You are here:
Harmful algal bloom smart device application: using image analysis and machine learning techniques for early classification of harmful algal blooms (SETAC presentation)
Waters, M., M. Steinitz-Kannan, Jim Lazorchak, AND Joel Allen. Harmful algal bloom smart device application: using image analysis and machine learning techniques for early classification of harmful algal blooms (SETAC presentation). SETAC World Congress, Orlando, FL, November 06 - 10, 2016.
This presentation will explain how a new iPhone App is being developed to detect bluegreen algal blooms for use by Citizen Scientists as well as monitoring crews
Reports of toxic cyanobacterial blooms, also known as Harmful Algal Blooms (HABS) have increased drastically in recent years. HABS impact human health from causing mild allergies to liver damage and death. The Ecological Stewardship Institute (ESI) at Northern Kentucky University is developing a smart device application that will permit accurate and quick identification of potential HABS. This new application, titled HAB APP, will be used to assist identification of HABS in recreational and drinking water supplies. Using support vector machine computer learning algorithms, the smart device extracts a color histogram from an image and compares it with a pre-loaded trained model of images for classification. More specifically, the algorithm distinguishes between relatively harmless green and potentially harmful blue-green algae. The algorithm will be extended to classify images taken from cameras situated at a nearby lake and along the Ohio River and will include other classes of algae.
Record Details:Record Type: DOCUMENT (PRESENTATION/SLIDE)
Organization:U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
NATIONAL EXPOSURE RESEARCH LABORATORY
SYSTEMS EXPOSURE DIVISION
ECOSYSTEM INTEGRITY BRANCH