Machine vision system for online inspection of freshly slaughtered chickens

Abstract

A machine vision system was developed and evaluated for the automation of online inspection to differentiate freshly slaughtered wholesome chickens from systemically diseased chickens. The system consisted of an electron-multiplying charge-coupled-device (EMCCD) camera used with an imaging spectrograph and controlled by a computer to obtain line-scan images quickly on a chicken processing line of a commercial poultry plant. The system scanned chicken carcasses on an eviscerating line operating at a speed of 140 chickens per minute. An algorithm was implemented in the system to automatically recognize individual carcasses entering and exiting the field of view, to locate the region of interest (ROI) of each chicken, to extract useful spectra from the ROI as inputs to the differentiation method, and to determine the condition for each carcass as being wholesome or systemically diseased. The system can acquire either hyperspectral or multispectral images without any cross-system calibration. The essential spectral features were selected from hyperspectral images of chicken samples. The differentiation of chickens on the processing line was then carried out using multispectral imaging. The high accuracy obtained from the evaluation results showed that the machine vision system can be applied successfully to automatic online inspection for chicken processing.

Quality classification of cooked, sliced turkey hams using NIR hyperspectral imaging system

“…On the other hand, imaging techniques in the form of computer or machine vision systems has the capability to provide superior spatial information. Therefore, imaging techniques have been applied for visual evaluation of meat and meat products and for rapidly identifying quality parameters on the processing line with the minimum of human intervention (Brosnan and Sun, 2004;Du and Sun, 2004;Yang et al, 2009;Fathi et al, 2009;Kaya et al, 2008). Although external attributes such as size, shape, colour, surface texture and external defects can easily be evaluated by imaging techniques, the compositional attributes such as moisture, fat and protein contents are very difficult to be determined with this relatively simple imaging approach due to very limited spectral information (Du and Sun, 2004).…”

Literature review: spectral imaging applied to poultry products

“…Traditional detection methods rely on inspectors to visually check for infection symptoms on poultry carcasses (e.g., lesions, degeneration of skeletal muscles), as indicators of septicaemia or toxaemia ( Fisher et al., 1998 ). Inspectors typically work at a rate of 35 birds per minute on processing lines ( Yang et al., 2009 ). Because traditional methods rely on inspectors, the process is slow, tedious, and prone to human error.…”

The development of a simple multispectral algorithm for detection of fecal contamination on apples using a hyperspectral line-scan imaging system

“…The EMCCD camera was capable of acquiring highresolution images at high speed under low illumination environments [11]. The imaging spectrograph dispensed light to form a spectrum for each pixel along the linear field of view during line-scanning.…”

“…Using the former could miss essential spectral data or mix such data with other useless data, while using the latter can unnecessarily slow the inspection system as a result of the huge amounts of data being acquired and processed. A multispectral algorithm based on selected essential wavebands can be implemented in the same imaging system as that used to acquire the hyperspectral data, without the need for hyperspectral/multispectral crosssystem calibration [8,11].…”