The Ivory Billed Woodpecker is a magnificent creature, believed to be extinct, that is of great interest to birdwatchers, ornithologists, and conservationists. The last confirmed U.S. sighting was in the 1940s. In January 2004, a potential sighting occurred in the Bayou DeView wildlife refuge in Arkansas, prompting comprehensive and systematic search led by researchers at Cornell University. In Fall 2005, we offered to assist the search effort by developing a high resolution robotic video system to observe the sky over an extended time period. We are now collaborating with the Cornell team and designing a new camera system in conjunction with our existing NSF project to develop collaborative observatories for natural environments. Detailed high resolution video images are required to distinguish an ivory-billed woodpecker from its cousin, the common pileated woodpecker. Our goal is to develop a computer vision system that detects when birds fly into the field of view, recording the associated video segments, and discarding video where there are no birds present. Subsequent analysis by human experts (perhaps with computer post processing) will be required to sift through the Gigabytes of video data collected. The system is to be installed in a clearing in the Bayou DeView, which is mostly swamp. Some challenges are: lack of network connectivity, variations and noise in sky images, limited power supply, high temperatures and moisture conditions. The first prototype Automated Collaborative Observatory for Natural Environments, ACONE 1.0, will incorporate two Arecont Vision 3 Megapixel video cameras and one 1.4 GHz MiniITX computer. We installed the first pair of ACONE cameras in October 2006.

 

 

Our group focuses on developing autonomous observatories to assist nature scientists to search rare birds in remote environments such as in a deep forest. Due to power and communication constraints, it is often prohibitive to install dislocated multiple cameras in deep forest to form stereo pairs. Also, the calibration and the precise synchronization of dislocated stereo cameras are very difficult. Therefore, monocular vision is preferred in such settings. Since the final bird identification has to be performed by human experts, it is necessary to develop a filter to reduce the huge volume of the video data to a manageable size for human experts. Since the chance of the rare bird flying in front of the camera is very small and missing the opportunity would be costly, the system requires an asymmetric filter design that emphasizes very low false negative with manageable false positive.

 

 

Consider a wide-angle camera is installed at an airport for human activity surveillance or in a forest for wildlife observation. The wide-angle camera can provide large, low resolution coverage of the scene. However, recognition and identification of humans and animals usually require close-up views at high resolution which needs PTZ cameras. The resulting autonomous observation system consists of a fixed wide-angle camera with multiple PTZ cameras as illustrated in figure. The wide-angle camera monitors the entire field to detect and track all moving objects. Each PTZ camera selectively covers a subset of the objects.

 

 

We build and maintain a flying bird video data set to support the research on biotelemetry and other potential fields. Details of each video clip and corresponding biological information of the bird species are provided. We hope users find this data set useful for relevant research. We also hope this data set would inspire new ideas and bring interets and insights into our research. Use or download of the content in this data set for research purpose is completely free.