From studies of single cells to investigations of large organisms, many biologists study digital imagery to understand how life interacts with the environment. As biological research becomes more dependent on technology, the need for rapid analysis of digital data grows. Unfortunately, the software needed for many analyses is either prohibitively expensive or requires substantial training. This project will generate image analysis software that will allow biologists to easily perform measurements on digital photographs. The software generated will have an intuitive user interface with built-in instructions developed by biologists, it will contain a wide range of analysis tools that will allow users to determine the position, distance, area, angles, and colors on selected objects, and it will be freely available. A second facet of this project involves the creation of sophisticated tracking and motion analysis software. By providing these digital analysis tools for a broad population of biologists this project will advance research and education from top-tier institutions to high school laboratories.
The software will automate most repetitive tasks and provide common calculations from within so that data can be quickly extracted. In the tracking programs these calculations include static parameters, such as count, position, and size, and kinematic parameters, such as velocity and acceleration. Additionally, new tracking algorithms will be developed that utilize a variety of segmentation techniques to resolve overlapping objects and improve object identification. A newly devised morphed segmentation technique will be tested and implemented. This technique has the potential to generate more accurate counting and tracking of complex objects, such as animals in flight. The algorithm will use stored images from frames where the animals are moving individually. Then, when overlaps occur, individual animals will be approximated based on the stored images. The software developed here will be principally tested on data from research on the flight behavior of bats in groups. The results from this research will provide key data to understand the rules governing guidance behavior and collision avoidance in bats. Additionally, this project will involve two supplemental collaborations with investigators researching biology on the microscopic scale. Together, these collaborations will motivate refinements to the software so that it is suitable to meet the diverse needs of biological research. The goal of this project is to develop image and video analysis programs that can become the go-to free image analysis software for biologists. The software produced by this work can be found at: www.saintmarys.edu/~ibentley/imageanalysis .
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
