Use of Electronic Nose to Identify Levels of Cooking Cookies

Currently, the baking of cakes using an electric oven is based on cooking duration. Usually, colors can be used to determine the levels of cooking food. However, many cakes have similar colors at each stage, which cannot be used as indicators of doneness. Through today’s technology, the sense of smell can be imitated using a gas sensor combined with artificial intelligence for food quality control. In this study, an electronic nose system was developed to distinguish levels of baking cookies. This process involved 20 gas sensors and 10 classification algorithms based on aroma. The optimization technique based on correlation analysis and distinguishing rate methods was carried out to obtain a small number of sensors that still maintained high accuracy values. Several sensors were eliminated, while the remaining 13 sensors were retained. The selected electronic nose system consisted of 6 gas sensors and convolutional neural networks. It succeeded in distinguishing cooking levels, including undercooked, cooked, and overcooked food, with an accuracy of 90.0%, a precision of 89.7%, a recall of 92.6%, and an F1-measure of 90.2%. This system has the potential to produce a consistent quality of cookies.