AIoT-Based Meat Quality Monitoring Using Camera and Gas Sensor With Wireless Charging

• DOI: 10.1109/jsen.2023.3328915
• Publication Date: 2023-11-28T19:32:44Z
• Authors: Dong-Eon Kim; Ngoc-Dau Mai; Wan-Young Chung

Automated monitoring technology is necessary to ensure the safe consumption and management of meat. Here we propose a system for determining food freshness on basis data collected by embedded cameras gas sensors using artificial intelligence things (AIoT). Using DeepLab V3+ model, images meat obtained from underwent semantic segmentation into lean fat, revealing nature rotting over time. convolutional neural network (CNN) learning, classified fresh, semi-fresh, rotten detected transition among these. The Raspberry Pi server online sensor data, which was then processed mobile application. were at 1-minute intervals, means standard deviations valuable features extracted. Meanwhile, camera preprocessed, their transfer-learning approach MobileNet as feature extractor. Next, proposed one-dimensional CNN trained combined predicting different quality states. accuracy detection deep learning model reaches up 99.44%. Our Internet Things (IoT)−connected platform, including an device, database server, smartphone application, allows quick easy quality. experimental results confirmed viability our deep-learning−based meat-monitoring with IoT applications in real world.