Abstract: A system, method and program product for implementing a machine learning platform that processes a data map having feature and operational information. A system is disclosed that includes an interpretable machine learning model that generates a function in response to an inputted data map, wherein the data map includes feature data and operational data over a region of interest, and wherein the function relates a set of predictive variables to one or more response variables; an integration/interpolation system that generates the data map from a set of disparate data sources; and an analysis system that evaluates the function to predict outcomes at unique points in the region of interest.

Abstract: The embodiments herein describe a forecasting system that uses captured images or a location to generate a weather forecast for that location. As used herein, a hyper-location is any location where images of that location are available to the forecasting system. For example, a hyper-location can be an airport where a security camera can provide historical images of the weather conditions at the airport. In one embodiment, the forecasting system can extract attributes from the images that indicate the historical weather conditions at the hyper-location. The forecasting system can then use those weather conditions to select which one of a plurality of historical scenarios best matches the weather conditions. The selected scenario can then be used to train a machine learning (ML) model that tunes a weather forecast for that location.

Abstract: The embodiments herein describe a forecasting system that uses captured images or a location to generate a weather forecast for that location. As used herein, a hyper-location is any location where images of that location are available to the forecasting system. For example, a hyper-location can be an airport where a security camera can provide historical images of the weather conditions at the airport. In one embodiment, the forecasting system can extract attributes from the images that indicate the historical weather conditions at the hyper-location. The forecasting system can then use those weather conditions to select which one of a plurality of historical scenarios best matches the weather conditions. The selected scenario can then be used to train a machine learning (ML) model that tunes a weather forecast for that location.

Abstract: A computer-implemented method for visualizing uncertainty in flow measurements is provided. A non-limiting exemplary method includes reading, by a processor, a plurality of fluid flow measurements having a magnitude, a direction, and a location. The method plots a plurality of points on a first polar graph, each point representing the difference in fluid flow measurements between two of the plurality of fluid flow measurements at different locations and calculates an uncertainty boundary line based on the plurality of plotted points. The method filters, by the processor, the plurality of plotted points to remove plotted points outside of the uncertainty boundary line leaving only remaining plotted points and defines at least one cluster of remaining plotted points based on similar behavior of a subset of the remaining plotted points. The method may use the remaining plotted points to provide adaptive setup to tune a response in the fluid flow measurements.

Abstract: A computer-implemented method for visualizing uncertainty in flow measurements is provided. A non-limiting exemplary method includes reading, by a processor, a plurality of fluid flow measurements having a magnitude, a direction, and a location. The method plots a plurality of points on a first polar graph, each point representing the difference in fluid flow measurements between two of the plurality of fluid flow measurements at different locations and calculates an uncertainty boundary line based on the plurality of plotted points. The method filters, by the processor, the plurality of plotted points to remove plotted points outside of the uncertainty boundary line leaving only remaining plotted points and defines at least one cluster of remaining plotted points based on similar behavior of a subset of the remaining plotted points. The method may use the remaining plotted points to provide adaptive setup to tune a response in the fluid flow measurements.

Abstract: A method for image capture for colorimetric analysis includes inserting a colorimetric sample into a slot formed in a base of a device. The device includes a lid surrounding and covering the base. The lid includes a central opening, and a plurality of lights are within the lid and surrounding the opening. The method also includes selecting one or more wavelengths and intensities for illuminating the sample within the device, illuminating the sample within the device with the selected one or more wavelengths and intensities by selectively activating one or more of the plurality of lights, adjusting a height of a camera over the sample to focus the sample in the camera, and capturing at least one image of the sample using the camera.

Abstract: A method of evaluating image quality includes capturing a first digital image of a target object using a first digital camera of a smart device. A second digital image of a user of the smart device is captured using a second digital camera of the smart device. The second digital image includes an image of the user's facial expression. A quality index is generated for the first digital image by analyzing one or more features of the second digital image. Analyzing the second digital image includes determining the user's sentiment. The quality index is then associated with the first digital image.

Abstract: A system, product, and method of automatic reading any flow metering device, including generating a first image of the device associated and secured with a mobile device with a camera, recognizing a type of device according to a database of metering's, generating a mask for any following image based on the mobile device and the first image, based on any following image, analyzing it to extract information based on the image recognition related to the type of device, and validating the extracted information with location, timestamp, and the mask.

Abstract: A computer-implemented method, an automatic pest control monitoring system and computer program product automatically monitor for pests on crops. An autonomous vehicle equipped with a normal camera and at least one alternate camera, simultaneously captures a normal image and a true alternate image containing a same portion of the crops. A composite image is generated using the difference of the captured images. If at least one pest is determined to present by applying an object recognition algorithm to the composite image, reactive measures are automatically deployed. The alternate camera may be an infrared camera or an ultraviolet camera. The composite image is generated by correlating the normal image with the true alternate image, determining the differences between the normal image and the true alternate image, and generating the composite image based on the differences between the normal image and the true alternate image.

Abstract: A system, method and program product for annotating visualizations with uncertainty information. A system is provided that includes a visualization importer that imports a generated visualization; an uncertainty processor that locates a region of uncertainty in the generated visualization; and a graphics annotator that generates an annotated visualization having uncertainty artifacts that visually identify the region of uncertainty.

Abstract: A method for image capture for colorimetric analysis includes inserting a colorimetric sample into a slot formed in a base of a device. The device includes a lid surrounding and covering the base. The lid includes a central opening, and a plurality of lights are within the lid and surrounding the opening. The method also includes selecting one or more wavelengths and intensities for illuminating the sample within the device, illuminating the sample within the device with the selected one or more wavelengths and intensities by selectively activating one or more of the plurality of lights, adjusting a height of a camera over the sample to focus the sample in the camera, and capturing at least one image of the sample using the camera.

Abstract: A system, product, and method of automatic reading any flow metering device, including generating a first image of the device associated and secured with a mobile device with a camera, recognizing a type of device according to a database of metering's, generating a mask for any following image based on the mobile device and the first image, based on any following image, analyzing it to extract information based on the image recognition related to the type of device, and validating the extracted information with location, timestamp, and the mask.

Abstract: A method of evaluating photographer satisfaction includes capturing a first digital image of a target object using a first digital camera of a smart device. A second digital image of a user of the smart device is captured using a second digital camera of the smart device. The second digital image includes an image of the user's facial expression. A satisfaction index is generated for the first digital image by analyzing the user's sentiment related to the first digital image. Analyzing the user's sentiment related to the first digital image includes evaluating the user's facial expression. The satisfaction index is then associated with the first digital image.

Abstract: A computer-implemented method, an automatic pest control monitoring system and computer program product automatically monitor for pests on crops. An autonomous vehicle equipped with a normal camera and at least one alternate camera, simultaneously captures a normal image and a true alternate image containing a same portion of the crops. A composite image is generated using the difference of the captured images. If at least one pest is determined to present by applying an object recognition algorithm to the composite image, reactive measures are automatically deployed. The alternate camera may be an infrared camera or an ultraviolet camera. The composite image is generated by correlating the normal image with the true alternate image, determining the differences between the normal image and the true alternate image, and generating the composite image based on the differences between the normal image and the true alternate image.

Abstract: A system, method and program product for annotating visualizations with uncertainty information. A system is provided that includes a visualization importer that imports a generated visualization; an uncertainty processor that locates a region of uncertainty in the generated visualization; and a graphics annotator that generates an annotated visualization having uncertainty artifacts that visually identify the region of uncertainty.