Abstract: This invention provides a system and method for detecting and acquiring one or more in-focus images of one or more barcodes within the field of view of an imaging device. A measurement process measures depth-of-field of barcode detection. A plurality of nominal coarse focus settings of a variable lens allow sampling, in steps, of a lens adjustment range corresponding to allowable distances between the one or more barcodes and the image sensor, so that a step size of the sampling is less than a fraction of the depth-of-field of barcode detection. An acquisition process acquires a nominal coarse focus image for each nominal coarse focus setting. A barcode detection process detects one or more barcode-like regions and respective likelihoods. A fine focus process fine-adjusts, for each high-likelihood barcode, the variable lens near a location of the barcode-like regions. The process acquires an image for decoding using the fine adjusted setting.

Abstract: This invention overcomes disadvantages of the prior art by providing a vision system and method of use, and graphical user interface (GUI), which employs a camera assembly having an on-board processor of low to modest processing power. At least one vision system tool analyzes image data, and generates results therefrom, based upon a deep learning process. A training process provides training image data to a processor (optionally) remote from the on-board processor to cause generation of the vision system tool therefrom, and provides a stored version of the vision system tool for runtime operation on the on-board processor. The GUI allows manipulation of thresholds applicable to the vision system tool and refinement of training of the vision system tool by the training process.

Abstract: A system may comprise a transport device for moving at least one object, wherein at least one substantially planar surface of the object is moved in a known plane locally around a viewing area, wherein the substantially planar surface of the object is occluded except when the at least one substantially planar surface passes by the viewing area, at least one 2D digital optical sensor configured to capture at least two sequential 2D digital images of the at least one substantially planar surface of the at least one object that is moved in the known plane around the viewing area, and a controller operatively coupled to the 2D digital optical sensor, the controller performing the steps of: a) receiving a first digital image, b) receiving a second digital image, and c) stitching the first digital image and the second digital image using a stitching algorithm to generate a stitched image.

Abstract: This invention provides a system and method for enhanced depth of field (DOF) advantageously used in logistics applications, scanning for features and ID codes on objects. It effectively combines a vision system, a glass lens designed for on-axis and Scheimpflug configurations, a variable lens and a mechanical system to adapt the lens to the different configurations without detaching the optics. The optics can be steerable, which allows it to adjust between variable angles so as to optimize the viewing angle to optimize DOF for the object in a Scheimpflug configuration. One, or a plurality, of images can be acquired of the object at one, or differing angle settings, with the entire region of interest clearly imaged. In another implementation, the optical path can include a steerable mirror and a folding mirror overlying the region of interest, which allows different multiple images to be acquired at different locations on the object.

Abstract: The techniques described herein relate to methods and systems for processing symbols quickly and robustly. The techniques can include using a pre-trained deep learning model to generate a region of interest (ROI) of an image captured by an imaging device according to a set of attributes associated with the imaging device. The techniques can include using a machine learning model to generate a quality metric for the image. The quality metric can indicate a measurement that the ROI of the symbol can be decoded. The set of attributes of the imaging device can be adjusted based on the quality metric before taking another image until a quality metric satisfies predetermined criteria. Such techniques enable fast and robust symbol processing with compact system configurations and easily setup components.

Abstract: This invention provides a system and method for using an area scan sensor of a vision system, in conjunction with an encoder or other knowledge of motion, to capture an accurate measurement of an object larger than a single field of view (FOV) of the sensor. It identifies features/edges of the object, which are tracked from image to image, thereby providing a lightweight way to process the overall extents of the object for dimensioning purposes. Logic automatically determines if the object is longer than the FOV, and thereby causes a sequence of image acquisition snapshots to occur while the moving/conveyed object remains within the FOV until the object is no longer present in the FOV. At that point, acquisition ceases and the individual images are combined as segments in an overall image. These images can be processed to derive overall dimensions of the object based on input application details.

Abstract: The techniques described herein relate to methods, apparatus, and computer readable media configured to identify a surface feature of a portion of a three-dimensional (3D) point cloud. Data indicative of a path along a 3D point cloud is received, wherein the 3D point cloud comprises a plurality of 3D data points. A plurality of lists of 3D data points are generated, wherein: each list of 3D data points extends across the 3D point cloud at a location that intersects the received path; and each list of 3D data points intersects the received path at different locations. A characteristic associated with a surface feature is identified in at least some of the plurality of lists of 3D data points. The identified characteristics are grouped based on one or more properties of the identified characteristics. The surface feature is identified based on the grouped characteristics.

Abstract: The techniques described herein relate to methods and systems for three-dimensional (3D) inspection using deep learning model pre-trained with two-dimensional (2D) images. The techniques include transforming a 3D representation (e.g., captured 3D point cloud, 3D profiles, meshes, voxels) to a 2D map, which can be input to a deep learning model pre-trained with 2D images. The 2D map includes elements disposed in an array. Each element includes a vector of a number of geometric features. Such a configuration enables the 2D map to be in a structure acceptable by the 2D deep learning model. The 2D deep learning model generates an output based on the 2D map and provides the output to a subsystem for generating an inspection result. The inspection result can have a 3D result such as a surface area and/or volume.

Abstract: Compact and robust machine vision systems are provided herein. A machine vision system includes an optoelectronic system configured to emit measurement beams towards objects and a motion module configured to move the objects and/or components of the optoelectronic system. The machine vision system uses the relative motions to reduce degrees of freedom required to sample the objects to produce 3D images. The optoelectronic system includes a light source configured to emit a beam and sweep its frequency. The optoelectronic system includes primary interferometers configured to measure distances to points of an object and auxiliary interferometers configured with reference delays. The optoelectronic system includes electronic circuitry configured to interpret outputs of the interferometers into distance measurements by simpler computations in the time domain.

Abstract: This invention provides a system and method that efficiently detects objects imaged using a 3D camera arrangement by referencing a cylindrical or spherical surface represented by a point cloud, and measures variant features of an extracted object including volume, height, and center of mass, bounding box, and other relevant metrics. The system and method, advantageously, operates directly on unorganized and un-ordered points, requiring neither a mesh/surface reconstruction nor voxel grid representation of object surfaces in a point cloud. Based upon a cylinder/sphere reference model, an acquired 3D point cloud is flattened. Object (blob) detection is carried out in the flattened 3D space, and objects are converted back to the 3D space to compute the features, which can include regions that differ from the regular shape of the cylinder/sphere. Downstream utilization devices and/or processes, such as part reject mechanism and/or robot manipulators can act on the identified feature data.

Abstract: This invention provides a system and method for finding patterns in images that incorporates neural net classifiers. A pattern finding tool is coupled with a classifier that can be run before or after the tool to have labeled pattern results with sub-pixel accuracy. In the case of a pattern finding tool that can detect multiple templates, its performance is improved when a neural net classifier informs the pattern finding tool to work only on a subset of the originally trained templates. Similarly, in the case of a pattern finding tool that initially detects a pattern, a neural network classifier can then determine whether it has found the correct pattern. The neural network can also reconstruct/clean-up an imaged shape, and/or to eliminate pixels less relevant to the shape of interest, therefore reducing the search time, as well significantly increasing the chance of lock on the correct shapes.

Abstract: This invention provides a vision system camera, and associated methods of operation, having a multi-core processor, high-speed, high-resolution imager, FOVE, auto-focus lens and imager-connected pre-processor to pre-process image data provides the acquisition and processing speed, as well as the image resolution that are highly desirable in a wide range of applications. This arrangement effectively scans objects that require a wide field of view, vary in size and move relatively quickly with respect to the system field of view. This vision system provides a physical package with a wide variety of physical interconnections to support various options and control functions. The package effectively dissipates internally generated heat by arranging components to optimize heat transfer to the ambient environment and includes dissipating structure (e.g. fins) to facilitate such transfer.

Abstract: Methods and systems for extracting a one-dimensional (1D) signal from a two-dimensional (2D) digital image along a projection line are provided herein. The methods and systems store the digital image in a memory hierarchy wherein non-blocking prefetch operations can fetch pixels from a main store to a data cache. A prefetch plan, pixel processing plan, and prefetch distance are selected responsive to the orientation of the projection line. The prefetch plan uses a first memory address order that is designed to be favorable for efficiently fetching pixels from the main store to the data cache for the given orientation. The pixel processing plan uses a second address order that is designed to be favorable for computing a 1D signal along the projection line.

Abstract: A system or method can analyze symbols on a set of objects having different sizes. The system can identify a characteristic object dimension corresponding to the set of objects. An image of a first object can be received, and, a first virtual object boundary feature (e.g., edge) in the image can be identified for the first object based on the characteristic object dimension. A first symbol can be identified in the image, and whether the first symbol is positioned on the first object can be determined, based on the first virtual object boundary feature.

Abstract: This invention provides a system and method for determining the location and characteristics of certain surface features that comprises elevated or depressed regions with respect to a smooth surrounding surface on an object. A filter acts on a range image of the scene. A filter defines an annulus or other perimeter shape around each pixel in which a best-fit surface is established. A normal to the pixel allows derivation of local displacement height. The displacement height is used to establish a height deviation image of the object, with which bumps, dents or other height-displacement features can be determined. The bump filter can be used to locate regions on a surface with minimal irregularities by mapping such irregularities to a grid and then thresholding the grid to generate a cost function. Regions with a minimal cost are acceptable candidates for application of labels and other items in which a smooth surface is desirable.

Abstract: In accordance with some embodiments of the disclosed subject matter, methods, systems, and media for generating images of multiple sides of an object are provided. In some embodiments, a method comprises receiving information indicative of a 3D pose of a first object in a first coordinate space at a first time; receiving a group of images captured using at least one image sensor, each image associated with a field of view within the first coordinate space; mapping at least a portion of a surface of the first object to a 2D area with respect to the image based on the 3D pose of the first object; associating, for images including the surface, a portion of that image with the surface of the first object based on the 2D area; and generating a composite image of the surface using images associated with the surface.

Abstract: Embodiments relate to predicting height information for an object. First distance data is determined at a first time when an object is at a first position that is only partially within the field-of-view. Second distance data is determined at a second, later time when the object is at a second, different position that is only partially within the field-of-view. A distance measurement model that models a physical parameter of the object within the field-of-view is determined for the object based on the first and second distance data. Third distance data indicative of an estimated distance to the object prior to the object being entirely within the field-of-view of the distance sensing device is determined based on the first distance data, the second distance data, and the distance measurement model. Data indicative of a height of the object is determined based on the third distance data.

Abstract: An on-axis aimer and distance measurement apparatus for a vision system can include a light source configured to generate a first light beam along a first axis. The first light beam can project an aimer pattern on an object and a receiver can be configured to receive reflected light from the first light beam to determine a distance between a lens of the vision system and the object. One or more parameters of vision system can be controlled based on the determined distance.

Abstract: This invention provides a system and method that performs 3D imaging of a complex object, where image data is likely lost. Available 3D image data, in combination with an absence/loss of image data, allows computation of x, y and z dimensions. Absence/loss of data is assumed to be just another type of image data, and represents the presence of something that has prevented accurate data from being generated in the subject image. Segments of data can be connected to areas of absent data and generate a maximum bounding box. The shadow that this object generates can be represented as negative or missing data, but is not representative of the physical object. The height from the positive data, the object shadow size based on that height, the location in the FOV, and the ray angles that generate the images, are estimated and the object shadow size is removed from the result.