Identifikasi Kandungan Formalin dan Kesegaran Daging Sapi dengan Image Processing

Abstract

The lack of public understanding in distinguishing between fresh and spoiled meat, coupled with the unrestricted use of formalin, provides opportunities for unscrupulous traders to gain greater profits. Therefore, the development of image processing for beef freshness and formalin identification is necessary. This image processing utilizes the YOLO method, while formalin identification is done using an HCHO sensor. The system was tested every 2 hours, starting from 9:00 AM, revealing that at 9:00 AM, fresh meat was identified with a 87.89% accuracy, at 11:00 AM, fresh meat was identified with a 74.43% accuracy, and at 1:00 PM, spoiled meat was identified with a 38.85% accuracy. The results indicate that the longer the meat is left untreated, the more its freshness diminishes, eventually being identified as spoiled meat. For formalin identification in beef, results from 4 samples show that the longer the meat is exposed to formalin solution, the higher the volatility of the sensor, indicating the presence of formalin in the meat.