Armil Monsura, A. R. Fernando, Rafael Ventura, Denise Antoinette Bañas, Dorothy Joy de Castro, John Michael Molleno, Krystal Mae Denise Rama, Eirron Carl Ramirez, Benjamin Norbert Vitug
{"title":"用颜色和化学特征检测掺假猪肉","authors":"Armil Monsura, A. R. Fernando, Rafael Ventura, Denise Antoinette Bañas, Dorothy Joy de Castro, John Michael Molleno, Krystal Mae Denise Rama, Eirron Carl Ramirez, Benjamin Norbert Vitug","doi":"10.1109/TENCON.2019.8929332","DOIUrl":null,"url":null,"abstract":"Pork in the Philippines is mostly sourced locally, and not all these meats come from well-sanitized slaughterhouses. Some carcasses of pigs that have died other than by slaughtering, called adulterated or double-dead meat, make their way to markets and get mixed up with legal and safe meat, only to be sold to unsuspecting consumers. These pose health hazards to meat handlers and consumers. One of the longtime practices in preventing the proliferation of double-dead meat in the markets is the manual inspection that analyzes the meat based on its organoleptic properties. This solution, however, is prone to limitations of subjectivity and human evaluation error. In this study, the proponents designed a system that will identify whether the meat sample is adulterated or not. The system is aimed at providing fast and accurate detection of double-dead meat using color and chemical determinants that is at least comparable to, or better than, humans. The application of the system can be utilized to replace the traditional human evaluation using electronic sensors as to avoid bias in the evaluation of the meat by different assessors as well as to eliminate the health risks on the meat inspectors and the consumers. The system utilizes three sensors: color sensor, methane sensor and pH level sensor. A microcontroller implements a logistic regression classifier. The model performs well, with an accuracy of 91.40% after validation, with an area of 0.95 under its ROC. When realized, the model has an accuracy of 93.18% compared to 84.09% of the human assessor's evaluation of the same meat samples.","PeriodicalId":36690,"journal":{"name":"Platonic Investigations","volume":"12 1","pages":"2099-2104"},"PeriodicalIF":0.0000,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Detection of Adulterated Pork Meat via Color and Chemical Characteristics\",\"authors\":\"Armil Monsura, A. R. Fernando, Rafael Ventura, Denise Antoinette Bañas, Dorothy Joy de Castro, John Michael Molleno, Krystal Mae Denise Rama, Eirron Carl Ramirez, Benjamin Norbert Vitug\",\"doi\":\"10.1109/TENCON.2019.8929332\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Pork in the Philippines is mostly sourced locally, and not all these meats come from well-sanitized slaughterhouses. Some carcasses of pigs that have died other than by slaughtering, called adulterated or double-dead meat, make their way to markets and get mixed up with legal and safe meat, only to be sold to unsuspecting consumers. These pose health hazards to meat handlers and consumers. One of the longtime practices in preventing the proliferation of double-dead meat in the markets is the manual inspection that analyzes the meat based on its organoleptic properties. This solution, however, is prone to limitations of subjectivity and human evaluation error. In this study, the proponents designed a system that will identify whether the meat sample is adulterated or not. The system is aimed at providing fast and accurate detection of double-dead meat using color and chemical determinants that is at least comparable to, or better than, humans. The application of the system can be utilized to replace the traditional human evaluation using electronic sensors as to avoid bias in the evaluation of the meat by different assessors as well as to eliminate the health risks on the meat inspectors and the consumers. The system utilizes three sensors: color sensor, methane sensor and pH level sensor. A microcontroller implements a logistic regression classifier. The model performs well, with an accuracy of 91.40% after validation, with an area of 0.95 under its ROC. When realized, the model has an accuracy of 93.18% compared to 84.09% of the human assessor's evaluation of the same meat samples.\",\"PeriodicalId\":36690,\"journal\":{\"name\":\"Platonic Investigations\",\"volume\":\"12 1\",\"pages\":\"2099-2104\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Platonic Investigations\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TENCON.2019.8929332\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"Arts and Humanities\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Platonic Investigations","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TENCON.2019.8929332","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Arts and Humanities","Score":null,"Total":0}
Detection of Adulterated Pork Meat via Color and Chemical Characteristics
Pork in the Philippines is mostly sourced locally, and not all these meats come from well-sanitized slaughterhouses. Some carcasses of pigs that have died other than by slaughtering, called adulterated or double-dead meat, make their way to markets and get mixed up with legal and safe meat, only to be sold to unsuspecting consumers. These pose health hazards to meat handlers and consumers. One of the longtime practices in preventing the proliferation of double-dead meat in the markets is the manual inspection that analyzes the meat based on its organoleptic properties. This solution, however, is prone to limitations of subjectivity and human evaluation error. In this study, the proponents designed a system that will identify whether the meat sample is adulterated or not. The system is aimed at providing fast and accurate detection of double-dead meat using color and chemical determinants that is at least comparable to, or better than, humans. The application of the system can be utilized to replace the traditional human evaluation using electronic sensors as to avoid bias in the evaluation of the meat by different assessors as well as to eliminate the health risks on the meat inspectors and the consumers. The system utilizes three sensors: color sensor, methane sensor and pH level sensor. A microcontroller implements a logistic regression classifier. The model performs well, with an accuracy of 91.40% after validation, with an area of 0.95 under its ROC. When realized, the model has an accuracy of 93.18% compared to 84.09% of the human assessor's evaluation of the same meat samples.