Master of Science, The Ohio State University, 2024, Computer Science and Engineering

Motion-activated cameras, commonly known as camera traps, play a crucial role in wildlife monitoring, biodiversity conservation, and basic ecological science, capturing a diverse range of species in their natural habitats. However, the sheer volume of images generated requires manual analysis by ecologists, making this a time-consuming and inefficient process. This is particularly frustrating since many of the images do not contain animals because camera traps are triggered by the motion of vegetation due to wind or miss the animal due to the slow camera response. This thesis presents an approach aimed at addressing these challenges through the automation of image classification and the development of smart camera traps that leverage artificial intelligence implementations in edge devices. First, we develop a species classifier pipeline consisting of object detection and a species classifier for a novel camera trap setting. We train a convolutional neural network to identify 45 trained species using 1.2M images, achieving an accuracy of around 89\% at the class level without manual intervention. This research demonstrates the combination of deep learning methodologies and human oversight, showcasing their ability to facilitate automatic or semiautomatic, precise, and efficient surveys of small animals in the wild, particularly in novel settings or for novel species. The suggested pipeline helps achieve 95\% accuracy in image classification, resulting in 11\% of images that require manual review at the class level. This pipeline helps to automatically annotate the image while maintaining accuracy. The camera trap generates a large number of images. The time delay between data capture and image processing leads to the loss of critical information. Real-time labeling while capturing images can help mitigate this issue. Following the traditional classifier, we investigate methods for deploying an AI model in a camera trap. Smart Camera Traps would result in real (open full item for complete abstract)

Committee: Tanya Berger-Wolf (Advisor); Christopher Stewart (Committee Member); Wei-Lun Chao (Committee Member) Subjects: Computer Engineering

Master of Science (MS), Bowling Green State University, 2012, Geology

Predicting and mapping invasive wetland plant species is an important process for future management decisions and strategies. Controlling and mapping such plant species requires robust methods that are applicable at different ecological scales to map and monitor their spread. In particular, this study tested the feasibility of classification tree analysis (CTA) by using a high resolution Applanix 439 Digital Sensor System (DSS) aerial imagery (< 20 cm) and linear spectral unmixing (LSU) analysis by using Landsat Thematic Mapper (TM) data to produce different distribution maps of invasive flowering rush (Butomus umbellatus L.) potential in the Ottawa National Wildlife Refuge (ONWR) wetlands, in Northwest Ohio. The classification accuracy from CTA maps derived from different splitting rules was evaluated by kappa statistics. The overall accuracy within the different runs varied between 35 to 56 % while the “Gini” splitting rule had the best performance. The endmembers from the best CTA performing map were utilized by the LSU method for estimating sub-pixel endmember fractions at a broader geographical scale. The results derived from the aerial imagery were slightly better than those from the Landsat imagery, as the goodness of fit between the flowering rush fraction map and the data measured in the field was lower. This study was intended to demonstrate the potential for flowering rush mapping over larger area using knowledge developed from smaller geographical scale using high resolution imagery. Results indicate that both methods show promising results for the prediction of flowering rush, but additional research that encompass different field data collection techniques, datasets of imagery and modeling methods need to be explored.

Committee: Peter Gorsevski (Advisor); Helen Michaels (Committee Member); Enrique Gomezdelcampo (Committee Member) Subjects: Remote Sensing