Honeywell International Inc.

Hand Held Products, Inc.

Ben Miller

Scott McCloskey

Saad J Bedros

Yunqian Ma

Isaac D Cohen

Gurumurthy Swaminathan

Michael Janssen

Michael Ray Elgersma

Thomas R Markham

Valerie Guralnik

Pedro Davalos

Mahesh Kumar Gellaboina

Kirk A Schloegel

Roland Miezianko

Dinesh Ramegowda

Andrew H Johnson

Yadhunandan U S

Yadhunandan Ullam Subbaraya

MaryRosaline Mariappan

Scott Mccloskey

An imaging device has a light field imager and a processor. The light field imager includes a microlens array, and a light field sensor positioned proximate to the microlens array, having a plurality of pixels and recording light field data of an optical target from light passing through the microlens array. The processor is configured to: receive the light field data of the optical target from the light field sensor, estimate signal to noise ratio and depth of the optical target in the light field data, select a subset of sub-aperture images based on the signal to noise ratio and depth, combine the selected subset of sub-aperture images, and perform image analysis on the combined subset of sub-aperture images.

Systems and methods for imaging

A method and system for automatic face recognition. A primary and a plurality of secondary video cameras can be provided to monitor a detection area. The primary video camera can detect people present in the detection zone. Data can be then transmitted to a prioritizor module that produces a prioritized list of detected people. The plurality of secondary video cameras then captures a high-resolution image of the faces of the people present in the detection area according to the prioritized list provided by the prioritizor module. The high-resolution images can be then provided to a face recognition module, which is used to identify the people present in the detection area.

Distant face recognition system

Devices, methods, and systems for detecting motion in a high resolution video are described herein. One method includes dividing an image frame of a high resolution video into a number of blocks, determining whether motion is present in each of the blocks, and identifying, within each block in which motion is determined to be present, pixels in which motion is present.

Detecting motion in a high resolution video

Devices, systems, and methods of performing feature detection on a set of multimedia files are disclosed. One method of organization includes identifying a feature from each multimedia file within the set of multimedia files wherein each file has one feature, organizing the features based on their similarities wherein similar features are grouped based upon a proximity in a feature space and a representative feature is identified for each group, receiving a detection model having one or more detection criteria the detection model having previously been trained for detection using the organized features, and using the representative features to apply the detection model in a decreasing order of detection probability in order to detect the files satisfying the detection criteria within the set of multimedia files.

Discriminative classification using index-based ranking of large multimedia archives

Methods, devices, and systems for object tracking are described herein. One or more method embodiments include receiving an initial set of track points associated with a trajectory of an object, compressing the initial set of track points into a plurality of track segments, each track segment having a start track point and an end track point, and storing the plurality of track segments to represent the trajectory of the object.

Object tracking

Methods, machine-readable media, and devices for face image prioritization based on face quality analysis are described herein. For example, one or more embodiments include detecting a facial image in an image that has been acquired by a camera that monitors a scene, passing the facial image through a number of quality analysis filters that include a number of quality analysis factors, wherein processing complexity associated with the number of quality analysis factors increases consecutively, and submitting the facial image to a facial recognition program upon a determination that the facial image has passed the number of quality analysis filters.

Face image prioritization based on face quality analysis

An apparatus includes at least one memory unit and at least one processing unit. The memory unit is configured to receive and store information associated with a particular geographical region. The processing unit configured to execute a computer program for discretizing an image of the geographical region into a plurality of sub-regions. Using these sub-regions, the processing unit may generate a graph comprising a plurality of nodes and a plurality of edges in which the nodes comprise the sub-regions. The processing unit may also geospatially partition an image of the geographical region according to the information that is associated with each of the sub-regions.

System and method for geospatial partitioning of a geographical region

Various embodiments illustrated and described herein include at least one of systems, methods, and software that utilizes imagery from a single high-resolution camera to capture images, locate individuals, and provide images to a facial recognition process. Some embodiments also include prioritization and filtering that choose which captured images from a stream of images to process and when to process them when there are many images to be processed.

Surveillance-based high-resolution facial recognition

A process includes receiving an archive of video data comprising a plurality of archive descriptor types, and applying a query to the archive. The query includes a number of N query descriptor types for a query object. The process further includes determining a difference between each query descriptor type and corresponding descriptor types of the archive, and storing each difference as a point in an N dimensional space. The process further includes identifying an archive object that is similar to the query object as a function of proximities of the differences to an origin of the N dimensional space.

Distance-space embedding for multi-descriptor matching and retrieval

A system having an approach for prioritizing targets for an order of capturing the targets photographically or otherwise. Prioritizing is based on cost of obtaining or capturing the target for viewing or photographing in high resolution. One acquisition mechanism is for obtaining a wide field of view of a scene of targets and another acquisition mechanism is for obtaining a narrow field of view of a target for capture. The cost for prioritizing is based on the time that the narrow field of view acquisition mechanism takes to pan and tilt to get a close-up image of a target divided by the width of the target. The targets may be faces of people.

Prioritizer system for target acquisition

A system includes a motion detection processor, a motion tracking processor, a people detection processor, a controller, a fusion processor, an appearance model generator processor, a database, a fast search processor, and a matching processor. The motion detection processor, the motion tracking processor, the controller, the people detection processor, the fusion processor, and the appearance model generator processor comprise an analytics pipeline, and the database and the fast search processor comprise a data index pipeline.

System for finding archived objects in video data

Video data matching includes both a single region of data and sequences of regions of data. In an embodiment, a video processing system selects, from a first matrix row corresponding to a test appearance model, one or more other appearance models as a function of similarity measures populating the first matrix row. After selection of the one or more other appearance models, the system then selects, from other matrix rows corresponding to the one or more other appearance models selected in the first step, one or more additional appearance models as a function of the similarity measures populating the other matrix rows. The system then ranks the appearance models selected from the first matrix row and the other matrix rows.

Video data matching using clustering on covariance appearance

A seed search of a subset of analytical data corresponding to video objects displayable in a plurality of video frames is carried out to identify video objects that most closely match a selected video object and then complete searches of the analytical data may be carried out so as to identify video objects that most closely match each video object identified during the seed search. The video objects having the greatest number of occurrences of being identified during the complete searches may be displayed by a graphical user interface (GUI). In this way, the GUI may display the video objects in an order based on how closely each video object matches the selected video object and/or a video object identified during the seed search, which may an order different than an order based on a time when each video object was captured.

Growing community of inventors

Minneapolis, MN, United States of America

Ben Miller