Abstract: Examples of the disclosure describe systems and methods for identifying, quantifying, and/or characterizing plant stomata. In an example method, a first set of two or more images of a plant leaf representing two or more focal distances is captured via an optical sensor. A reference focal distance is determined based on the first set of images. A second set of two or more images of the plant leaf is captured via the optical sensor, including at least one image captured at a focal distance less than the reference focal distance, and at least one image captured at a focal distance greater than the reference focal distance. A composite image is generated based on the second set of images. The composite image is provided to a trainable feature detector in order to determine a number, density, and/or distribution of stomata in the composite image.

Abstract: Existing techniques in precision farming comprise supervised event detection and need labeled training data which is tedious considering the large number of crops, differences therein and even larger number of diseases and pests. The present disclosure provides an unsupervised method and uses images of any size captured in an uncontrolled environment. The methods and systems disclosed find application in automatically localizing and classifying events, including health state and growth stage and also estimating an extent of manifestation of the event. Information of spatial continuity in pixels and boundaries in a given image is used to update the feature representation and label assignment to every pixel using a fully convolutional network. Back propagation of the pixel labels modified according to the output of a graph based method helps the neural network to converge and provide a time efficient solution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for identification and re-identification of fish. In some implementations, first media representative of aquatic cargo is received. Second media based on the first media is generated, wherein a resolution of the second media is higher than a resolution of the first media. A cropped representation of the second media is generated. The cropped representation is provided to the machine learning model. In response to providing the cropped representation to the machine learning model, an embedding representing the cropped representation is generated using the machine learning model. The embedding is mapped to a high dimensional space. Data identifying the aquatic cargo is provided to a database, wherein the data identifying the aquatic cargo comprises an identifier of the aquatic cargo, the embedding, and a mapped region of the high dimensional space.

Abstract: Various techniques employ digital fingerprints, alone or in conjunction with other data related to a physical object, to characterize the physical object in a way allows an identity of the physical object to be confirmed or authenticated within a threshold of confidence when the physical object is encountered at a subsequently after a digital fingerprint thereof was previously ingested, without capture, retention and/or reference to certain types or categories of characterizing information which may be of a sensitive nature and/or subject to misuse or abuse. Such digital fingerprints may include feature vectors and non-positional characterization values, while advantageously omitting positional information that would otherwise allow recreation of a recognizable image of the physical object or an image from which one or more immutable characteristics or traits (e.g.

Abstract: An inspection device inspects of goods that include plural articles that sometimes overlap each other. An X-ray inspection device irradiates goods containing plural articles having a predetermined shape and inspects the goods on the basis of inspection images obtained from radiation that has passed through the goods or radiation that has reflected off the goods. The X-ray inspection device includes a storage component, a learning component, and an inspection component. The storage component stores, as teaching images, at least the inspection images of the goods that are in a state in which the plural articles overlap each other. The learning component acquires, by machine learning using the teaching images stored in the storage component, features relating to the goods that are in a state in which the plural articles overlap each other. The inspection component inspects the goods using the features that the learning component has acquired.

Abstract: A rapid detection system for water and impurity of machine-harvested seed cotton in purchase link, including: a seed cotton large-impurity-cleaning device, a seed cotton sample case, a driving and transmission device, a cotton pressing device, an image collection and processing device, and a moisture regain rate detection device and a control cabinet. The seed cotton sample case consists of a positioning plate and a high light transmittance glass case, the weighing device consists of a weighing sensor, a pallet, and an electric push rod. The driving and transmission device consists of the rollers, the motor of conveyor belts and conveyor belts, the cotton pressing device consists of an electric push rod, a connecting rod, a connecting plate, and a high light transmittance glass plate. Said rapid detection system could realize the rapid detection to the impurity rate and moisture regain rate of the seed cotton.

Abstract: The method for item recognition can include: optionally calibrating a sampling system, determining visual data using the sampling system, determining a point cloud, determining region masks based on the point cloud, generating a surface reconstruction for each item, generating image segments for each item based on the surface reconstruction, and determining a class identifier for each item using the respective image segments.

Abstract: Embodiments of the disclosed technologies are capable of inputting, to a machine-learned model that has been trained to recognize a horticultural product in digital imagery, digital video data comprising frames that represent a view of the horticultural product in belt-assisted transit from a picking area of a field to a harvester bin; outputting, by the machine-learned model, annotated video data; using the annotated video data, computing quantitative data comprising particular counts of the individual instances of the horticultural product associated with particular timestamp data; using the timestamp data, mapping the quantitative data to geographic location data to produce a digital yield map; causing display of the digital yield map on a field manager computing device.

Abstract: Described herein are systems and methods for capturing images of a field and performing agricultural data analysis of the images. In one embodiment, a computer system for monitoring field operations includes a database for storing agricultural image data including images of at least one stage of crop development that are captured with at least one of an apparatus and a remote sensor moving through a field. The computer includes at least one processing unit that is coupled to the database. The at least one processing unit is configured to execute instructions to analyze the captured images, to determine relevant images that indicate a change in at least one condition of the crop development, and to generate a localized view map layer for viewing the field at the at least one stage of crop development based on at least the relevant captured images.

Abstract: A food-recognition engine can be used with a mobile device to identify, in real-time, foods present in a video stream. To capture the video stream, a user points a camera of the mobile device at foods they are about to consume. The video stream is displayed, in real-time, on a screen of the mobile device. The food-recognition engine uses several neural networks to recognize, in the video stream, food features, text printed on packaging, bar codes, logos, and “Nutrition Facts” panels. The neural-network outputs are combined to identify foods with high probabilities. The foods may be packaged or unpackaged, branded or unbranded, and labeled or unlabeled, and may appear simultaneously within the view of the mobile device. Information about recognized foods is displayed on the screen while the video stream is captured. The user may log identified foods with a gesture and without typing.

Abstract: A calibration tool for a coffee grinder can receive an image from a camera. The image can depict coffee grinds produced by the coffee grinder. The calibration tool can analyze the image using one or more image processing techniques to determine a representative grind size for the coffee grinds. The calibration tool can determine whether the representative grind size of the coffee grinds deviates from a predefined reference size by at least a threshold amount. In response to determining that the representative grind size of the coffee grinds deviates from the predefined reference size by at least the threshold amount, the calibration tool can output an instruction indicating how to calibrate the coffee grinder to more closely adhere to the predefined reference size.

Abstract: Systems and methods for determining a mass of a crop by using at least one X-ray scanner is provided. The method includes obtaining at least two scan images of the crop, where a first of the at least two images is obtained along a first plane relative to the crop and a second of the at least two images is obtained along a second plane relative to the crop, and where the first plane is angularly displaced relative to the second plane, registering the first image and the second image, correcting the registered first and second images, and determining the mass of the crop from the corrected first and second images.

Abstract: The embodiments disclose a method comprising creating 3D models of an orchard with multiple plants in the form of a densified point cloud using oblique aerial RGB imaging and photogrammetry, identifying and segmenting individual plants of the orchard from the 3D models, simulating sunlight radiation in the 3D models, determining a shading effect of branches and neighboring plants on each individual plant at any time of the day, determining canopy light interception of each plant, analyzing canopy geometry of each plant in the 3D models, forecasting potential yield of each plant based on the measured canopy light interception and calculating nitrogen and water requirements of each plant based on the potential yield and other predetermined field, environmental and climate factors and validating the yield forecasting model using the canopy light interception data by measuring the actual yield for each plant.

Abstract: In an illustrative embodiment, a system for identifying products on a production line includes image capturing devices that acquire images of containers moving along a production line at an inspection location. The system also includes a rejection device and a controller that configures the image capturing devices for image acquisition based on properties of the containers, identifies a product associated with each of the containers based on a portion of a product identification code and a portion of additional features detected in the images, and determines whether the identified product matches predetermined properties or characteristics, resulting in a pass result, otherwise a non-pass result occurs. When a non-pass result occurs, the controller outputs a signal to actuate the rejection device that removes the container from the production line.

Abstract: A method for controlling application of agrichemical products, comprises acquiring remotely sensed digital image data; developing a prescription to apply at least one agrichemical product in a variable manner based on at least the digital image data, wherein the prescription describes a plurality of passes of a particular autonomous vehicle over a field to apply the at least one agrichemical product; applying the at least one agrichemical product to a crop in the variable manner by the particular autonomous vehicle according to the prescription.

Abstract: Implementations set forth herein relate to using fiducial markings on one or more localized portions of an agricultural apparatus in order to generate local and regional data that can be correlated for planning and executing agricultural maintenance. An array of fiducial markings can be disposed onto plastic mulch that surrounds individual crops, in order that each fiducial marking of the array can operate as a signature for each individual crop. Crop data, such as health and yield, corresponding to a particular crop can then be stored in association with a corresponding fiducial marking, thereby allowing the certain data for the particular crop to be tracked and analyzed. Furthermore, autonomous agricultural devices can rely on the crop data, over other sources of data, such as GPS satellites, thereby allowing the autonomous agricultural devices to be more reliable.

Abstract: System and method for imaging-based quality assessment of raw produce loads, including an imaging structure comprising a mounting beam, light sources; optical sensors, and additional sensors; wherein the mounting beam is configured to allow vertical, lateral and/or longitudinal and movement of the measurement chamber. The system also includes a processing unit configured to control the vertical, lateral and/or longitudinal movement of the measurement chamber relative to the mounting beam, control opto-mechanical parameters of the optical sensors, control the position, orientation, light projection and/or light intensity of the light sources; trigger signal acquisition, receiving and fusing spectral data from the optical sensors with data from the additional sensors; and feeding the fused data into machine learning algorithm and/or big-data analytics model, to derive internal and/or external quality attributes of the produce.

Abstract: A method of estimating chlorophyll content of a leaf including providing a device wherein the device includes a mobile computing device with a digital camera; and a peripheral removably attached to the mobile computing device, the peripheral including a light source aligned with the digital camera, wherein the peripheral provides for a space between the light source and the digital camera; capturing by the digital camera a first image of the light source turned on without the leaf interposed between the digital camera and the light source; retrieving exposure data of the first image; capturing by the digital camera a second image with the leaf interposed between the digital camera and the turned on light source; retrieving exposure data of the second image; and using a processor, estimating the chlorophyll content of the leaf based on the retrieved exposure data of the first and second images.

Abstract: This disclosure relates generally to estimating age of a leaf using morphological features extracted from segmented leaves. Traditionally, leaf age estimation requires a single leaf to be plucked from the plant and its image to be captured in a controlled environment. The method and system of the present disclosure obviates these needs and enables obtaining one or more full leaves from images captured in an uncontrolled environment. The method comprises segmenting the image to identify veins of the leaves that further enable obtaining the full leaves. The obtained leaves further enable identifying an associated plant species. The method also discloses some morphological features which are fed to a pre-trained multivariable linear regression model to estimate age of every leaf. The estimated leaf age finds application in estimation of multiple plant characteristics like photosynthetic rate, transpiration, nitrogen content and health of the plants.

Abstract: The present disclosure relates to methods and systems for obtaining image information of an organism including a set of optical data; calculating a growth index based on the set of optical data; and calculating an anticipated harvest time based on the growth index, where the image information includes at least one of: (a) visible image data obtained from an image sensor and non-visible image data obtained from the image sensor, and (b) a set of image data from at least two image capture devices, where the at least two image capture devices capture the set of image data from at least two positions.

Abstract: A computer system is provided comprising a classification model management server computer configured, by instructions, to: receive a new image from a user device; apply a first digital model to first regions within the new image for classifying each of the first regions into a particular class; apply a second digital model to second regions within the new image for classifying each of the second regions into a particular class; and transmit classification data related to the class of the first regions and the class of the second regions to the user device. In connection therewith, the second regions each generally correspond to a combination of multiple first regions.

Abstract: A system and method for detecting the origin of wooden planks in a sawmill is provided. The method scans surfaces of processed planks and, with the help of an AI algorithm comprising a deep-learning algorithm, determines the origin of said planks based on analysed parameters of the planks. The parameters used in the analysis are mainly properties of tool marks and the resulting analysis provides tools and equipment used. The deep learning algorithm may be in a self-learning mode or in a supervised training mode.

Abstract: A human body identification method, an electronic device and a storage medium, related to the technical field of artificial intelligence such as computer vision and deep learning, are provided. The method includes: inputting an image to be identified into a human body detection model, to obtain a plurality of preselected detection boxes; identifying a plurality of key points from each of the preselected detection boxes respectively according to a human body key point detection model, and obtaining a key point score of each of the key points; determining a target detection box from each of the preselected detection boxes, according to a number of the key points whose key point scores meet a key point threshold; and inputting the target detection box into a human body key point classification model, to obtain a human body identification result for the image to be identified.

Abstract: A system and processing methods for configuring and utilizing a convolutional neural network (CNN) for plant disease recognition are disclosed. In some embodiments, the system is programmed to collect photos of infected plants or leaves where regions showing symptoms of infecting diseases are marked. Each photo may have multiple marked regions. Depending on how the symptoms are sized or clustered, one marked region may include only one lesion caused by one disease, while another may include multiple, closely-spaced lesions caused by one disease. The system is programmed to determine anchor boxes having distinct aspect ratios from these marked regions for each convolutional layer of a single shot multibox detector (SSD). For certain types of plants, common diseases lead to relatively many aspect ratios, some having relatively extreme values. The system is programmed to then train the SSD using the marked regions and the anchor boxes and apply the SSD to new photos to identify diseased plants.

Abstract: A system for managing the dosing of inputs to be applied to a surface includes an electronic device housed on board an agricultural machine and a remote management server. The electronic device includes a weather sensor for measuring a weather condition, a central unit for collecting a weather datum containing information relating to the measured weather condition, and first communication means for transmitting the weather datum to the management server. The management server includes a processor for processing the weather datum and usage-specific data so as to generate a dosing recommendation for the inputs on the basis of a predetermined and learning dosing model, and second transmission means for transmitting the dosing recommendation to a communication terminal.

Abstract: Example systems, methods, and non-transitory computer readable media are directed to determining a geographic location to be assessed; obtaining information associated with the geographic location, the information including at least one data point of the geographic location; classifying one or more habitats within the geographic location based on at least one machine learning model that processes the at least one data point of the geographic location; determining at least one respective metric for the one or more classified habitats based at least in part on the at least one data point of the geographic location; and providing an interface that includes at least a map of the geographic location and the at least one respective metric for the one or more classified habitats, the one or more classified habitats are visually segmented in the map by habitat type.

Abstract: A technique for controlling a human machine interface is provided. A first task includes outputting a predefined criterion applicable to each of a plurality of physical objects. Each object is associated with a category within a group of at least pairwise disjoint categories with the criterion fulfilled for each object in a first category and not fulfilled for each object in a second category. Controlling the HMI for the first task includes repeatedly performing the steps of rendering an object; monitoring the HMI for an input during a predefined first time period after the rendering of the object; and updating a first metric indicative of a performance measurement in the first task. The method further comprises rendering a plurality of receptacles each enclosing one of the objects for a second task.

Abstract: The present disclosure relates to a system and a method for identification of individual animals based on images, such as 3D-images, of the animals, especially of cattle and cows. When animals live in areas or enclosures where they freely move around, it can be complicated to identify the individual animal. The present disclosure relates to a method for determining the identity of an individual animal in a population of animals with known identity, the method comprising the steps of acquiring at least one image of the back of a preselected animal, extracting data from said at least one image relating to the anatomy of the back and/or topology of the back of the preselected animal, and comparing and/or matching said extracted data against reference data corresponding to the anatomy of the back and/or topology of the back of the animals with known identity, thereby identifying the preselected animal.

Abstract: There is achieved short-time measurement of a measurement target such as a farm field with high accuracy. Therefore, processing is performed on a sampling image obtained by the imaging of a part of a range as a measurement target with a multi spectrum camera that performs imaging to capture images in a plurality of different wavelength bands. Then, arithmetic processing is performed on the sampling image as a processing target to generate a calculation result image serving as evaluation information for the entire measurement target.

Abstract: A maize ear scanning device and computer vision kernel counting pipeline is used to analyze a maize ear. The process of analyzing the maize ear comprises: placing the ear on a rotating motor, fastened at the top with a metal pin on the sliding portion of the scanner; starting the motor's rotation; capturing an image of rotating ear with camera; converting rotating ear video into flat image by extracting individual frames, cropping to center horizontal pixel row, and concatenating all pixel rows; and running computer vision pipeline on image to identify seed locations.

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 diagnostic support for a skin includes a radio-transparent structure that defines a folding surface of the skin and on which the skin may be stretched and consequently folded, thereby defining folded, mutually superimposed portions spaced apart from each other. The support may be used for radiographic inspection of a folded animal skin.

Abstract: The disclosure relates to methods and systems for determining muscle fascicle fracturing in a meat sample. In one embodiment, the methods and systems disclosed herein use automated methods to determine muscle fascicle fracturing. Methods and systems disclosed herein comprise imaging a meat sample with an imaging device to obtain one or more images; and using an iteratively trained training detection model to determine an objective classification, including the presence or absence of muscle fascicle fracturing in the images of the meat sample.

Abstract: The present application relates to the use of soft nano material integrated to micro-electronics circuit to detect signals from nutritional substances and the subsequent conversion of these signals into format suitable for a trained machine learning model through pre-processing steps prior to sending to the machine learning model as well as the output format of the machine learning model and post processing in order to obtain a final determination of the quality and predicted expiration date of a nutritional substance within a probabilistic confidence level. The application encompasses a soft nano-material integrated with micro-electronics for signal capture and an analytics platform for machine learning model training, classification and prediction with a probabilistic confidence level result.

Abstract: A method and a system for determining an indicator of processing quality of an agricultural harvested material using a mobile device is disclosed. A computing unit analyzes image data of a prepared sample of harvested material containing grain components and non-grain components in an analytical routine to determine the indicator of the processing quality of the agricultural harvested material. Further, the computing unit uses a trained machine learning model in the analytical routine to perform at least one step of determining the indicator of the processing quality of the agricultural harvested material and that the computing unit adjusts at least one machine parameter of the forage harvester based on the indicator of processing quality.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for escape detection and mitigation for aquaculture. In some implementations, a method includes obtaining one or more images that depict one or more fish within a population of fish that are located within an enclosure; providing, to one or more detection models configured to classify fish that are depicted within the images as likely being member or as likely not being member of a type of fish, the one or images; generating, as a result of providing the one or more images to the one or more detection models, a value that reflects a quantity of fish that are depicted in the images that are likely a member of the type of fish; and detecting a condition based at least on the value.

Abstract: One or more images are used to analyze permanent cropland. The images can be obtained from one or more sensors on an aerial vehicle including, but not limited to, UAVs. The analysis of the permanent cropland includes, but is not limited to, analyzing the plants growing in the permanent cropland or analyzing the permanent crops growing on the plants. In one embodiment, the analysis can include estimating a total yield of the permanent crops where the estimated permanent crop yield is more accurate than the estimated yield obtained using traditional methods.

Abstract: A residue vision system includes a harvesting machine configured to traverse a field and harvest an agricultural material, a residue distribution system carried by the harvesting machine and configured to distribute a residue of the agricultural material onto a first harvested area of the field, at least one camera coupled to the harvesting machine and configured to acquire an image of a second harvested area of the field, and an electronic control unit in communication with the at least one camera and the residue distribution system. The electronic control unit is configured to analyze the image acquired by the at least one camera, and in response to the analysis adjust the residue distribution system to adjust distribution of the residue onto the first harvested area of the field.

Abstract: A device may receive sensor data and video data associated with a vehicle, and may process the sensor data, with a rule-based detector model, to determine whether a probability of a vehicle accident satisfies a first threshold. The device may preprocess acceleration data of the sensor data to generate calibrated acceleration data, and may process the calibrated acceleration data, with an anomaly detector model, to determine whether the calibrated acceleration data includes anomalies. The device may filter the sensor data to generate filtered sensor data, and may process the filtered sensor data and anomaly data, with a decision model, to determine whether the probability of the vehicle accident satisfies a second threshold. The device may process the filtered sensor data, the anomaly data, and the video data, with a machine learning model, to determine whether the vehicle accident has occurred, and may perform one or more actions.

Abstract: The method for classifying ambiguous objects, including: determining initial labels for an image set; determining N training sets from the initially-labelled image set; training M annotation models using the N training sets; determining secondary labels for each image of the image set using the M trained annotation models; and determining final labels for the image set based on the secondary labels. The method can optionally include training a runtime model using images from the image set labeled with the final labels; and optionally using the runtime model.

Abstract: An electronic data processor is configured to estimate a spatial region of interest of plant pixels of one or more target plants in the obtained image data for a harvestable plant component and its associated harvestable plant component pixels of the harvestable plant component. The electronic data processor is configured to identify the component pixels of a harvestable plant component within the obtained image data of plant pixels of the one or more target plants. An edge, boundary or outline of the component pixels is determined. The data processor is configured to detect a size of the harvestable plant component based on the determined edge, boundary or outline of the identified component pixels. A user interface is configured to provide the detected size of the harvestable plant component for the one or more target plants as an indicator of yield of the one or more plants or standing crop in the field.

Abstract: Providing objective analysis of performance results of power equipment devices are detailed throughout this disclosure. Image data of vegetation cut by a power equipment device can be captured and uploaded to a server for analysis. Data pertinent to the performance results, such as environment conditions, mechanical parameter settings, type of vegetation or turf, and so forth can be submitted with the image data in some disclosed aspects. Analysis results, optionally in conjunction with adjustment data for adjusting mechanical settings of the power equipment can be provided in response to the image data.

Abstract: In some embodiments, a method can include receiving first images of produce. The method can further include executing a first machine learning model to generate second images of produce based on the first images of produce. The first images of produce can include (1) images of non-bagged produce or (2) images of bagged produce. The second images of produce can include the other of (1) images of non-bagged produce or (2) images of bagged produce. The method can further include training a second machine learning model based on the first images of produce and the second images of produce. The method can further include executing, after the training, the second machine learning model to classify as a bagged produce or a non-bagged produce an image not included in the first images and not included in the second images.

Abstract: A method and a system for automated plant surveillance and manipulation are provided. Pursuant to the method and the system, images of target plants are obtained through a machine vision system having multiple cameras. The obtained images of the target plants are processed to determine tissue candidates of the target plants and to determine a position and an orientation of each tissue candidate. A tool is manipulated, based on the position and the orientation of each tissue candidate, to excise each tissue candidate to obtain tissue samples. The tissue samples are transported for subsequently manipulation including live processing of the tissue samples or destructive processing of the tissue samples.

Abstract: A method for generating a localized data map, the method including (a) traversing an area with a machine, the machine including at least one sensor, wherein the sensor is configured to receive data; (b) collecting data of the area utilizing the sensor; and (c) communicating the data to generate a localized data map. A system and method for generating a localized turf grass data map, the method including (a) traversing an area of turf grass with an outdoor power equipment machine, the outdoor power equipment machine including at least one sensor, wherein the sensor is configured to receive data; (b) collecting turf grass data utilizing the sensor; and (c) communicating the turf grass data to generate a localized turf grass data map.

Abstract: Techniques are described for generating population counts for seed production plants based on one or more images obtained from a camera on an aerial vehicle including, but not limited to, UAVs. The image(s) is processed to identify the plurality of rows of seed production plants, classify each one of the identified rows as either male or female, and produce a count of the number of seed production plants in the male rows and produce a count of the number of seed production plants in the female rows. The seed production plants can be corn seed plants or any other type of seed production plant in which male and female seed production plants can be distinguished from one another based on one or more aerial images.

Abstract: In one aspect of the present application, a computer-implemented method for setting a local coordinate system of a tooth 3D digital model is provided, the method comprises: obtaining a first 3D digital model representing a first tooth, wherein the first 3D digital model is based on a world coordinate system; and setting a local coordinate system for the first 3D digital model using a first artificial neural network based on the first 3D digital model, where the first artificial neural network is a trained deep learning artificial neural network.

Abstract: Various embodiments of an apparatus, methods, systems and computer program products described herein are directed to an agricultural observation and treatment system and method of operation. The agricultural treatment system may determine a first real-world geo-spatial location of the treatment system. The system can receive captured images depicting real-world agricultural objects of a geographic scene. The system can associate captured images with the determined geo-spatial location of the treatment system. The treatment system can identify, from a group of mapped and indexed images, images having a second real-word geo-spatial location that is proximate with the first real-world geo-spatial location. The treatment system can compare at least a portion of the identified images with at least a portion of the captured images. The treatment system can determine a target object and emit a fluid projectile at the target object using a treatment device.

Abstract: A first obtaining unit obtains two images obtained by capturing images of an inspection target object under different capturing environments with different illumination conditions inputted by a user. A second obtaining unit obtains area information that specifies two types of areas set by the user regarding an area in each image. The two types of areas include a reference area serving as a reference for comparison between the areas and a comparison target area serving as a target of comparison. A converting unit generates color conversion parameters for carrying out color conversion such that a color value of the reference area in one image coincides with a color value of the reference area in the other image. A deriving unit derives a color difference between the reference area and the comparison target area regarding each of the two images.

Abstract: Computer vision grouping recognition is provided by receiving training images that include unpackaged items; identifying, by a computer vision model, a candidate identities for unpackaged items in a given training image; receiving, from a human user, a selected identity for the unpackaged item as feedback for the candidate identity; constructing a confusion matrix tallying matches and mismatches between candidate identities and the selected identities as analyzed across the training images for each unpackaged item; identifying at least one product category that includes at least a first unpackaged item and a second unpackaged item that the confusion matrix indicates as being misidentified for each other by the computer vision model; and reconfiguring the computer vision model to identify the product category instead of the first unpackaged item or the second unpackaged item when analyzing a given image including one or more of the first unpackaged item or the second unpackaged item.