Submitted to: Applied Optics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: May 5, 2005
Publication Date: February 1, 2006
Citation: Ding, F., Chen, Y.R., Chao, K., Chan, D.E. 2006. Two-color mixing for classifying agricultural products for safety and quality. Applied Optics. 45(4):668-677. Interpretive Summary: Presently, the Instrumentation and Sensing Laboratory (ISL) is developing a low cost optically enhanced device that can assist inspectors or plant processors in small meat and poultry plants to conduct inspection in situ. In this paper, two-band color mixing is shown to be directly related to the two-color band ratio criterion, which has been used for successful separation of wholesome and unwholesome agricultural products. This paper also reports the results of selecting optimal waveband pairs from color difference and chromaticness difference indices for chicken carcass inspection and for cucumber chilling injuries detection. It is shown that two color-mixing binoculars have potential for practical use in a processing plant environment. Based on the results of this study, binoculars equipped with special two-narrow-band interference optical filters are being tested for safety inspection of poultry carcasses and plant produce. This information is useful to FSIS and researchers who are interested in developing low cost vision-based systems for in situ inspection of agricultural products.
Technical Abstract: In this paper, it was shown that the chromaticness of the visual signal resulting from the two-color mixing achieved through our device is directly related to the band ratio of light intensity at the two selected wavebands. A technique that implements the band-ratio criterion in a visual device using two-color mixing is presented here. The device will allow the inspectors visually identify the target in accordance with two wavelength band ratio. It is a method of inspection with human vision assisted by an optical device, offering flexibility and significant cost savings compared to a multi-spectral machine vision system that implements the band ratio criterion. With proper selection of the two narrow wavebands, discrimination by chromaticness that has a direct relation with band ratio can work very well. An example application of this technique is given for the inspection of chicken carcasses with various diseases. An optimal wavelength pair of 454 nm and 578 nm is selected which optimizes differences in the saturation and hue in the CIELUV color space among different types of targets. Another example application for on the detection of chilling injury in cucumbers is given with the wavelength pair of 504 nm and 652 nm. The novel two-color mixing technique for visual inspection can be implemented on visual devices for various applications, ranging from target detection to food safety inspection.