000			02708nam a22002297a 4500
003			ft8850
005			20251205151528.0
008			251205b ||||| |||| 00| 0 eng d
041			_aengtag
082			_a.
100	1		_a Hizole, Marc Gabriel V.; Panganiban, Justin Razzi V.
245			_aSmart pet micturition behavior corrector: Utilization of computer vision for pet urinary activity detection, with adaptive deterrence, and progress monitoring
264		1	_a. _b. _cc2025
300			_bCapstone Project: (Bachelor of Science in Information Technology) - Pamantasan ng Lungsod ng Maynila, 2025
336			_2text _atext _btext
337			_2 unmediated _a unmediated _b unmediated
338			_2 volume _a volume _b volume
505			_aABSTRACT: Pet owners often struggle to prevent their pets from urinating in unwanted areas, especially since they cannot always be present to monitor and deter this behavior in real time. This study addresses this issue by developing an automated system for real-time detection, deterrence, and monitoring of pet urination activity. The study had three key objectives: to develop a real-time pet urination detection system using YOLOv8 to identify and record pets urination events based on visual cues, to create an adaptive audio deterrence system that adjusts alert intensity based on the severity of urination behavior, and to provide pet owners with progress reports, including a dashboard with urination activity logs and heat maps. The system was developed using Django for the backend and Arduino IDE for the ESP32. Additionally, the AGILE Scrum SDLC methodology was employed throughout the development process. The system uses a custom-trained object detection model in YOLOv8 for real-time identification of pet urination events, and a pre-trained YOLOv8 model along with Bot-SORT for general dog detection. For triggering adaptive audio alerts, an ESP32-controlled Bluetooth speaker. The system was effective in detecting urination activity through the combined custom-trained and pre-trained YOLOv8 models. Furthermore, the adaptive audio deterrence system was effective based on the frequency of urination incidents, with louder deterrents for repeated offenses and softer cues for minor ones. Moreover, the third objective was realized through the system’s dashboard which tracked pet urination activity, displaying data through a pie chart and a table of comprehensive information of urination incidents. Finally, the heat map feature provided real-time data visualizing frequent urination spots. These demonstrated the system’s success in fulfilling the study’s objectives.
526			_aF
655			_aacademic writing
942			_2lcc _cMS
999			_c37291 _d37291