From Event: SPIE Commercial + Scientific Sensing and Imaging, 2017
A hyperspectral image provides multidimensional figure rich in data consisting of hundreds of spectral dimensions. Analyzing the spectral and spatial information of such image with linear and non-linear algorithms will result in high computational time. In order to overcome this problem, this research presents a system using a MapReduce-Graphics Processing Unit (GPU) model that can help analyzing a hyperspectral image through the usage of parallel hardware and a parallel programming model, which will be simpler to handle compared to other low-level parallel programming models. Additionally, Hadoop was used as an open-source version of the MapReduce parallel programming model. This research compared classification accuracy results and timing results between the Hadoop and GPU system and tested it against the following test cases: the CPU and GPU test case, a CPU test case and a test case where no dimensional reduction was applied.
Andres Ramirez and Maryam Rahnemoonfar, "Classification of hyperspectral imagery using MapReduce on a NVIDIA graphics processing unit (Conference Presentation)," Proc. SPIE 10213, Hyperspectral Imaging Sensors: Innovative Applications and Sensor Standards 2017, 102130D (Presented at SPIE Commercial + Scientific Sensing and Imaging: April 12, 2017; Published: 9 June 2017); https://doi.org/10.1117/12.2268363.5460497456001.
Conference Presentations are recordings of oral presentations given at SPIE conferences and published as part of the conference proceedings. They include the speaker's narration along with a video recording of the presentation slides and animations. Many conference presentations also include full-text papers. Search and browse our growing collection of more than 12,000 conference presentations, including many plenary and keynote presentations.