A novel approach for metal detection in food using curve fitting technique

Food product contamination is a hazard to human health. Therefore, we can see increasing demand for Metal Detectors. To guarantee the safety and robustness of food product metal detectors are used in the food industry to detect any trace element of metal that may have got into the food during the production process. This paper presents a novel method of detecting contamination in food products by using curve fitting on the data set collected from the metal detector provided by the EME Automation, Goa, India. Features such as amplitude and phase were extracted. The curve fitting is done on the data sets through the use of various algorithms/functions. Implementation of the product effect, zero-band tolerance on the data is taken into consideration. Polar graphs approach have been used for the discrimination between the different products. On the basis of these polar graphs, better identification should be undertaken. This is implemented using MATLAB.