Abstract: A controller using image processing to identify produce is disclosed herein. The controller may include one or more memories and one or more processors communicatively coupled to the one or more memories. In some implementations, a controller may receive a trigger associated with presence of an item at a product processing zone. The controller may capture, via an imager, an image representing the item. Accordingly, the controller may apply a first convolutional neural network (CNN) to the image to determine a bounding box associated with the item. The controller may determine that the item within the bounding box satisfies an occlusion threshold and may apply a second CNN to the image to determine a query image representation. Accordingly, the controller may receive, from a database, an indication of one or more candidate items based on the query image representation and indicate, via a user interface, the candidate item(s).

Abstract: Systems and methods for calibrating a camera using a calibration system are provided. The calibration system includes a mat, a camera, and a controller. The systems and methods include the controller configuring the mat to operation in an illumination mode in which the mat energizes one or more LEDs to form an illumination pattern. The controller receives image data generated by the camera that features the illumination pattern. The controller analyzes the image data to derive heading data for the camera and sets the camera heading data to be the derived heading data.

Abstract: Systems and methods for performing three-dimensional measurements are disclosed herein. An example system includes an imaging system with a scan platter defining a spatial region within which an object may be imaged. The imaging system has first optics configured to image a first field of view of the imaging system, and second optics configured to image a second field of view of the imaging system. The second field of view spatially overlaps at least partially with the first field of view to form an overlap region, with the overlap region being a three-dimensional volume extending along a length of the scan platter. The system may further include a processor configured to analyze images of the first and second fields of view and determine at least one common point between the first and second images. The processor then further determines three-dimensional information of an object from the determined common point.

Abstract: Devices and methods for computer vision guided analysis of a display module are disclosed herein. The method captures a burst of images of at least a portion of an object for displaying at least one item. The method detects at least one attribute of the object present in a first image of the burst of images and extracts the at least one attribute of the object present in the first image from each image of the burst of images. The method aligns the extracted at least one attribute from each image of the burst of images with the extracted at least one attribute of the first image and generates a reconstructed image based on the aligned at least one attribute of the burst of images where the resolution of the first image is different from the resolution of the reconstructed image.

Abstract: A method includes: capturing, by a scanner device comprising an image sensor, first image data representing at least a portion of a first item; decoding, by the scanner device, a first barcode represented in the first image data; determining a first item template associated with the first barcode, the first item template comprising first identifier data identifying the first item from among other items and first region-of-interest data specifying a first region-of-interest of the first item; generating second image data comprising the first region-of-interest of the first image data; determining, by a first machine learning model, that the second image data corresponds to the first identifier data identifying the first item; and generating first data indicating that the first barcode is matched with the first item.

Abstract: A method includes: capturing, by a scanner device comprising an image sensor, first image data representing at least a portion of a first item; detecting, by the scanner device, a first barcode represented in the first image data; decoding the first barcode to determine first identifier data for the first item; performing a lookup using the first identifier data to determine a first entry for the first item stored in a data store; generating a sub-image representing a region-of-interest of the first item; generating, using an image encoder, a first vector representation of the sub-image; comparing the first vector representation of the sub-image to at least one second vector representation stored in the data store in association with the first entry; and generating first data indicating that the first barcode is matched with the first item.

Abstract: Systems and methods for calibrating a camera using a calibration system are provided. The calibration system includes a mat, a camera, and a controller. The systems and methods include the controller configuring the mat to operation in an illumination mode in which the mat energizes one or more LEDs to form an illumination pattern. The controller receives image data generated by the camera that features the illumination pattern. The controller analyzes the image data to derive heading data for the camera and sets the camera heading data to be the derived heading data.

Abstract: Disclosed is a robot management system comprising a control circuit configured to receive data indicative of a worker task being completed by a human from a human task management system coupled to the robot management system, generate a new robot task based on the data, add the new robot task to a robot task queue, detect an available robot, determine capabilities of the available robot, and filter the robot task queue based on the capabilities of the available robot to generate a group of robot tasks, the group being a subset of the robot task queue and each task in the group can be performed by the available robot. The control circuit is further configured to determine a priority group comprising a priority for the available robot for each task in the group and assign a robot task from the group to the available robot based on the priority group.

Abstract: Devices and methods for planogram generation are disclosed herein. The method detects at least one first item and at least one label present in a captured image and associates the at least one first item with the at least one label based on a boundary between the at least one first item and at least one second item different from the at least one first item. The method identifies the at least one first item based on at least one attribute of the at least one first item and determines an area indicative of a position of the identified at least one first item based on the association. The area can be one or more of an aisle, a module, a shelf, a rack, a bay, and a bin. The method generates a planogram based on the association, the identified at least one first item and the area.

Abstract: A method in a computing device includes: in a facility containing a plurality of support structures, capturing an image of a first support structure; detecting, in the image, a first feature set of the first support structure; selecting obtaining at least one reference feature set by proximity to an estimated location of the mobile computing device in the facility coordinate system, the at least one reference feature set selected from a repository defining feature locations for each of the support structures in a facility coordinate system; comparing the first feature set with the at least one reference feature set; and in response to determining that the first feature set matches the at least one reference feature set, determining a location of the mobile computing device in the facility coordinate system based on the image and the feature locations from the repository.

Abstract: Systems and methods for lost asset management using photo-matching are disclosed herein. An example method includes capturing a lost asset image corresponding to a lost asset, and generating, by a feature extractor model, a lost asset descriptor that represents features of the lost asset image. The example method also includes storing the lost asset descriptor and the lost asset image in an asset database that includes known asset descriptors, and performing, by a visual search engine, a nearest neighbor search within the asset database to determine a respective metric distance between the lost asset descriptor and the known asset descriptors. The example method also includes determining, by the visual search engine, a ranked list of known assets corresponding to the lost asset, and displaying, at a user interface, the ranked list of known assets for viewing by a user.

Abstract: Systems and methods for enhancing decoding of indicia are disclosed herein. An example implementation includes: generating a visual embedding from image data; identifying, in the image data, a partial indicia and decoding the partial indicia to determine at least a partial payload; responsive to the at least partial payload not being a full payload, determining one or more candidate visual embeddings corresponding to the partial payload; comparing the one or more candidate visual embeddings to the visual embedding generated from the image data and assigning a correspondence score for each of the one or more candidate visual embeddings based on the comparison; and determining a recommendation based on the correspondence score for each of the one or more candidate visual embeddings.

Abstract: Systems and apparatuses for dual-mode pressure-sensing and illumination mats are provided. The systems include a surface that includes a plurality of row wires and a plurality of column wires forming intersection points. At the intersection points, the systems include a pressure sensor configured to sense a pressure exerted upon the surface and a light emitting diode (LED) connected in parallel. The systems include a controller coupled to the selector and configured to control operation of the surface, wherein the controller is configured to operate in one of a pressure-sensing mode or an illumination mode. When the controller is configured to operate in the pressure-sensing mode, the controller scans pressure values provided by the pressure sensors. When the controller is configured to operate in the illumination mode, the controller configures the surface to enable the LEDs in accordance with an illumination pattern.

Abstract: Techniques for three-dimensional product reconstruction using multiple images collected at checkout lanes are disclosed herein. An example method includes capturing, by a barcode reader associated with a point of sale (POS) workstation, first image data associated with each of a plurality of products passing through a product scanning region of the POS workstation; analyzing barcode data from the first image data captured by the barcode reader to identify each product of the plurality of products passing through the product scanning region of the POS workstation; capturing, by one or more color cameras associated with the POS workstation, second image data associated with the identified product, of the plurality of products passing through the product scanning region of the POS workstation; and generating, by a processor, a textured three-dimensional mesh reconstruction for the identified product based on the second image data associated with the identified product.

Abstract: A method in a computing device of detecting boundary deformation for an item includes: obtaining a current image of the item; obtaining, from the current image, a current attribute of a boundary of the item; retrieving a reference attribute of the item boundary, the reference attribute corresponding to the item in an initial state; comparing the current attribute to the reference attribute; determining, based on the comparison, whether the item boundary is deformed relative to the initial state of the item; and when the item boundary is deformed, generating an exception notification.

Abstract: A controller using image processing to identify produce is disclosed herein. The controller may include one or more memories and one or more processors communicatively coupled to the one or more memories. In some implementations, a controller may receive a trigger associated with presence of an item at a product processing zone. The controller may capture, via an imager, an image representing the item. Accordingly, the controller may apply a first convolutional neural network (CNN) to the image to determine a bounding box associated with the item. The controller may determine that the item within the bounding box satisfies an occlusion threshold and may apply a second CNN to the image to determine a query image representation. Accordingly, the controller may receive, from a database, an indication of one or more candidate items based on the query image representation and indicate, via a user interface, the candidate item(s).

Abstract: Techniques for three-dimensional product reconstruction using multiple images collected at checkout lanes are disclosed herein. An example method includes capturing, by a barcode reader associated with a point of sale (POS) workstation, first image data associated with each of a plurality of products passing through a product scanning region of the POS workstation; analyzing barcode data from the first image data captured by the barcode reader to identify each product of the plurality of products passing through the product scanning region of the POS workstation; capturing, by one or more color cameras associated with the POS workstation, second image data associated with the identified product, of the plurality of products passing through the product scanning region of the POS workstation; and generating, by a processor, a textured three-dimensional mesh reconstruction for the identified product based on the second image data associated with the identified product.

Abstract: A method and system for operating an indicia reader are disclosed herein. An example method includes detecting a first object in the scanning region; capturing one or more images of the first object to create first image data; determining that a successful decode of an indicia has not occurred; retrieving the first image data of the first object; generating, using the first image data, an embedding of the first image data; detecting a second object in the scanning region; retrieving one or more support samples from an image database; comparing the embedding of the image data to each support sample of the one or more support samples; based on the comparison that the embedding of the image data does not match any support sample, performing an operation.

Abstract: A method and system for operating an indicia reader are disclosed herein. An example method includes detecting a first object in the scanning region; capturing one or more images of the first object to create first image data; determining that a successful decode of an indicia has not occurred; retrieving the first image data of the first object; generating, using the first image data, an embedding of the first image data; detecting a second object in the scanning region; retrieving one or more support samples from an image database; comparing the embedding of the image data to each support sample of the one or more support samples; based on the comparison that the embedding of the image data does not match any support sample, performing an operation.

Abstract: Systems and methods for lost asset management using photo-matching are disclosed herein. An example method includes capturing a lost asset image corresponding to a lost asset, and generating, by a feature extractor model, a lost asset descriptor that represents features of the lost asset image. The example method also includes storing the lost asset descriptor and the lost asset image in an asset database that includes known asset descriptors, and performing, by a visual search engine, a nearest neighbor search within the asset database to determine a respective metric distance between the lost asset descriptor and the known asset descriptors. The example method also includes determining, by the visual search engine, a ranked list of known assets corresponding to the lost asset, and displaying, at a user interface, the ranked list of known assets for viewing by a user.