Sentrify: Microcontroller-based surveillance and alarm system efficientyolo-fine for hand signal detection

By: Alon, Patrick Miguell B.; Calso, Christian Paul R.; Corpuz, Kio B.; Disuanco, Ryan Paul L.; Lajom, Sean Francis D.; Ong, Franz Khyl M
Language: English Publisher: . . c2023Description: Undergraduate Thesis: (Bachelor of Science in Computer Engineering) - Pamantasan ng Lungsod ng Maynila, 2023Content type: text Media type: unmediated Carrier type: volumeGenre/Form: academic writingDDC classification: . LOC classification: TK7895 A46 2023
Contents:
ABSTRACT: Single Stage Convolutional Neural Networks (CNNs) have been lately dominating the field of object detection due to their real-time inference speeds while exhibiting competitive performance with heavy state-of-the-art multiple-stage models. In this paper, we use CNNs for gesture detection as a public security measure and equip them with microcontrollers to transform detections into meaningful by emitting light and sound signals. Particularly, we use a modified form of YOLOv3 (Redmon & Farhadi, 2018) to target small-scale object detection suited for frames captured from surveillance cameras. We train the model using over custom annotated images and the results demonstrate good generalizability. We integrate the model with ESP8266 and camera module to build a functional alarm system that can alert, record, and send detections using mailing systems.
Tags from this library: No tags from this library for this title. Log in to add tags.
    Average rating: 0.0 (0 votes)
Item type Current location Home library Collection Call number Status Date due Barcode Item holds
Thesis/Dissertation PLM
PLM
Filipiniana Section 		Filipiniana-Thesis TK7895 A46 2023 (Browse shelf) Available FT8824
Total holds: 0

ABSTRACT: Single Stage Convolutional Neural Networks (CNNs) have been lately dominating the field of object detection due to their real-time inference speeds while exhibiting competitive performance with heavy state-of-the-art multiple-stage models. In this paper, we use CNNs for gesture detection as a public security measure and equip them with microcontrollers to transform detections into meaningful by emitting light and sound signals. Particularly, we use a modified form of YOLOv3 (Redmon & Farhadi, 2018) to target small-scale object detection suited for frames captured from surveillance cameras. We train the model using over custom annotated images and the results demonstrate good generalizability. We integrate the model with ESP8266 and camera module to build a functional alarm system that can alert, record, and send detections using mailing systems.

Filipiniana

There are no comments for this item.

to post a comment.

© Copyright 2024 Phoenix Library Management System - Pinnacle Technologies, Inc. All Rights Reserved.