<![CDATA[This study aims to develop a tempe quality detection model using the MobileNetV2 architecture, trained with a tempe image dataset from the Sanan Tempe Industry Center. Ensuring the quality of tempe, distinguishing between fit-for-consumption and spoiled tempe, is crucial for consumer safety. MobileNetV2 was selected due to its efficient image classification capabilities with a lightweight model, making it suitable for practical applications. The dataset consists of images of both fit and spoiled tempe, which are used to train the model. Evaluation results demonstrate that the MobileNetV2 model can classify tempe quality with high accuracy. This research aims to provide a solution for accelerating and simplifying automatic tempe quality detection, enhancing efficiency in quality control for the tempe industry. Additionally, this technology holds potential for use in other food industries requiring image-based detection. Keywords: Deep Learning, Image Processing, MobileNetV2, Tempe, Tempe Quality Detection.]]>
Copyrights © 2025
