International Business Machines Corporation

Johannes Walter Schmude

Hendrik F Hamann

Siyuan Lu

Conrad M Albrecht

Akihisa Sakurai

Taku Izumiyama

Masao Hasegawa

Oki Gunawan

Fernando Jimenez Marianno

Marcus Oliver Freitag

Carlo Siebenschuh

From each of a plurality of cameras, a visual input of a location is received over a network. For each visual input from the plurality of cameras, a coupling correction is performed between a shaking of the camera with respect to the visual input by subtracting velocity vectors of the plurality of cameras from velocity vectors of pixels defining the visual input to provide a processed input. It is determined whether a shaking identified in the processed input is above a predetermined threshold based on the processed input, thereby detecting one or more anomalies. From the one or more anomalies, at least one of a location, magnitude, or depth of an earthquake are inferred based on the shaking identified in the processed input of each of the plurality of cameras.

Earthquake detection and response via distributed visual input

A computer produces predictions throughout a raster field in response to point data, by obtaining a partially empty matrix of point data, filling a matrix of extrapolated raster data by dilating the point data in a first convolutional neural network, and generating a matrix of aggregate raster data by combining the extrapolated raster data with organic raster data in a second convolutional neural network.

Combining point observations with raster data for machine learning

A method of generating an aggregate forecast includes obtaining historical forecasts for a number of time steps and at least one location, obtaining historical conditions for the time steps and the at least one location, training a machine learning algorithm to produce an aggregate historical forecast in response to the historical conditions and the historical forecasts, and producing an aggregate current forecast by running the trained machine learning algorithm on current forecasts. The historical forecasts and the current forecasts vary in at least one of spatial resolution or temporal resolution, and include a first forecast that is valid for a first time step and a second forecast that is valid for a second time step.

Combining forecasts of varying spatial and temporal resolution

A computer-implemented method includes accessing, by a processing unit, an existing layer representing geospatial-temporal data at a selected timestamp. A first overview layer of the existing layer is generated by iteratively aggregating each cluster of cells of the existing layer into a corresponding lower-resolution cell of the first overview layer. The first overview layer therefore has a lower resolution than the existing layer. A query is received related to the geospatial-temporal data in the existing layer, and the query is processed with reference to the first overview layer.

Efficient querying using overview layers of geospatial-temporal data in a data analytics platform

Growing community of inventors

White Plains, NY, United States of America

Johannes Walter Schmude