NPS UNIT: South Florida Carribean Inventory and Monitoring Network
LOCATION: Palmetto Bay, FL
Computer Science Researcher working on natural resource Data Collection. The incumbent will have the ability to use software program to analyze photographic data and determine the present of nesting birds. Preferably using machine learning (i.e. artificial intelligence) to be able to determine if a bird nest is present on a photograph and if so graphically indicate that. Then to compare the software outcome against human processed Data Collection that is currently being done.
Nesting colonies of wading birds and seabirds are important indicators of ecosystem health, as they respond to changes in food abundance, food quality, contaminants, and disturbances. The acts of selecting mates, building nests, laying eggs and rearing chicks, are energy intensive activities. If the habitat is insufficient to support these activities, nesting success will suffer and may indicate a problem in the ecosystem.
The National Park Service, South Florida/Caribbean Inventory and Monitoring Network (SFCN) has been monitoring colonial nesting birds monthly via low level aerial photography from a helicopter platform in Biscayne National Park since 2010. These photographs are processed to determine the number of active nests for specific focal species: Double-crested Cormorants, Great Egrets, Great White Herons, Great Blue Herons, White Ibises, and Roseate Spoonbills. We now have 10 years of nesting data and have an extensive data set of bird nesting data (44,376 nest and 45,317 birds representing 35 taxon).
This monitoring is time consuming and needs to be accomplished consistently and accurately. Currently, the aerial photos are processed by individuals. We propose to have the intern explore the creation of automatic script that would allow these nest to be identified using computer software. The goal of this internship is to explore software programs and create automatic script which could indicate potential bird nest is present on the photograph and to highlight this area on the photograph. Ideally by using a machine learning model trained on previous data with a program such as the open-source annotation service called Accelerating image-based ecology (AIDE) as the user interface and then using Microsoft Azure like cloud computing to "point out" the active bird nests to train a machine learning model capable of identifying birds' nests. The machine learning model will automatically draw bounding boxes around the bird nests, and the user will see it on AIDE. The end goal being that data processing and allows the U.S. National Park Service to more quickly and accurately monitor the nesting pattern of these colonial birds
The objective of the internship is to understand if automatic software can find bird nest and correctly count the number of bird nest. The intern will: 1) train the software to find bird nest based on the already processed photographs. 2) assess the accuracy of the software found bird nest compared to a traditional processing method. 3) report on next steps if they are needed. The intern will present these results to park Resource management staff.
Learning Goals:
The intern will gain the skills necessary to run an analysis on a large dataset with the goal of producing a well document product that will allow a computer software program to accelerate Data Collection and increase repeatability in that Data Collection.
The intern will more fully develop computer science type skills, image recognition, machine learning, interacting with professional in the field of image analysis.
The intern will develop and refine coding skills and statistical analysis abilities. The intern will develop the ability to develop code to automate graphical display and to facilitate analysis of the data set for long term monitoring and for event driven impacts. The intern will generate and follow standard operating procedures to perform the above activities.
The intern will write a document reporting the findings.
Once the project is completed the intern will present the results to the Science and Technical Board member from the park being monitored. The presentation will indicate the general results along with any caveats that need to be understood.