Joel Sartwell

Real-time video recording of food selection by captive white-tailed deer

Abstract Knowledge of animal behavior and resource selection is most accurate when direct observations of animals are made. It is impractical, if not impossible, however, to directly and efficiently observe free-ranging animals for extended periods of time without affecting their behavior. To help address these difficulties, we designed a miniature animal-mounted wireless video camera system that remotely records a portion of the sighting field of white-tailed deer (Odocoileus virginianus). The system consisted of a miniaturized charge-coupled-device (CCD) video camera, a miniaturized UHF transmitter (channel 30 or 42), a light-activated on-off relay circuit, protective plastic housing, a combination VCR/TV, a UHF TV antenna on a 9.1-m mast, and 12 9V lithium-ion DC batteries on a leather neck collar. The real-time color video image was transmitted up to 500 m and recorded on VHS tape. Using this system, we videotaped daytime observations of white-tailed deer food choices for up to 2 weeks. We assessed food choices from 70 hours of video footage from 1 antlered deer recorded during fall 2002 to demonstrate the utility of the technique. Our video system allowed us to monitor detailed food choices without the logistical difficulties present in direct observational studies. This tool could prove useful in studying detailed behavioral observations of captive animals; future development of electronic components could offer applications to free-ranging deer.

Energy-aware portable video communication system design for wildlife activity monitoring

In this paper, we introduce our recent research and development effort on energy-efficient portable video communication system design for wildlife activity monitoring. The capability of seeing what an animal sees in the field is very important for wildlife activity monitoring and research. We design an integrated video and sensor system, called DeerCam and mount it on animals so as to collect important video and sensor data about their activities in the field. From the video and sensor data collected by DeerCam, wildlife researchers will be able to extract a wealth of scientific data for studying the behavior patterns of wildlife species and understanding the dynamic of wildlife systems. We present the system architecture of DeerCam, explain our system design goals, and discuss major design issues. One of the central challenges in DeerCam system design is energy minimization. We present a new approach for energy minimization of portable video devices: power-rate-distortion (P-R-D) analysis and optimization. We discuss various approaches to minimizing the energy consumption of DeerCam, which can be also applied to other portable video devices. Results demonstrate that, by incorporating the third dimension of power consumption into conventional rate-distortion (R-D) analysis, P-R-D analysis gives us one extra dimension of flexibility in resource allocation and energy minimization and allows us to significantly reduce energy consumption.

Physiological Stress Response of Captive White-Tailed Deer to Video Collars

Abstract Animal-borne video and environmental data collection systems (AVEDs) are an advanced form of biotelemetry that combines video with other sensors. As a proxy for physiological stress, we assessed fecal glucocorticoid metabolite (FGM) excretion in 7 white-tailed deer (Odocoileus virginianus) fitted with AVED dummy collars; 9 additional deer served as controls. We collected fecal samples over 3 2-week periods: pretreatment, treatment, and posttreatment periods. There were no differences in FGMs across time periods (F2,218 = 1.94, P = 0.147) and no difference between FGMs of control and treatment individuals (F1,14 = 0.72, P = 0.411). Fecal glucocorticoid metabolite excretion in AVED-collared deer was indistinguishable from uncollared animals and within the normal, baseline range for this species. Absence of an adrenal response to collaring suggested that AVED collaring does not induce physiological stress in deer.

Digital Image Vegetation Analysis with Machine Learning

We propose computer vision based approach for effectively computing the vegetation coverage of the image to determine the structure of the vegetation and to understand wildlife habitat. To deal with the variation of lighting condition, two-stage segmentation strategy is applied. Firstly, texture information is used to roughly classify the vegetation and the reference blackboard at each position using Support Vector Machine. And then a K-means based adaptive color model is used to refine the segmentation result in pixel level. We evaluate our approach on our dataset, and the results demonstrate that the proposed method is robust to environment changing, and color instability. For blackboard localization, we tested 200 images and the accuracy is approximately 93%. For grass detection and coverage computation, the error rate is approximately 3%.

Mobile Data Collection and Analysis in Conservation

Mobile computing and big data analytics have great potential for improving efficiency, productivity, and knowledge discovery in conservation tasks. This paper presents some recent development of mobile computing and data analysis systems for the Missouri Department of Conservation (MDC), including a mobile data collection system, an improved bear tracking website, and new analytics results obtained from real Missouri bear and deer GPS trajectories. The mobile data collection system using Android tablets is developed for surveying the usage and status of conservation areas. The new bear tracking website is developed to be mobile friendly and can dynamically present information of tracked bears. Lastly, for analyzing real GPS trajectories obtained from Missouri bears and deer, a density-based spatial and temporal clustering method is developed for identifying stay regions in trajectories of low sampling rate GPS points. Using real world data, interesting movement patterns of a large number of bears and white-tailed deer have been obtained, therefore advancing a step in achieving a better understanding of the behaviors of various types of animals in their natural environments.