Open AccessPERSON RE-IDENTIFICATION USING DEEP METRIC LEARNING
Author(s) -
Shruti Jalapur,
Bibi Ayesha Hundekar
Publication year - 2021
Publication title -
international journal of engineering applied science and technology
Language(s) - English
Resource type - Journals
ISSN - 2455-2143
DOI - 10.33564/ijeast.2021.v06i04.047
Subject(s) - safer , computer science , metric (unit) , identification (biology) , variety (cybernetics) , deep learning , artificial intelligence , computer security , machine learning , data science , human–computer interaction , engineering , operations management , botany , biology
Today everyone faces a multitude and varietyof threats ranging from robbery, kidnapping andterrorism to murder. To avoid these threats, authoritiesneed to collect real-time information about what's goingon in and around the city. New technologies aretherefore being developed to make cities safer and morerisk-free. Here we have built a reliable system thatrecognizes the person from every angle from a recordedimage. We can get the input into the systems throughCCTV cameras installed in public places where thesetypes of life threatening events take place. It is easy toinstall these cameras in public places, and it is easier tomonitor and store the data. The developed system usesdeep metric learning and the machine learning platform,Tensor Flow and Keras. It's a type of machine learningwhere the system iteratively performs calculations toknow the patterns. The system processes recordedimages and compares them with existing data records inorder to identify the person. The comparison is madebased on certain selected features. The results are moreaccurate (98.18%) compared to existing systems.