Texture Classifying Neural Network Algorithm (TCNNA) from Synthetic Aperture Radar (SAR) nearshore potential oiling footprints collected collected during the Deepwater Horizon oil spill response from April 2010 to August 2010 in the Northern Gulf of Mexico (NCEI Accession 0163819)

This archival information package (AIP) contains Environmental Response Management Application (ERMA) GIS layers of outputs from Synthetic Aperture Radar (SAR) imagery that has been processed using the Texture Classifying Neural Network Algorithm (TCNNA). This algorithm classifies SAR data on a pixel by pixel basis, into oil or not-oil classes. The fully implemented TCNNA routine produces approximately a one megabyte georectified, raster image in which all pixels receive a binary classification of 1 for no oil and 0 for oil. Each classified raster image was converted to polygons and clipped to within approximately 20 kilometers of the coastline. This data was collected from April 29th 2010 to August 11th 2010. These data were collected during the response to the Mississippi Canyon 252 Deepwater Horizon oil spill in the Northern Gulf of Mexico and used as part of the Programmatic Damage Assessment and Restoration Plan (PDARP).本档案信息包（Archival Information Package, AIP）包含经纹理分类神经网络算法（Texture Classifying Neural Network Algorithm, TCNNA）处理得到的合成孔径雷达（Synthetic Aperture Radar, SAR）影像输出成果的环境响应管理应用程序（Environmental Response Management Application, ERMA）地理信息系统图层。该算法以像素为单位对SAR数据进行分类，将其划分为含油与不含油两类。完整执行的TCNNA流程可生成约1MB大小的地理校正栅格图像，其中所有像素均被赋予二分类标签：1代表不含油，0代表含油。每张经分类的栅格图像均被转换为多边形要素，并裁剪至距海岸线约20公里的范围内。本数据集采集于2010年4月29日至2010年8月11日期间，采集工作是在针对墨西哥湾北部密西西比峡谷252号海域深水地平线漏油事故的应急响应过程中开展的，并被纳入《项目性损害评估与恢复计划》（Programmatic Damage Assessment and Restoration Plan, PDARP）作为组成部分。