Abstract: A computing system including one or more processing devices configured to receive a first labeled image set including first images that each include one or more first regions of interest. The one or more processing devices are further configured to receive an unlabeled image set including a plurality of second images, and to identify second regions of interest included in the second images. The one or more processing devices are further configured to compute feature similarity values between the second and first regions of interest, identify a subset of the second regions of interest that have feature similarity values above a predetermined similarity threshold, and apply respective second labels to those second regions of interest. The one or more processing devices are further configured to construct a second labeled image set and train an image classification machine learning model with a training data set that includes the labeled image sets.

Abstract: A method of creating custom datasets is provided. The method comprises capturing, with a camera, a sequence of images of a surface in a specified crawl direction over a number of sequential crawl steps that represent changes of camera position. For each crawl step, the method matches a first image and a second image captured at sequential camera positions. The method models each crawl step as a pairwise quadratic transformation between the first image and second image according to a number of respective anchor points in each image selected as best candidates to fit a regression model.

Abstract: A computing system includes an input sensor, a processor, and a memory storing executable instructions that, in response to execution by the processor, cause the processor to collect input data related to at least a portion of an object from the input sensor, generate a UV coordinate map based on the input data, use a predetermined process ink density lookup table to produce an ink density bitmap corresponding to the UV coordinate map, the ink density bitmap including process ink densities for each bit in the ink density bitmap, produce and output a dithering of bits of the ink density bitmap to thereby generate dithered image pixel data, and compile and output a control plan, based on the dithered image pixel data, to control a printing array to print an image on a contoured surface of the object.

Abstract: A computing system including one or more processing devices configured to receive a first labeled image set including first images that each include one or more first regions of interest. The one or more processing devices are further configured to receive an unlabeled image set including a plurality of second images, and to identify second regions of interest included in the second images. The one or more processing devices are further configured to compute feature similarity values between the second and first regions of interest, identify a subset of the second regions of interest that have feature similarity values above a predetermined similarity threshold, and apply respective second labels to those second regions of interest. The one or more processing devices are further configured to construct a second labeled image set and train an image classification machine learning model with a training data set that includes the labeled image sets.

Abstract: A method of creating custom datasets is provided. The method comprises capturing, with a camera, a sequence of images of a surface in a specified crawl direction over a number of sequential crawl steps that represent changes of camera position. For each crawl step, the method matches a first image and a second image captured at sequential camera positions. The method models each crawl step as a pairwise quadratic transformation between the first image and second image according to a number of respective anchor points in each image selected as best candidates to fit a regression model.

Abstract: A printing system is provided comprising a print head including a plurality of ink nozzles; an actuator configured to move the print head relative to a substrate; a print head position sensing system configured to detect an actual position of the print head relative to the substrate; and a controller. The controller is configured to: receive the actual position of the print head detected by the print head position sensing system corresponding to a target print head position; determine a positional offset between the actual position and the target print head position; generate a nozzle firing pattern based on the positional offset; and control the print head to print the nozzle firing pattern at the target print head position using the plurality of ink nozzles.

Abstract: A computing system includes an input sensor, a processor, and a memory storing executable instructions that, in response to execution by the processor, cause the processor to collect input data related to at least a portion of an object from the input sensor, generate a UV coordinate map based on the input data, use a predetermined process ink density lookup table to produce an ink density bitmap corresponding to the UV coordinate map, the ink density bitmap including process ink densities for each bit in the ink density bitmap, produce and output a dithering of bits of the ink density bitmap to thereby generate dithered image pixel data, and compile and output a control plan, based on the dithered image pixel data, to control a printing array to print an image on a contoured surface of the object.

Abstract: Apparatuses and methods train a model and then use the trained model to determine a global three dimensional (3D) position and orientation of a fiduciary marker. In the context of an apparatus for training a model, a wider field-of-view sensor is configured to acquire a static image of a space in which the fiducial marker is disposed and a narrower field-of-view sensor is configured to acquire a plurality of images of at least a portion of the fiducial marker. The apparatus also includes a pan-tilt unit configured to controllably alter pan and tilt angles of the narrower field-of-view sensor during image acquisition. The apparatus further includes a control system configured to determine a transformation of position and orientation information determined from the images acquired by the narrower field-of-view sensor to a coordinate system for the space for which the static image is acquired by the wider field-of-view sensor.

Abstract: A printing system is provided comprising a print head including a plurality of ink nozzles; an actuator configured to move the print head relative to a substrate; a print head position sensing system configured to detect an actual position of the print head relative to the substrate; and a controller. The controller is configured to: receive the actual position of the print head detected by the print head position sensing system corresponding to a target print head position; determine a positional offset between the actual position and the target print head position; generate a nozzle firing pattern based on the positional offset; and control the print head to print the nozzle firing pattern at the target print head position using the plurality of ink nozzles.

Abstract: Apparatuses and methods train a model and then use the trained model to determine a global three dimensional (3D) position and orientation of a fiduciary marker. In the context of an apparatus for training a model, a wider field-of-view sensor is configured to acquire a static image of a space in which the fiducial marker is disposed and a narrower field-of-view sensor is configured to acquire a plurality of images of at least a portion of the fiducial marker. The apparatus also includes a pan-tilt unit configured to controllably alter pan and tilt angles of the narrower field-of-view sensor during image acquisition. The apparatus further includes a control system configured to determine a transformation of position and orientation information determined from the images acquired by the narrower field-of-view sensor to a coordinate system for the space for which the static image is acquired by the wider field-of-view sensor.

Abstract: Methods and systems for facilitating detecting features in sensor data are described. One example method implemented by a computing device includes receiving a first set of sensor data about a geographical region, and generating a second set of sensor data. The first set of sensor data includes data in a plurality of bands. The second set of sensor data is generated by receiving a first input designating a first sub-region of the geographical region, and determining a single band representation of at least a portion of the first set of sensor data associated with the first sub-region.