Building a portable device based on DSP for meat discrimination

More than ever, people today demand quality in the food that they consume. In this work, we will present an instrument and a method to discriminate horse and beef meat. Meat color is a long established property that is used to classify meat. The proposed device employs Charge-coupled Device (CCD) imaging techniques, digital image processing, Digital Signal Processor (DSP) control, processing techniques and Liquid Crystal Display (LCD) screen. Image analysis methods were developed for discriminating beef and horse meat. Samples were placed under cold storage at 4°C for two weeks. Two color models are used to define beef and horse meat color: the RGB (Red, Green and Blue) and HSI (Hue, Saturation and Intensity) model. Principal Component Analysis (PCA) was employed to optimize the data matrix. Results show that the portable device was able to distinguish between beef and horse meat using HSI model.