Abstract:
Abstract: Quality of fresh meat affects the health of consumers and also receives much attention from government and industry. Recent years' study shows that visible and near-infrared reflectance spectroscopy (VIS-NIRS) was applied to fast and non-invasively detect meat's quality factors which include water content, tenderness, pH value, TVB-N and color. Inspection departments have created the need for a cost-effective and nondestructive quality-control analysis system. But the traditional detection system consist of spectrographs, illuminant, optical fiber, computer and dark box which cannot be moved and operated easily, so it has not been widely used.. Portable detection devices are powerful instruments offering several advantages for nondestructive, online analysis: small size, low cost, robustness, sample user interface, portability, and ergonomic design. Domestic and abroad scholars have developed portable spectrometers to detect internal and external parameters of fruits and vegetables, but characterization of meat parameters is still needed. In order to meet the demand of inspection departments for portable devices, a rapid and nondestructive detection device for fresh meat quality based on ARM (advanced RISC machines) processor was designed. Working principal, hardware composition, software system and functional test were introduced. The hardware system consisted of ARM processing unit, light source and detection unit, spectral data acquisition unit, LCD (liquid crystal display) touch screen display unit and the cooling unit. Linux operating system and quality parameters acquisition processing application were designed. In this paper, ARM processor was applied to collect spectrum data, to control the serial port, to process data, and to export results on LCD. ARM processor received spectral data via the serial port from Micro spectrometer. A halogen lamp was used as a light source because of its wide spectrum range. USB4000 spectrograph was selected as the collecting spectrum unit due to its small size and high-performance. In order to transmit optical signals, there was a coupling lens fixed in front of the USB4000's slit. As the object of detection is meat, it is easy to contaminate the probe by blood, so an optical window was set between the meat and the probe. As to software, the C++ language was used to compile to achieve offering a friendly user interface. Users can get values of the meat's parameter from the LCD. This system has realized collecting spectral signal, storing, displaying and processing as integration with the weight of 3.5 kg and volume of 184 mm×127 mm×114 mm. 76 pieces of meat were used in the experiment to validate the stability and reliability. The experimental results show that this device can be used for portable detection of meat quality with small volume and the advantage of being portable, with no damage and being fast.