New Student Drug Tests at College Using Principal Component Analysis Method

Drugs are substances or illegal drugs that can endanger human life. Someone who consumes it in an inappropriate way will become dependent and even result in death. The physical characteristics of people who use drugs vary, but the more obvious characteristics are on the faces of drug users such as red eyes, stiff facial muscles, dark spots, pupils susceptible to light, sunken face shape, and dullness. The lack of physical characteristics of drug users due to similarities with other diseases makes it difficult for people to recognize them initially. However, for users whose face data has been tracked by the National Narcotics Agency, the facial data is stored in the dataset. This research was conducted with the aim of building a system that can detect and recognize prospective students whether they have ever been included in drug users recorded in the National Narcotics Agency dataset or not as one of the requirements for new student admissions to universities. The system built using the Principal Component Analysis method to process and extract images of the physical characteristics of drug users through the facial image data of drug users stored in the dataset. If the detected face has similarities with the characteristics in the dataset, it is necessary to suspect that the detected face is a drug user. The results of this study are the system is able to detect the faces of drug users using the Principal Component Analysis method with an accuracy of 90% and it is hoped that with this research the system can be one solution in helping universities as an identification effort to minimize drug use so that it can be an additional identification tool which strengthens someone detected using drugs.