A Proposal of Real-Time foreign object detection for Food Through Conveyor Belt Using hyperspectral Camera

Abstract

Every year, foreign substances such as insects and hair are detected in the various foods we eat. Foreign substances found in these foods may be offensive to the consumer or adversely affect humans when consumed with the food. Considering the process of food passing through to consumers in terms of manufacturing and transportation, we propose a solution to the original problem in the manufacturing process. In manufacturing, hyperspectral cameras are used to inspect the raw materials and finished products of food passing through the conveyor belt. From the human eye, it is difficult to distinguish due to the small amount of foreign matter or the color similar to the food. Also, the speed of the conveyor belt is high and the ability to detect at high speed is also required. Therefore, by comparing the spectroscopy for each pixel in the VNIR (Visible and Near Infrared) region, the possibility of foreign material detection using a hyperspectral camera is suggested.