发明名称 Automated and scalable object and feature extraction from imagery
摘要 Feature extraction of image data using feature extraction modules. The feature extraction modules may be provided in an architecture that allows for modular, decoupled generation and/or operation of the feature extraction modules to generate feature data corresponding to image data. In this regard, the feature extraction modules may communicate with a file system storing image data and feature data by way of a common interface format. Accordingly, regardless of the nature of the execution of the feature extraction module, each feature extraction module may be communicative by way of the common interface format, thereby providing a modular approach that is highly scalable, flexible, and adaptive.
申请公布号 US9141872(B2) 申请公布日期 2015.09.22
申请号 US201314024399 申请日期 2013.09.11
申请人 DigitalGlobe, Inc. 发明人 Marchisio Giovanni B.;Tabb Mark D.;Tusk Carsten;Koperski Krzysztof;Shafer Jeffrey D.
分类号 G06K9/00;G06K9/46;H04N1/407;G06T5/40 主分类号 G06K9/00
代理机构 Marsh Fischmann & Breyfogle LLP 代理人 Marsh Fischmann & Breyfogle LLP
主权项 1. A method for extracting feature data from very high resolution (VHR) remotely-sensed multispectral Earth imagery, comprising: storing VHR remotely-sensed multispectral Earth image data in a file system; providing a plurality of feature extraction modules, wherein each respective feature extraction module is operable to analyze the image data to generate feature data regarding the image data; accessing the VHR remotely-sensed multispectral Earth image data in the file system with one or more of the plurality of feature extraction modules, wherein the one or more feature extraction modules each access the file system using a common interface format; executing the one or more feature extraction modules to generate feature data corresponding to the VHR remotely-sensed multispectral Earth image data; and receiving the feature data at the file system from the one or more feature extraction modules, wherein the feature data is received from each of the one or more feature extraction modules using the common interface format; wherein at least one of the feature extraction modules comprises a clustering histogram feature extraction module for extraction of feature based on clustered and classified spetra values, and wherein the clustering histogram feature extraction module is operative to generate feature data by: assigning each pixel of an image into one of a plurality of predefined classes; identifying kernel surrounding each pixel containing a plurality of other pixels from the image, wherein each of the plurality of other pixels have been assigned into one of the plurality of predefined classes; building a histogram for each pixel based on the number of other pixels in each of the plurality of predefined classes relative to the pixel for which the histogram is built.
地址 Longmont CO US