公开(公告)号：US10089712B2

公开(公告)日：2018-10-02

申请号：US15281149

申请日：2016-09-30

Applicant: ORACLE INTERNATIONAL CORPORATION

Inventor： Avishkar Misra ,  Zazhil Ha Herena Ulloa ,  Juan Carlos Reyes Martinez ,  Siva Ravada

IPC: G06K9/00 ,  G06T3/00 ,  G06T7/00 ,  G06T5/00 ,  G06T5/30 ,  H04N5/232

Abstract: Systems, methods, and other embodiments are disclosed for correcting errors in the geo-spatial locations of acquired image data. In one embodiment, acquired aerial or satellite image data is segmented to generate extracted boundary data. The extracted boundary data represents boundaries of features of a portion of the Earth's surface, but at incorrect geo-spatial coordinates. The extracted boundary data is matched to expected boundary data derived from ground truth data. The expected boundary data represents boundaries of the features at correct geo-spatial coordinates. Adjustment parameters are generated that represent a geo-spatial misalignment between the extracted boundary data and the expected boundary data. Metadata in a header of the acquired image data is modified to include the adjustment parameters. The adjustment parameters may be applied to the acquired image data to generate corrected image data at correct geo-spatial coordinates.

公开(公告)号：US20190243799A1

公开(公告)日：2019-08-08

申请号：US15887234

申请日：2018-02-02

Applicant: Oracle International Corporation

Inventor： Kenny C. Gross ,  Mengying Li ,  Alan Paul Wood ,  Steven T. Jeffreys ,  Avishkar Misra ,  Lawrence L. Fumagalli, JR.

IPC: G06F15/18 ,  G06F17/30 ,  H04W4/38 ,  G06F17/18 ,  G06F17/14 ,  G06K9/62 ,  G05B19/048

CPC classification number: G06N20/00 ,  G05B19/048 ,  G06F16/2474 ,  G06F17/14 ,  G06F17/18 ,  G06K9/6256 ,  H04W4/38

Abstract: The disclosed embodiments relate to a system that facilitates development of machine-learning techniques to perform prognostic-surveillance operations on time-series data from a monitored system, such as a power plant and associated power-distribution system. During operation, the system receives original time-series signals comprising sequences of observations obtained from sensors in the monitored system. Next, the system decomposes the original time-series signals into deterministic and stochastic components. The system then uses the deterministic and stochastic components to produce synthetic time-series signals, which are statistically indistinguishable from the original time-series signals. Finally, the system enables a developer to use the synthetic time-series signals to develop machine-learning (ML) techniques to perform prognostic-surveillance operations on subsequently received time-series signals from the monitored system.