ± 0.05, respectively. The subtractions of these fat components from the WAXS composite signals provided estimates of μs for chicken and beef fibrous tissue. The differences between the estimates and μs of fibrous obtained with a pure sample were calculated as a function of the momentum transfer x. A t-test showed that the mean of the differences did not vary from zero in a statistically significant way thereby validating the methods. Conclusions: The methodology to estimate vfat in a ROI of a tissue sample via CCD x-ray imaging was quantitatively accurate. The WAXS fat subtraction model allowed ß s of fibrous tissue to be obtained from a ROI which had some fat. The fat estimation method coupled with the WAXS models can be used to compare ßs coefficients of fibroglandular and cancerous breast tissue. [ABSTRACT FROM AUTHOR]