Spatial frequency domain imaging for detecting bruises of pears

A spatial frequency domain imaging system (SFDI) was developed to estimate the optical properties of biological samples. The system was calibrated by using a series of self-made solid phantoms which covering a wide range of absorption ( µ a ) and reduced scattering coefficients ( \({\mu ^{\prime}_s}\) ). The relative errors between the reference and calibrated values were regarded as the evaluation parameters for validation effectiveness of the system, the results showed that the maximum relative errors of µ a and \({\mu ^{\prime}_s}\) are 3.92 and 2.39% respectively. Quantitative absorption and scattering maps at the wavelength of 527 nm were obtained for four kinds of pear samples (normal and bruised with three levels of bruised severity with impact energy of 0.025, 0.075 and 0.125 J). The normal pears could be distinguished from the bruised pears by comparing values of coefficient of variation ( CV ) of \({\mu ^{\prime}_s}\) maps, and the correct rates of the discriminant analysis were 100% for non-bruised pears,and 98.33% for bruised pears. Bruises induced using 0.025 J (class 1) were distinguishable from those induced using 0.075 and 0.125 J (class 2) by comparing ratio between mean values of bruised region and normal region of \({\mu ^{\prime}_s}\) maps, the results turned out that the class 1 had the correct rate of 90%, and class 2 had the correct rate of 87.5%. The research showed that SFDI has the potential to be used in the detection of bruises on ‘Crown’ pears and the proposed calibration method lay the foundation for future studies. Future work will be conducted with an emphasis on the acquisition of more spectral information, surface profile correction and the acceleration of detection speed.