Theses and Dissertations at Montana State University (MSU)
Permanent URI for this communityhttps://scholarworks.montana.edu/handle/1/732
Browse
3 results
Search Results
Item Smart wildlife monitoring: evaluating a camera trap enabled with artificial intelligence(Montana State University - Bozeman, College of Agriculture, 2024) Kaltenbach, Taylor Louise Gregory; Chairperson, Graduate Committee: Jared T. Beaver; Jeffrey C. Mosley (co-chair)Wildlife-livestock conflicts, including depredation, disease transmission, and resource competition, present significant challenges to both the ecological and economic aspects of ranching operations. These conflicts can undermine the sustainability of ranching operations as well as the conservation of wildlife in working landscapes. Leveraging timely and precise data on wildlife activity, distribution, and their interactions with livestock are crucial for enhancing ongoing conflict mitigation efforts and to help sustain wildlife on working landscapes. I evaluated the potential of an artificial intelligence (AI)-enabled camera trap to limit false positive images and provide real-time monitoring of wildlife presence while reducing data overload. In Study 1, I compared the performance of a prototype, edge AI-enabled camera trap (Grizzly Systems) with 2 traditional, non-AI camera traps (Browning and Reconyx) at 8 sites across 3 ranches in south-central Montana, USA, from mid-June through mid-September 2023. I also evaluated the influence of site-specific environmental conditions, including air temperature, wind speed, cloud cover, and vegetation type on camera trap performance. The Grizzly Systems camera trap captured fewer false positive images but exhibited a higher rate of missed detections compared to the Browning and Reconyx camera trap models. Across all 3 camera trap models, the probability of positive detections declined with warmer air temperatures and greater wind speeds. In addition, warmer air temperatures positively influenced missed detections by Reconyx and Grizzly Systems camera trap models, but warmer air temperatures negatively influenced missed detections by Browning camera traps. In Study 2, I compared the performance of a cellular-connected AI-enabled Grizzly Systems camera trap, equipped with an automated image processing and notification reduction workflow, to a traditional, non-AI, cellular-connected Reconyx camera trap at 2 sites in south-central Montana, USA from mid-April to mid-June 2023. The AI-enabled, cellular-connected Grizzly Systems camera trap successfully sent real-time notifications of wildlife presence and transmitted significantly fewer false positive images than the cellular-connected Reconyx camera trap. However, the Grizzly Systems camera trap sent substantially fewer notifications of positive detections than the Reconyx camera trap, which are likely attributed to missed detections by the Grizzly Systems camera trap.Item The psychology of camera observation: how the camera affects human behavior(Montana State University - Bozeman, College of Arts & Architecture, 2021) Trainor, Catherine; Chairperson, Graduate Committee: Theo LipfertThis paper explores the influence of an observational camera on human behavior, particularly in documentary films. Whether it is a surveillance camera that represents the eyes of an authority figure, or a camera with a human operator, the presence of an observer impacts our behavior. The paper hypothesizes that the presence of a camera activates the same pathway in the brain as when a person senses that they are being watched. The paper uses observations from several documentary films, reality television shows, and the author's documentary film as supporting evidence in exploring this concept.Item Development of a smart camera system using a system on module FPGA(Montana State University - Bozeman, College of Engineering, 2017) Dack, Connor Aquila; Chairperson, Graduate Committee: Ross K. SniderImaging systems can now produce more data than conventional PCs with frame grabbers can process in real-time. Moving real-time custom computation as close as possible to the image sensor will alleviate the bandwidth bottle-neck of moving data multiple times through buffers in conventional PC systems, which are also computation bottlenecks. An example of a high bandwidth, high computation application is the use of hyperspectral imagers for sorting applications. Hyperspectral imagers capture hundreds of colors ranging from the visible spectrum to the infrared. This masters thesis continues the development of the hyperspectral smart camera by integrating the image sensor with a field programmable gate array (FPGA) and by developing an object tracking algorithm for use during the sorting process, with the goal of creating a single compact embedded solution. An FPGA is a hardware programmable integrated circuit that can be reprogrammed depending on the application. The prototype integration involves the development of a custom printed circuit board to connect the data and control lines between the sensor, the FPGA, and the control code to read data from the sensor. The hyperspectral data is processed on the FPGA and is combined with the object edges to make a decision on the quality of the object. The object edges are determined using a line scan camera, which provides data via the Camera Link interface, and a custom object tracking algorithm. The object tracking algorithm determines the horizontal edges and center of the object while also tracking the vertical edges and center of the object. The object information is then passed to the air jet sorting subsystem which ejects bad objects. The solution is to integrate the hyperspectral image sensor, the two processing algorithms, and Camera Link interface into a single, compact unit by implementing the design on the Intel Arria 10 System on Module with custom printed circuit boards.