Performance of hyperspectral imaging system for poultry surface fecal contaminant detection

Abstract

A hyperspectral imaging system demonstrated potential to detect surface fecal and ingesta contaminants on poultry carcasses. Hyperspectral imagery were processed and analyzed with four pre-processing methods considering calibration and spectral smoothing approaches. A band-ratio image processing algorithm, using 2-wavelengths ( I 565/ I 517) and 3-wavelengths ( I 576 − I 616)/( I 529 − I 616) equations, was then applied for each pre-processed method that included background masking for ratioed images, and fecal thresholding. Based on a high accuracy of 96.2% for predicting surface contaminants and significantly less false positives on the 64 birds measured, the method with calibrated and smoothed images was the best pre-processing method for fecal and ingesta contaminant detection. In conjunction with appropriate image processing algorithms, the hyperspectral imaging system is an effective technique for identifying fecal and ingesta contaminants on broiler carcasses. Specifically, band-ratio with 2-wavelength equation ( I 565/ I 517) performed very well with 96.4% accuracy and 147 false positives for detecting both feces (duodenum, ceca, colon) and ingesta contaminants.