Abstract: A method for correcting top-of-atmosphere reflectance data in high altitude imagery to a ground surface reflectance data. Transmission of light through Earth's atmosphere and its suspended load of aerosol particles degrades light within the visible through near infrared portion of the spectrum. This can severely affect the quality of the data recorded by orbiting Earth observation satellites. The method first measures the degree of atmospheric effects upon reflectance, then reverses these effects to deliver surface reflectance data and imagery cleaned of haze and thin clouds.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying one or more regions of a brain of a biological organism that are predicted to be functionally-specialized for performing a task. In one aspect, a method comprises: obtaining data defining a synaptic connectivity graph representing synaptic connectivity between neurons in the brain of the biological organism; identifying a plurality of sub-graphs of the synaptic connectivity graph; determining, for each sub-graph of the plurality of sub-graphs, a performance measure characterizing a performance of a neural network having a neural network architecture that is specified by the sub-graph in accomplishing the task; and determining, based on the performance measures, that one or more sub-graphs of the plurality of sub-graphs correspond to regions of the brain of the biological organism that are predicted to be functionally-specialized for performing the task.

Abstract: The present disclosure provides a method for recognizing a small target based on a deep learning network. The method comprises: determining, based on a collected image, spot defect information through a recognition model including a first feature determination layer, a second feature determination layer, and a spot defect determination layer, determining, based on the collected image, a first feature map, determining, based on the first feature map, a second feature map by fusing with positional encoding, determining, based on the second feature map, a third feature map through the second feature determination layer, and obtaining, based on the third feature map, positional information of the spot defect through a first determination layer, and determining, based on the third feature map, classification information of the spot defect through a second determination layer.

Abstract: A tree crown extraction method based on UAV multi-source remote sensing includes: obtaining a visible light image and LIDAR point clouds, taking a digital orthophoto map (DOM) and the LIDAR point clouds as data sources, using a method of watershed segmentation and object-oriented multi-scale segmentation to extract single tree crown information under different canopy densities. The object-oriented multi-scale segmentation method is used to extract crown and non-crown areas, and a tree crown distribution range is extracted with the crown area as a mask; a preliminary segmentation result of single tree crown is obtained by the watershed segmentation method based on a canopy height model; a brightness value of DOM is taken as a feature, the crown area of the DOM is performed secondary segmentation based on a crown boundary to obtain an optimized single tree crown boundary information, which greatly increases the accuracy of remote sensing tree crown extraction.

Abstract: Implementations are described herein for predicting soil organic carbon (“SOC”) content for agricultural fields detected in digital imagery. In various implementations, one or more digital images depicting portion(s) of one or more agricultural fields may be processed. The one or more digital images may have been acquired by a vision sensor carried through the field(s) by a ground-based vehicle. Based on the processing, one or more agricultural inferences indicating agricultural practices or conditions predicted to affect SOC content may be determined. Based on the agricultural inferences, one or more predicted SOC measurements for the field(s) may be determined.

Abstract: Embodiments of the present disclosure provide a method and apparatus for identifying an item. The method includes: acquiring an item image of a to-be-identified item; setting initial position coordinates of the to-be-identified item on the item image; and executing following identifying: inputting the item image and the initial position coordinates into a pre-trained attention module to output an item feature of the to-be-identified item; inputting the item feature into a pre-trained long short-term memory network to output a predicted category and predicted position coordinates of the to-be-identified item; determining whether a preset condition is satisfied; and determining, in response to the preset condition being satisfied, a predicted category of the to-be-identified item outputted by the long short-term memory network a last time for use as a final category of the to-be-identified item.

Abstract: A citrus identification method comprises: performing first-time image acquisition processing on a target citrus tree to obtain a first image; inputting the first image into a first citrus fruit identification model to be processed to obtain a first identification result sequence; performing area interception processing on the first image to obtain a citrus fruit area; obtaining roundness integrity numerical values; selecting an appointed roundness integrity numerical value, and acquiring a defect position of an appointed citrus fruit in the first image; determining a first spatial range and a second spatial range; performing both first spray injection treatment and second spray injection treatment; performing second-time image acquisition processing to obtain a second image; inputting the second image into a second citrus fruit identification model to be processed to obtain a second identification result; and generating a citrus fruit identification result.

Abstract: An object association method, apparatus and system, and a storage medium are provided. The method includes: obtaining a first image and a second image; and determining an association relationship between objects in the first image and objects in the second image based on surrounding information of the objects in the first image and surrounding information of the objects in the second image, where the surrounding information of one object is determined according to pixels within a set range around a bounding box of the object in the image where the object is located.

Abstract: In one embodiment, a method includes a computer system accessing a curvilinear image captured using a camera lens, generating multiple rectilinear images from the curvilinear image based at least in part on one or more calibration parameters associated with the camera lens, identifying semantic information in one or more of the rectilinear images by processing each of the multiple rectilinear images using a machine-learning model configured to identify semantic information in rectilinear images, and identifying semantic information in the curvilinear image based on the identified semantic information in the one or more rectilinear images.

Abstract: A farming machine moves through a field and includes an image sensor that captures an image of a plant in the field. A control system accesses the captured image and applies the image to a machine learned plant identification model. The plant identification model identifies pixels representing the plant and categorizes the plant into a plant group (e.g., plant species). The identified pixels are labeled as the plant group and a location of the pixels is determined. The control system actuates a treatment mechanism based on the identified plant group and location. Additionally, the images from the image sensor and the plant identification model may be used to generate a plant identification map. The plant identification map is a map of the field that indicates the locations of the plant groups identified by the plant identification model.

Abstract: A computing system includes a processor and a non-transitory, computer-readable media including instructions that, when executed by the one or more processors, cause the computing system to access an initial machine data set; label the machine data set; process the labeled machine data set; and modify one or more parameters of the machine-learned model. A method includes accessing an initial machine data set; labeling the machine data set; processing the labeled machine data set; and modifying one or more parameters of the machine-learned model. A computing system for predicting a variety profile index includes a processor; and a non-transitory, computer-readable media including a trained machine-learned model; and instructions that, when executed by the one or more processors, cause the computing system to process a second machine data set to generate one or more predicted variety profile index values; and provide the one or more predicted variety profile index values.

Abstract: Disclosed herein is an apparatus for the real-time monitoring of construction objects. The apparatus for the real-time monitoring of construction objects includes: a communication unit configured to receive image data acquired by photographing a construction site, and to transmit safety information to an external device; and a monitoring unit provided with a prediction model pre-trained using binary image sequences of construction objects at the construction site as training data, and configured to detect a plurality of construction objects from image frames included in image data received via the communication unit and convert the detected construction objects into binary images, to generate future frames by inputting the resulting binary images to the prediction model, and to derive proximity between the construction objects by comparing the generated future frames with the resulting binary images and generate the safety information.

Abstract: An apparatus for weed control includes a processing unit that receives at least one image of an environment. The processing unit analyses the at least one image to determine at least one vegetation control technology from a plurality of vegetation control technologies to be used for weed control for at least a first part of the environment. An output unit outputs information that is useable to activate the at least one vegetation control technology.

Abstract: A system determines spatial locations of pixels of an image. The system includes a processor configured to: receive location data from devices located within a hotspot; generate a density map for the hotspot including density pixels associated with spatial locations defined by the location data, each density pixel having a value indicating an amount of location data received from an associated spatial location; match the density pixels of the density map to at least a portion of the pixels of the image; and determine spatial locations of the at least a portion of the pixels of the image based on the spatial locations of the matching density pixels of the density map. In some embodiments, the image and density map are converted to edge maps, and a convolution is applied to the edge maps to match the density map to the pixels of the image.

Abstract: A method for determining residue length within a field includes receiving, with a computing system, a captured image depicting an imaged portion of the field from one or more imaging devices. Furthermore, the method includes determining, with the computing system, an image gradient orientation at each of a plurality of pixels within the captured image. Additionally, the method includes identifying, with the computing system, a residue piece present within the image portion of the field based at least in part on the determined image gradient orientations. Moreover, the method includes determining, with the computing system, a length of the identified residue piece.

Abstract: Systems and methods for iteratively computing an image registration or an image segmentation are driven by an optimization function that includes a similarity measure component whose effect on the iterative computations is relatively mitigated based on a monitoring of volume changes of volume elements at image locations during the iterations. A system and a related method quantify a registration error by applying a series of edge detectors to input images and combining related filter responses into a combined response. The series of filters are parameterized with a filter parameter. An extremal value of the combined response is then found and a filter parameter associated with said extremal value is then returned as output. This filter parameter relates to a registration error at a given image location.

Abstract: A method of compiling performance metrics for racing competitors by analyzing videos of the competitors' prior races, includes determining: (a) the GPS coordinates of race track landmarks that appear in the videos, (b) within each of the videos' images, the position and orientation of a part for each of the competitors, (c) for each of the videos' images, the camera's extrinsic and intrinsic parameters, (d) the path of each of the competitors from the race's starting line to its finish line, (e) for each of the videos' images, the GPS coordinates of the part for each of the competitors, and (f) using the camera's frame rate, the performance metrics for each of the racing competitors.

Abstract: The disclosure directed to methods for measuring an analyte alone or in combination with total protein in biological samples. More particularly, the disclosure relates to methods for measuring an analyte and/or total protein using one or more colorimetric reagents alone or in combination with protein precipitation reagents.

Abstract: In accordance with various aspects of the present disclosure, methods and systems for fire detection are disclosed. In some embodiments, a method for fire detection includes: acquiring data related to a monitored area, wherein the data comprises image data related to the monitored area; determining, based on the acquired data, whether a first mode or a second mode is to be executed for fire detection, wherein the first mode comprises a first smoke detection, and the second mode comprises a first flame detection; and executing, by a hardware processor, at least one of the first mode or the second mode based on a result of the determination.

Abstract: An image recognition device includes: an image processing device that acquires a feature amount from an image; and an identification device that determines whether a prescribed identification object is present in the image, and identifies the identification object. The identification device includes a BNN that has learned the identification object in advance, and performs identification processing by performing a binary calculation with the BNN on the feature amount acquired by the image processing device. Then, the identification device selects a portion effective for identification from among high-dimensional feature amounts output by the image processing device to reduce the dimensions used in identification processing, and copies low-dimensional feature amounts output by the image processing device to increase dimensions.