arXiv subjects

Scott Fazackerley

Publications and source records attributed to Scott Fazackerley.

4 recordsLinked to original sources

arXiv2026

Space-efficient B-tree Implementation for Memory-Constrained Flash Embedded Devices

Small devices collecting data for agricultural, environmental, and industrial monitoring enable Internet of Things (IoT) applications. Given their critical role in data collection, there is a need for optimizations to improve on-device data processing. Edge device computing allows processing of the data closer to where it is collected and reduces the amount of network transmissions. The B-tree has been optimized for flash storage on servers and solid-state drives, but these optimizations often require hardware and memory resources not available on embedded devices. The contribution of this work is the development and experimental evaluation of multiple variants for B-trees on memory-constrained embedded devices. Experimental results demonstrate that even the smallest devices can perform efficient B-tree indexing, and there is a significant performance advantage for using storage-specific optimizations.

cs.DB↗

arXiv2023

An Efficient B-tree Implementation for Memory-Constrained Embedded Systems

Embedded devices collect and process significant amounts of data in a variety of applications including environmental monitoring, industrial automation and control, and other Internet of Things (IoT) applications. Storing data efficiently is critically important, especially when the device must perform local processing on the data. The most widely used data structure for high performance query and insert is the B-tree. However, existing implementations consume too much memory for small embedded devices and often rely on operating system support. This work presents an extremely memory efficient implementation of B-trees for embedded devices that functions on the smallest devices and does not require an operating system. Experimental results demonstrate that the B-tree implementation can run on devices with as little as 4 KB of RAM while efficiently processing thousands of records.

cs.DB↗

arXiv2023

An Application of Deep Learning for Sweet Cherry Phenotyping using YOLO Object Detection

Tree fruit breeding is a long-term activity involving repeated measurements of various fruit quality traits on a large number of samples. These traits are traditionally measured by manually counting the fruits, weighing to indirectly measure the fruit size, and fruit colour is classified subjectively into different color categories using visual comparison to colour charts. These processes are slow, expensive and subject to evaluators' bias and fatigue. Recent advancements in deep learning can help automate this process. A method was developed to automatically count the number of sweet cherry fruits in a camera's field of view in real time using YOLOv3. A system capable of analyzing the image data for other traits such as size and color was also developed using Python. The YOLO model obtained close to 99% accuracy in object detection and counting of cherries and 90% on the Intersection over Union metric for object localization when extracting size and colour information. The model surpasses human performance and offers a significant improvement compared to manual counting.

cs.CV↗

arXiv2023

Estimation of Average Annual Daily Bicycle Count Using Bike-Share GPS Data and Bike Counter Data for an Urban Active Transportation Network

In 2018, the City of Kelowna entered into a license agreement with Dropbike to operate a dockless bike-share pilot in and around the downtown core. The bikes were tracked by the user's cell phone GPS through the Dropbike app. The City's Active Transportation team recognized that this GPS data could help understand the routes used by cyclists which would then inform decision-making for infrastructure improvements. Using OSMnx and NetworkX, the map of Kelowna was converted into a graph network to map inaccurate, infrequent GPS points to the nearest street intersection, calculate the potential paths taken by cyclists and count the number of trips by street segment though the comparison of different path-finding models. Combined with the data from four counters around downtown, a mixed effects statistical model and a least squares optimization were used to estimate a relationship between the different traffic patterns of the bike-share and counter data. Using this relationship based on sparse data input from physical counting stations and bike share data, estimations and visualizations of the annual daily bicycle volume in downtown Kelowna were produced. The analysis, modelling and visualization helped to better understand how the bike network was being used in the urban center, including non-traditional routes such as laneways and highway crossings.

cs.DB↗