Development of Electronic Sensing Devices of Artificial Neural Network Programming Application for Borax in Food Detection

Abstract

The use of Food Additives is regulated in BPOM regulation number 11 of 2019. In practice, there are food entrepreneurs who use prohibited BTP, one of which is borax to thicken meatballs, noodles, and rice cakes. The prohibition on the use of borax in food is stated in the Regulation of
the Minister of Health No. 722/MenKes/Per/IX/88 where borax is a hazardous material. The development of electronic sensing equipment to detect borax can be used for rapid identification in the field as early detection for food safety monitoring. The research phase consists of the design of the E-sensing device followed by the programming of the Artificial Terms Network method using Matlab software. Meatball samples with the addition of 0%, 1%, 2%, and 3% borax were measured for RGB color image values using E sensing as input for Matlab programming. The design of E Sensing uses an NCU ESP8266 node microcontroller, TCS 3200 color sensor, and LCD can be used to measure the color image of meatballs with RGB value output. Matlab ANN programming results are less sensitive for borax detection at 1% and 2% borax content, as seen from the low test accuracy value (<75%), while the 3% borax content is quite sensitive with a test accuracy value of 90.9091%. Display graphical user interface (GUI) on a computer screen can be used easily by users.

Keywords: Borax, NCU ESP 8266 microcontroller, Artificial neural network, TCS 3200 sensor, Electronic sensing