1. ARM-Based Video Intercom System with Next-Gen Human Presence Detection using Deep Learning : magistrsko deloMario Gavran, 2022, magistrsko delo Opis: This master's thesis presents an advanced video system with human presence detection based on deep learning and an ARM microcontroller. The objective of the thesis is to develop a system that works as a smart video intercom, which could be installed, e.g. on the entrance door, and autonomously alert the owner that a guest is in front of the door. The main goal is to use an AI algorithm, namely the neural network model on a constrained device, such as an ARM microcontroller, as their main advantage is lower power consumption and cost.
The thesis also describes commonly used methods to reduce the power and memory footprint and to implement and accelerate the deep learning algorithms more effectively. Further, the most notable deep learning hardware and some general platforms are described in more detail.
The thesis also presents the development of a human presence detection system based on an ARM microcontroller, VGA camera, and LCD, where Tensorflow Lite Micro, an open-source C++ framework for deploying deep learning models to embedded platforms and a pre-trained neural network model for person presence detection are used.
Ključne besede: TensorFlow Lite Micro, Video intercom system, ARM Cortex-M microcontroller, Human presence detection, Neural network
Objavljeno v DKUM: 08.07.2022; Ogledov: 783; Prenosov: 55
 Celotno besedilo (7,86 MB) |
2. Energy efficient system for detection of elephants with Machine Learning : master's thesisMarko Sagadin, 2020, magistrsko delo Opis: Human-Elephant Conflicts are a major problem in terms of elephant conservation.
According to WILDLABS, an average of 400 people and 100 elephants are killed every year in India alone because of them.
Early warning systems replace the role of human watchers and warn local communities of nearby, potentially life threatening, elephants, thus minimising the Human-Elephant Conflicts.
In this Master's thesis we present the structure of an early warning system, which consists of several low-power embedded systems equipped with thermal cameras and a single gateway.
To detect elephants from captured thermal images we used Machine Learning methods, specifically Convolutional Neural Networks.
The main focus of this thesis was the design, implementation and evaluation of Machine Learning models running on microcontrollers under low-power conditions.
We designed and trained several accurate image classification models, optimised them for on-device deployment and compared them against models trained with commercial software in terms of accuracy, inference speed and size.
While writing firmware, we ported a part of the TensorFlow library and created our own build system, suitable for the libopencm3 platform.
We also implemented reporting of inference results over the LoRaWAN network and described a possible server-size solution.
We finally a constructed fully functional embedded system from various development and evaluation boards, and evaluated its performance in terms of power consumption.
We show that embedded systems with Machine Learning capabilities are a viable solution to many real life problems.
Ključne besede: machine learning, microcontroller, on-device inference, thermal camera, low-power system
Objavljeno v DKUM: 06.01.2021; Ogledov: 1469; Prenosov: 174
Celotno besedilo (13,35 MB) |
3. Programmable ultrasonic sensing system for targeted spraying in orchardsDenis Stajnko, Peter Berk, Mario Lešnik, Viktor Jejčič, Miran Lakota, Andrej Štrancar, Marko Hočevar, Jurij Rakun, 2012, izvirni znanstveni članek Opis: This research demonstrates the basic elements of a prototype automated orchard sprayer which delivers pesticide spray selectively with respect to the characteristics of the targets. The density of an apple tree canopy was detected by PROWAVE 400EP250 ultrasound sensors controlled by a Cypress PSOC CY8C29466 microcontroller. The ultrasound signal was processed with an embedded computer built around a LPC1343 microcontroller and fed in real time to electro-magnetic valves which open/close spraying nozzles in relation to the canopy structure. The analysis focuses on the detection of appropriate thresholds on 15 cm ultrasound bands, which correspond to maximal response to tree density, and this was selected for accurate spraying guidance. Evaluationof the system was performed in an apple orchard by detecting deposits of tartrazine dye (TD) on apple leaves. The employment of programmable microcontrollers and electro-magnetic valves decreased the amountof spray delivered by up to 48.15%. In contrast, the reduction of TD wasonly up to 37.7% at some positions within the tree crown and 65.1% in the gaps between trees. For all these reasons, this concept of precise orchard spraying can contribute to a reduction of costs and environmental pollution, while obtaining similar or even better leaf deposits.
Ključne besede: air-assisted sprayer, ultrasound, algorithm, programmable, microcontroller, spray distribution, orchard
Objavljeno v DKUM: 22.06.2017; Ogledov: 2341; Prenosov: 490
Celotno besedilo (890,30 KB)
Gradivo ima več datotek! Več... |
4. A flexible microcontroller-based data acquisition deviceDarko Hercog, Bojan Gergič, 2014, izvirni znanstveni članek Opis: This paper presents a low-cost microcontroller-based data acquisition device. The key component of the presented solution is a configurable microcontroller-based device with an integrated USB transceiver and a 12-bit analogue-to-digital converter (ADC). The presented embedded DAQ device contains a preloaded program (firmware) that enables easy acquisition and generation of analogue and digital signals and data transfer between the device and the application running on a PC via USB bus. This device has been developed as a USB human interface device (HID). This USB class is natively supported by most of the operating systems and therefore any installation of additional USB drivers is unnecessary. The input/output peripheral of the presented device is not static but rather flexible, and could be easily configured to customised needs without changing the firmware. When using the developed configuration utility, a majority of chip pins can be configured as analogue input, digital input/output, PWM output or one of the SPI lines. In addition, LabVIEW drivers have been developed for this device. When using the developed drivers, data acquisition and signal processing algorithms as well as graphical user interface (GUI), can easily be developed using a well-known, industry proven, block oriented LabVIEW programming environment.
Ključne besede: data acquisition, DAQ, microcontroller, analogue-to-digital converter, ADC, USB, HID, LabVIEW, GUI, data logging
Objavljeno v DKUM: 22.06.2017; Ogledov: 1462; Prenosov: 389
Celotno besedilo (1,98 MB)
Gradivo ima več datotek! Več... |
