Gutter oil detection for food safety based on multi-feature machine learning and implementation on FPGA with approximate multipliers

• Published in: PeerJ. Computer science Vol. 7; p. e774
• Main Authors: Jiang, Wei, Ma, Yuhanxiao, Chen, Ruiqi
• Format: Journal Article
• Language: English
• Published: United States PeerJ. Ltd 16.11.2021; PeerJ, Inc; PeerJ Inc
• Subjects: Algorithms; Analytical chemistry; Approximate multiplier; Classification; Cooking; Data mining; Data Mining and Machine Learning; Data processing; Digital integrated circuits; Drains; Embedded Computing; Embedded systems; Fatty acids; Field tests; Food safety; FPGA; Gutter oil detection; Gutters; K-NN; Laboratories; Laboratory tests; Machine learning; Oils & fats; Power consumption; Real-Time and Embedded Systems; Research methodology; Semiconductor industry; Source code; Three dimensional models; Vegetable oils; United States; China; K-NN; FPGA; Approximate multiplier; Machine learning; Gutter oil detection
• ISSN: 2376-5992; 2376-5992
• DOI: 10.7717/peerj-cs.774

Since consuming gutter oil does great harm to people’s health, the Food Safety Administration has always been seeking for a more effective and timely supervision. As laboratory tests consume much time, and existing field tests have excessive limitations, a more comprehensive method is in great need. This is the first time a study proposes machine learning algorithms for real-time gutter oil detection under multiple feature dimensions. Moreover, it is deployed on FPGA to be low-power and portable for actual use. Firstly, a variety of oil samples are generated by simulating the real detection environment. Next, based on previous studies, sensors are used to collect significant features that help distinguish gutter oil. Then, the acquired features are filtered and compared using a variety of classifiers. The best classification result is obtained by k-NN with an accuracy of 97.18%, and the algorithm is deployed to FPGA with no significant loss of accuracy. Power consumption is further reduced with the approximate multiplier we designed. Finally, the experimental results show that compared with all other platforms, the whole FPGA-based classification process consumes 4.77 µs and the power consumption is 65.62 mW. The dataset, source code and the 3D modeling file are all open-sourced.