Abstract: In various examples, image processing techniques are presented for reducing artifacts in images for autonomous or semi-autonomous systems and applications. Systems and methods are disclosed for deferring at least a portion of a strength associated with a color correction matrix (CCM) of an image processing pipeline to one or more other stages associated with the image processing pipeline. For instance, the CCM is used to determine a first CCM and a second, deferred CCM. The first CCM may be associated with an earlier stage of the image processing pipeline while the deferred CCM is associated with a later stage of the image processing pipeline. In some examples, the deferred CCM is combined with another matrix, such as a color space conversion matrix. By deferring part of the CCM, the image processing pipeline may reduce a number of artifacts and/or eliminate the artifacts that occur when processing image data.

Abstract: In some embodiments, a first image may be captured from a first field of view using a first exposure time. A second image may be captured from a second field of view using a second exposure time that is different than the first exposure time. An overlapping field of view may be defined by an overlapping portion of the first field of view and the second field of view. Histograms may be created for the first image and the second image, and possibly more images that include different exposure times and represent the overlapping field of view. The histograms may be analyzed to determine a presence or an absence of a light source in the overlapping field of view.

Abstract: A system for detecting transparent elements in a vehicle environment are described. In some examples, this may include accessing an image of a scene captured by an image capture device attached to a vehicle. A reflected image present in the image may be detected. The reflected image may include a portion of the vehicle. It may be determined that the scene includes a transparent element based at least in part on detecting the reflected image present in the image.

Abstract: Techniques for verifying a location and identification of a landing marker to aid an unmanned aerial vehicle (UAV) to deliver a payload to a location may be provided. For example, upon receiving an indication that a UAV has arrived to a delivery location, a server computer may process one or more images of an area that are provided by the UAV and/or a user interacting with a user device. A landing marker may be identified in the image and a representation of the landing marker along with instructions to guide the UAV to deliver the payload to the landing marker may be transmitted to the UAV and implemented by the UAV.

Abstract: Camera calibration may be performed in a mobile environment. One or more cameras can be mounted on a mobile vehicle, such as an unmanned aerial vehicle (UAV) or an automobile. Because of the mobility of the vehicle the one or more cameras may be subjected to inaccuracy in imagery caused by various factors, such as environmental factors (e.g., airflow, wind, etc.), impact by other objects (e.g., debris, vehicles, etc.), vehicle vibrations, and the like. To reduce the inaccuracy in imagery, the mobile vehicle can include a mobile camera calibration system configured to calibrate the one or more cameras while the mobile vehicle is traveling along a path. The mobile camera calibration system can cause the one or more cameras to capture an image of an imaging target while moving, and calibrate the one or more cameras based on a comparison between the image and imaging target data.

Abstract: Systems and corresponding methods for monitoring of a state of charge or a state of health of a battery are described. For example, the battery monitoring system may include strain gauges to measure dimensions or changes in dimension of cells of the battery, and may determine states of charge of the cells based on dimensional changes over time, e.g., based on data from strain gauges. The dimensional changes may also be corrected for dimensional changes due to thermal strain. In addition, states of health of individual cells of the battery may be determined by comparing the dimensional changes of each of the cells over time. Further, the data from strain gauges may be received at various frequencies, e.g., based on an expected operational duration of the battery and/or based on a current state of charge or state of health.

Abstract: Systems and methods for detecting contamination on a sensor cover. In an example, an image sensor of a vehicle may be used to capture an image of an environment of the vehicle. A sensor cover may be positioned outwardly from the image sensor and may form a covered volume around at least a sensing surface of the image sensor. A light source may be positioned in the covered volume. A management module may be coupled with the image sensor. The management module may be configured to receive test image data of the environment generated by the image sensor while the light source is on, compare the test image data to baseline image data generated while the light source is off, and detect whether contamination exists on a surface of the sensor cover based on the comparison.

Abstract: Described herein are systems and methods of determining a distance between an object depicted in an image and an imaging device that captured that image. In particular, the disclosure discusses that an image may be separated into multiple derivative images, each of which is associated with a different wavelength of light. In some embodiments, an image may be separated into images associated with wavelengths of primary colors (e.g., red, green, and blue). Once separate images have been created, a sharpness value may be determined for each image. A distance between the object and the imaging device may then be calculated based on sharpness values associated with each of the separate images.

Abstract: This disclosure is directed to monitoring exposure levels for individual cameras of a stereo camera system to reduce unwanted lens flaring which may disrupt stereo vision capabilities of an unmanned aerial vehicle (UAV). The UAV may identify lens flaring by monitoring histogram data received from the cameras and then mitigate the lens flaring by performing aerial maneuvers with respect to the light source and/or moving camera componentry such as deploying a lens hood. The UAV may also identify lens flaring patterns that on a camera's sensor and determine a location of a peripheral light source based on these patterns. The UAV may then deploy a visor between the location of the peripheral light source and the cameras.

Abstract: Images captured by a camera system of an unmanned aerial vehicle (UAV) can be used to determine a weather condition in an environment of the UAV. The camera system of the UAV can capture one or more images, and a characteristic of at least one image of the one or more images can be determined from image data associated with the at least one image. A weather condition of the environment of the UAV can be determined based at least in part on the characteristic of the at least one image.

Abstract: A location marker that may be used to provide information to a vehicle, such as an unmanned aerial vehicle (UAV). The location marker may include a plurality of retroreflectors that may form a pattern readable by a vehicle or other device. The pattern may be read to extract an identifier of a location, such as an address or an identifier of a person. A global positioning system (GPS) device may transmit a general location of the location marker to the vehicle, while the location marker may provide a unique visual location identifier to a device within visual range of the location marker. In some embodiments, the location marker may also include lights that may be individually sequenced on and off at different times to create a time domain signal sequence that is readable by the vehicle.

Abstract: Systems and corresponding methods for monitoring of a state of charge or a state of health of a battery are described. For example, the battery monitoring system may include strain gauges to measure dimensions or changes in dimension of cells of the battery, and may determine states of charge of the cells based on dimensional changes over time, e.g., based on data from strain gauges. The dimensional changes may also be corrected for dimensional changes due to thermal strain. In addition, states of health of individual cells of the battery may be determined by comparing the dimensional changes of each of the cells over time. Further, the data from strain gauges may be received at various frequencies, e.g., based on an expected operational duration of the battery and/or based on a current state of charge or state of health.

Abstract: Systems and corresponding methods for optical monitoring of a state of charge or a state of health of a battery are described. For example, the battery monitoring system may include an imaging device that captures images of cells of the battery and determines states of charge of the cells based on observed dimensional changes over time, e.g., relative to reference markers. The observed dimensional changes may also be corrected for dimensional changes due to thermal strain. In addition, states of health of individual cells of the battery may be determined by comparing the dimensional changes of each of the cells over time. Further, the images may be captured at various frequencies, e.g., based on an expected operational duration of the battery and/or based on a current state of charge or state of health.

Abstract: Techniques for providing an object awareness guidance to clear a landing space may be provided. For example, during delivery an unmanned aerial vehicle (UAV) may capture an image of a potential landing zone and identify one or more objects in the image that may impede or obstruct delivery of the item in the potential landing zone. The UAV may be configured to generate and provide instructions to a user device to move or remove the identified one or more objects from the potential landing zone thereby creating a safe and unobstructed landing zone to deliver the item.

Abstract: A method, device, and system for detecting transparent elements in a vehicle environment are described. In some examples, this may include accessing an image of a scene captured by an image capture device attached to a vehicle. A reflected image present in the image may be detected. The reflected image may include a portion of the vehicle. It may be determined that the scene includes a transparent element based at least in part on detecting the reflected image present in the image.

Abstract: Examples of the disclosure enable a mobile device to generate high quality images. In some examples, the mobile device includes a lens assembly that includes a first lens configured to provide positive optical power, and a telephoto stage including a second lens configured to provide negative optical power, and a third lens configured to perform one or more of increasing a sharpness and decreasing a distortion of an image associated with light transmitted through the lens assembly. The lens assembly has a track length that is less than or equal to approximately 6.0 mm and a focal length that is greater than or equal to approximately 7.3 mm. Aspects enable a lens assembly to be used in a mobile device environment, such that a mobile device including and/or associated with the lens assembly is configured to take high quality images.

Abstract: Techniques for verifying a location and identification of a landing marker to aid an unmanned aerial vehicle (UAV) to deliver a payload to a location may be provided. For example, upon receiving an indication that a UAV has arrived to a delivery location, a server computer may process one or more images of an area that are provided by the UAV and/or a user interacting with a user device. A landing marker may be identified in the image and a representation of the landing marker along with instructions to guide the UAV to deliver the payload to the landing marker may be transmitted to the UAV and implemented by the UAV.

Abstract: A method, device, and system for detecting transparent elements in a vehicle environment are described. In some examples, this may include accessing a first image depicting a scene. The first image may include a first polarization characteristic. A second image depicting the scene may also be accessed. The second image may include a second polarization characteristic. It may be determined that the scene includes an element having transparent properties based at least in part on the first image and the second image.

Abstract: Images captured by a camera system can be processed to detect precipitation in one or more of the images, and to generate a reconstructed image(s) without the precipitation, or with a reduced amount of the precipitation. Detection of precipitation can be based on a difference between a first feature in a first image and a second feature in a second image that corresponds to the first feature, where the first and second images were captured by different cameras at different times. A determination as to whether precipitation is present in the first image and the second image can be based at least in part on a disparity between the first feature and the second feature.

Abstract: Systems and methods are provided herein for detecting a marker (e.g., a marker that identifies a delivery location) utilizing an image captured by a camera of an unmanned aerial vehicle. A method may include obtaining marker information associated with a marker, the marker comprising a repetitive visual pattern, the marker being associated with delivery of an item by an unmanned aerial vehicle. Optical pattern information may be obtained that indicates a moiré pattern associated with the marker and the one or more cameras of the unmanned aerial vehicle. Image capture information that is associated with an image comprising the marker may be received. The marker may be detected in the image based at least in part on the image capture information and the moiré pattern associated with the marker.