A method and apparatus of a device for generating an above ground biomass density prediction model is described. In an exemplary embodiment, the device receives a first set of satellite and optionally environmental data for the target landmass. In addition, the device trains an above ground biomass density model using at least the satellite data and Light Detection and Ranging (LIDAR) data. Furthermore, the device applies the above ground biomass density model using a second set of satellite and environmental biomass to generate the ground biomass density map.

A system and method for generating soil moisture data from satellite images of a geographical area using a deep learning model 108 is provided. The system includes one or more satellites 102A-C, a soil moisture data generator server 106. The method includes, (i) receiving, by a soil moisture data generator server, satellite images of the geographical area, (ii) pre-processing first set of satellite images, second set of satellite images, and third set of satellite images, (iii) interpolating, using spline interpolation, pre-processed first set of images, pre-processed second set of images, and pre-processed third set of images to generate high-resolution set of images, (iv) generating hydrological parameters from the high-resolution set of images, (v) training, a deep learning model, by providing historical hydrological parameters and historical soil moisture data associated with historical satellite images as training data to generate trained deep learning model, (v) generating soil moisture data on daily basis.

The present invention provides a processing system (20) including an image capturing unit (21) that captures an image of a person passing through a first area and generates a person image indicating an appearance of the person, an electromagnetic wave transmission/reception unit (22) that emits an electromagnetic wave having a wavelength of 30 micrometers or more and 1 meter or less toward the person passing through the first area, and receives a reflected wave, a determination unit (23) that determines whether the person possesses a pre-designated object, based on a signal of the received reflected wave, and a registration unit (24) that registers, in a first list, the person image of the person determined to possess the pre-designated object.

The invention relates to a method for processing time series of noisy images of a same area, the method comprising: generating a set of time series of images from an input image time series by combining by first linear combinations each pixel of each image of the input image time series with selected neighboring pixels in the image and in an adjacent image of the input image time series; applying filtering operations in cascade to the set, each filtering operation combining each pixel of each image of each time series of the set by second linear combinations with selected neighboring pixels in the image and in an adjacent image in each time series of the set; performing an image combination operation to reduce each time series of the set to a single image; introducing a model image of the area as a filtered image in the set; and combining each image in the set into an output image, by third linear combinations.

Systems and methods for road typology scene annotation are provided. A method for road typology scene annotation includes receiving an image having a road scene. The image is received from an imaging device. The method populates, using a machine learning model, a set of attribute settings with values representing the road scene. An annotation interface is implemented and configured to adjust values of the attribute settings to correspond with the road scene. Based on the values of the attribute settings, a simulated overhead view of the respective road scene is generated.

Examples disclosed herein are related to using a machine learning model to generate image data. One example provides a system, comprising one or more processors, and storage comprising instructions executable by the one or more processors to obtain image data comprising an image with unoccluded features, apply a mask to the unoccluded features in the image to form partial observation training data comprising a masked region that obscures at least a portion of the unoccluded features, and train a machine learning model comprising a generator and a discriminator at least in part by generating image data for the masked region and comparing the image data generated for the masked region to the image with unoccluded features.

A dam slope deformation monitoring system and method are provided. The monitoring system monitors an entire dam in a reservoir area by using an unmanned aerial vehicle (UAV) photogrammetry system, and determines an encrypted monitoring area (steep slope) with the relatively large deformation and a relatively large digital elevation difference; determines, in the intensive monitoring area, a first level key monitoring area with the larger deformation by using a ground-based radar interferometry measurement system; determines, in the first level key monitoring area, a second level key monitoring area with the larger deformation by using a ground-based three-dimensional lidar measurement system; determines, in the second level key monitoring area, a key monitoring particle with a high deformation speed by using a global navigation satellite system (GNSS). The core chip stack is used to monitor and warn the collapse process in the area where the key monitoring particles are located.

This document describes techniques, apparatuses, and systems for sensor fusion for object-avoidance detection, including stationary-object height estimation. A sensor fusion system may include a two-stage pipeline. In the first stage, time-series radar data passes through a detection model to produce radar range detections. In the second stage, based on the radar range detections and camera detections, an estimation model detects an over-drivable condition associated with stationary objects in a travel path of a vehicle. By projecting radar range detections onto pixels of an image, a histogram tracker can be used to discern pixel-based dimensions of stationary objects and track them across frames. With depth information, a highly accurate pixel-based width and height estimation can be made, which after applying over-drivability thresholds to these estimations, a vehicle can quickly and safely make over-drivability decisions about objects in a road.

A method (100) for referencing an optical image (19) including: obtaining (110, 120) a stereoscopic image pair (19, 23) of the optical image (19) and a SAR image (35), the surface areas covered by the images (19, 23, 35) on the ground having an overlapping area (39); selecting (130) an area of interest (42) in the overlapping area (39); from the area of interest (42): obtaining (140) a 3D model (40); calculating (150) a simulated radar image (44); estimating (160) an offset (di, dj) between the simulated image (44) and the radar image (35); selecting (170) a reference point (46); projecting (180) and shifting (di, dj) the reference point (46) in the radar image (35) to correct the radar connection point (46′″); determining (175) a pair of connection points (46′, 46″) in the image pair; and referencing the optical image (19) based on the connection points (46′, 46″, 46′″).

Methods and systems are provided for performing radar-to-lidar calibration. In some aspects, a process can include steps for receiving, at an autonomous vehicle system, radar data from a radar of an object, receiving, at the autonomous vehicle system, lidar data from a lidar of the object, generating, by the autonomous vehicle system, a plurality of cost functions based on the radar data and the lidar data of the object, and adjusting, by the autonomous vehicle system, at least one setting based on the plurality of cost functions of the radar data and the lidar data of the object.