Datalogic IP Tech S.r.l.

Francesco D'Ercoli

Marco Cumoli

Simone Cilli

Claudio Saporetti

Francesco Paolo Muscaridola

Alessandro Brunetta

Matteo Del Castello

Federico Canini

Marco Bassani

Alessandro Franchi

Vittorio Rago

Silvia Marinoni

Mattia Mantovani

Giorgio Sanzeni

Donald Bressan

Davide Benghi

Michele Gruppioni

Nicola Fioraio

Ali Al-Erwi

Amedeo Pisapia

Federica Albertini

The disclosure includes imaging systems employing event-based sensors in various applications for analyzing a scene, such as those involved in conveyor systems, mass flow detection systems, portals, machine vision systems, retail settings including checkout stations, and the like. Event-based sensors may be operated in conjunction with frame-based cameras to select frames or subparts of frames or otherwise trigger actions related to the imaging system based on analysis of the events generated by the event-based sensors.

Imaging systems with enhanced functionalities with event-based sensors

Systems and methods for autofocus may include a light source for generating a light wave and a varifocal lens arranged in front of the light source. The varifocal lens may receive the light wave and generate a Fourier transform of a known semi-transparent pattern positioned on the rear focal plane (or input plane) of the varifocal lens therefrom. An image sensor may receive the Fourier transform carried by the light wave after being reflected from an object. A focus tunable lens may be arranged in front of the image sensor and through which the reflected light wave passes. A processor may adjust a focal length of the varifocal lens to cause the Fourier transform carried by the light wave to form a predefined (expected) pattern detected by the image sensor, and adjust a control parameter of the focus tunable lens until one or more spatial frequencies of the predefined pattern detected at the image sensor match one or more predefined spatial frequencies.

Optical systems and methods for active autofocus

A predictive system and process that predicts safety system activation in industrial environments when collaborative robots (COBOTs), automated guidance vehicles (AGVs), and other robots (individual or collectively 'robots') are interacting (i) between one another or (ii) between a robot and human. As provided herein, the predictive system is not meant to substitute traditional safety systems, but rather to detect and classify robot-to-robot and robot-to-human interactions and potential interactions thereof so as to limit or avoid those interactions altogether, thereby increasing safety and efficiency of the robots.

Multi-sensor optimization of automatic machines to prevent safety issues

A computer-implemented method of performing a three-dimensional 3D point cloud registration with multiple two-dimensional (2D) images may include estimating a mathematical relationship between 3D roto-translations of dominant planes of objects in a 3D point cloud and bi-dimensional homographies in a 2D image plane, thereby resulting in a 3D point cloud registration using multiple 2D images. A trained classifier may be used to determine correspondence between homography matrices and inferred motion of the dominant plane(s) on a 3D point cloud for paired image frames. A homography matrix between the paired images of the dominant plane(s) on the 2D image plane may be selected based on the correspondence between the inferred motions and measured motion of the dominant plane(s) on the 3D point cloud for the paired image frames. The process may be less computationally intensive than conventional 2D-3D registration approaches.

System and method of 3D point cloud registration with multiple 2D images

The present disclosure relates to a method for image classes definition and to a method for image multiprocessing and related vision system, which implement said method for image classes definition. The latter comprising an image splitting operation for each image of M input images, the image splitting operation comprising the steps of: a) splitting the image into image portions; b) executing the algorithm onto each image portion with at least one processing unit; c) identifying the image portion associated with a maximum execution time of said algorithm; d) splitting said identified image portion into further image portions; e) checking if a stop criterion is met: e1) if the stop criterion is met, iterating steps a) to e) onto another one of the M input images; e2) if the stop criterion is not met, executing a predefined image processing algorithm onto each of the further image portions; identifying the image portion or further image portion associated with a maximum execution time of said algorithm; and iterating steps d) to e) on the so identified image portions portion or further image portion; wherein after executing steps a) to e) on all of the M input images, the method for image classes definition further comprises the steps of: f) identifying in an image space, for all the M input images, the positions of each split image portion/further image portion and defining clusters (A, B, C) based thereon; g) defining a set of Q image classes (A′, B′, C, AC) based on said clusters (A, B, C), each class (A′, B′, C, AC) being univocally associated with a splitting pattern representing in the image space a plurality of regions to be allocated to a corresponding plurality of processing units of the vision system for image multiprocessing.

Image multiprocessing method for vision systems

A system and method for measuring volume dimensions of objects may include flying a UAV to measuring points around an object within a defined area. Images of the object may be captured by the UAV at each of the measuring points. The captured images may be communicated by the UAV to a computing device remotely positioned from the UAV. Volume dimensions of the object may be computed based on the captured images. The volume dimensions of the object may be presented. In presenting the volume dimensions, the volume dimensions may be presented to a user via an electronic display.

Apparatus and method for pallet volume dimensioning through 3D vision capable unmanned aerial vehicles (UAV)

Systems and methods for applying and detecting cross dependent marks incorporated into an electronic or digital image to form a watermark. The electronic or digital image may include encoded information for example a machine-readable symbol. The watermarking may include an encoding and insertion sub-process that inserts one or more marks into an image at a first point in time for form a marked image, an extraction sub-process that extracts the marks at a second point in time, and a detection sub-processthat determines if any modifications have been made to the marked image. The marked image may be formed by determining a first original descriptor and first original mark within the image, determining a second original descriptor and second original mark within the image, and incorporating the first original mark into the second original descriptor and incorporating the second original mark into the first original descriptor.

Systems and methods for watermarking digital images

Embodiments disclosed herein describe systems and methods for authenticating a new device to operate on a network using peer-to-peer protocol key exchange. An existing network node sharing common secret seed information may initially exchange public keys with the new device. After the initial exchange, the network node and the new device may exchange one or more protocol messages. A received protocol message may include a pseudo-random number generated based upon a challenge position in a previously sent protocol message. If the network node determines a match between the received pseudo-random number at the challenge position and a locally generated pseudo-random number, the network node may authenticate the new device to the network.

Systems and methods for public key exchange employing a peer-to-peer protocol

A system and method of executing a deep neural network (DNN) in a local area network (LAN) may include executing a partitioned deep neural network in multiple computational nodes (CPs) in devices operating on the LAN. An image frame may be captured by a device. The image frame may be processed by a first layer of the partitioned neural network by a CP operating on the device. In response to the device that captured the image frame determining to request processing assistance from another CP, a request using a peer-to-peer protocol to other CPs on the LAN may be performed. A feature map may be communicated to another CP selected using the peer-to-peer protocol to process the feature map by a next layer of the DNN.

Deployment of deep neural networks (DNN) in embedded devices by means of peer-to-peer routing between computational points

Simple device replacement (SDR), low-cost class A compatible devices that store and use configuration tables for use in performing a node replacement may be used. The use of SDR using 802.3 Ethernet-enabled class A devices allows for dispensing with an external management computer or device-mounted additional hardware (e.g., push-button, SD card, etc.), and for performing node replacement while the network remains operating. A modified token ring or logical ring configuration and process may also be utilized. In an embodiment, a logical ring with a token, which may be used as a distributed scheduler that is circulated amongst network nodes, may be provided. The use of a modified token ring may provide for managing low-priority traffic on an 802.3 network for sensors to utilize low-priority communications without impacting high-priority communications.

Simple device replacement in a Profinet IO conformance class A (CCA) network through ubiquitous computing paradigm and combining a token ring approach with a ubicomp paradigm to prevent real-time performance drop

Systems and methods for providing reliability and redundancy of data for networked nodes (e.g., sensors and/or actuators) is provided. Each sensor may operate as network node in the sensor network that may operate as a peer to peer (P2P) consensus network. Each network node may maintain its local copy of a data chain and may generate a data block for updating the data chain. After a threshold number of network nodes generate the same consensus block from the data block, each network node may update its local data chain by appending the consensus block thereto.

Systems and methods for consensus-based data security for networked devices

A system and method of processing and analyzing data captured by automatic data collection ('ADC') readers. The ADC readers encapsulate the captured data in markup language pages and associate various metadata with these pages. A markup language document analyzer identifies associations or links between the meta-data stored in multiple markup language pages. Once an association is detected, the analyzer stores a corresponding hyperlink in one or both of the associated HTML pages. After the hyperlinks are stored, a rankings engine may rank the markup language pages based on the hyperlinks that have been stored. The rankings engine may continue to rank the pages as new links are identified and as new pages are created and stored. Queries related to the data received from the ADC readersmay be received and processed based upon the rankings of the processed markup language pages.

Method and algorithms for auto-identification data mining through dynamic hyperlink search analysis

A hand-held barcode reader has a decoding module that analyzes barcodes on a surface. The decoding module determines a displacement based on a current position of the hand-held barcode reader and a desired new position of the hand-held barcode reader with respect to a barcode on the surface. An LED of the hand-held barcode reader produces a first-colored light-spot at a position on the surface to direct a desired movement of the hand-held barcode reader to move from the current position to the desired new position.

Light projected feedback for guiding hand movements

A system and method for managing a shared resource in a network including a plurality of nodes including, in response to joining the network, broadcasting, by a node, an identification message to the network, and receiving, by the node, a token data structure including computational data including a threshold total resource consumption value of the plurality of nodes, current resource consumption values of respective nodes, and current utility function values of the respective nodes, the computational data being used by the node to adjust consumption of the shared resource by the node.

Automatic shared resource management system and associated methods

Various embodiments of the invention are implemented for an entry level, compact and lightweight single package apparatus combining a conventional high-resolution two-dimensional (2D) camera with a low-resolution three-dimensional (3D) depth image camera, capable to learn, through depth information, how to improve the performance of a set of 2D identification and machine vision algorithms in terms of speed-up (e.g. through regions of interests (ROIs)) and raw discriminative power. The cameras simultaneously capture images that are processed by an Intelligent Segmentation Engine in the system to facilitate object recognition.

Growing community of inventors

Bologna, Italy

Francesco D'Ercoli