Please use this identifier to cite or link to this item:
https://dspace.ctu.edu.vn/jspui/handle/123456789/45778
Title: | SECURITY CAMERA EMBEDDED MODULES |
Authors: | Trần, Công Án Nguyễn, Văn Lợi |
Keywords: | CÔNG NGHỆ THÔNG TIN |
Issue Date: | 2021 |
Publisher: | Trường Đại Học Cần Thơ |
Abstract: | Video surveillance systems have come a long way from fixed security cameras. This technology now functions as miniature computers, offering features such as motion sensors and automatic mobile notifications or automatically contacting law enforcement. The goal of this project was to develop a fast and light-weight control program that can embedded into low cost and low power device. The program contains three main modules: Object detection module, streaming module, state machine based controlling module. Streaming module was implemented by using FFMPEG library. The streaming module was able to stream live video from camera to relay server throw RMTP protocol with around 10 seconds latency. Object detection is key feature in this project. Most state of the art models for object detection, however, are computationally complex. The goal of this module was to develop a fast and light-weight framework for object detection in a sequence of images using a Raspberry Pi 3 Model B, a low cost and low power computer. As even the most light-weight state of the art object detection models, i.e. Tiny-YOLO and SSD300 with MobileNet, were considered too computationally complex, a simplified approach had to be taken. This approach assumed a stationary camera and access to a background image. With these constraints, background subtraction was used to locate objects, while a light weight object recognition model based on MobileNet was used to classify any objects that were found. A tracker that primarily relied on object location and size was used to track distinct objects between frames. The suggested framework was able to achieve framerates as high as 1.5 FPS with 1 object in the scene, and 1.3 FPS when 4 objects were present. These values are significantly higher than those achieved using the mentioned state of the art models. This performance, however, comes at a price. While the suggested framework was seen to work well in many situations, it does have several weaknesses. Some of these include poor handling of occlusion, a lack of ability to distinguish between objects in close proximity, and false detections when lighting conditions change. Additionally, its processing speed is affected by the number of objects in an image to a larger degree than what the state of the art models are. None of the mention models have deterministic processing speeds.. |
Description: | 61 Tr |
URI: | https://dspace.ctu.edu.vn/jspui/handle/123456789/45778 |
Appears in Collections: | Trường Công nghệ Thông tin & Truyền thông |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
_file_ Restricted Access | 2.74 MB | Adobe PDF | ||
Your IP: 3.138.178.162 |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.