International Business Machines Corporation

Charles A Otto

Sharathchandra UmapathiRao Pankanti

Norman C Haas

Ying Li

Hoang Trinh

Arun Hampapur

Rogerio Schmidt Feris

Myron Dale Flickner

Duan D Tran

Yun Zhai

Russell Patrick Bobbitt

Ankur Datta

Jonathan Hudson Connell, Ii

Sharathchandra Umapathirao Pankanti

Akira Yanagawa

Unsang Park

Yuichi Fujiki

Balamanohar Paluri

Jun Jun Li

Yuichi Fujiji

Video image data is acquired from synchronized cameras having overlapping views of objects moving past the cameras through a scene image in a linear array and with a determined speed. Processing units generate one or more object detections associated with confidence scores within frames of the camera video stream data. The confidence scores are modified as a function of constraint contexts including a cross-frame constraint that is defined by other confidence scores of other object detection decisions from the video data that are acquired by the same camera at different times; a cross-view constraint defined by other confidence scores of other object detections in the video data from another camera with an overlapping field-of-view; and a cross-object constraint defined by a sequential context of a linear array of the objects, spatial attributes of the objects and the determined speed of the movement of the objects relative to the cameras.

Multisensor evidence integration and optimization in object inspection

A system and computer program product for performing visual surveillance of one or more moving objects include registering one or more images captured by one or more cameras, wherein registering the one or more images comprises region-based registration of the one or more images in two or more adjacent frames, performing motion segmentation of the one or more images to detect one or more moving objects and one or more background regions in the one or more images, and tracking the one or more moving objects to facilitate visual surveillance of the one or more moving objects.

Detection and tracking of moving objects

A global position of an observed object is determined by obtaining a first global position of an observed object with at least one positioning device. A determination is made as to whether a set of stored visual characteristic information of at least one landmark matches a visual characteristic information set obtained from at least one captured image comprising a scene associated with the observed object. In response to the set of stored visual characteristic information matching the obtained visual characteristic information set, a second global position of the observed object is determined based on a set of stored location information associated with the at least one landmark and the first global position.

Location determination for an object using visual data

An imaging system includes an image capturing device and a plurality of reflective devices. The image capturing device is configured to receive a plurality of images reflected by the plurality of reflective devices. Responsive to receiving the plurality of images, the image capturing device is further configured to capture within a single frame at least a first image corresponding to a first side of a first railroad track rail, a second image corresponding to a second side of the first railroad track rail, a third image corresponding to a first side of a second railroad track rail, and a fourth image corresponding to a second side of the second railroad track rail.

System and method for inexpensive railroad track imaging for inspection

Techniques for segmenting an object are provided. The techniques include capturing an image of an object, dividing the image into one or more blocks, computing a confidence value for each of the one or more blocks, and eliminating one or more blocks from consideration based on the confidence value for each of the one or more blocks.

Object segmentation at a self-checkout

An apparatus, computer program product, and method for reducing persistent shadows within an image. The apparatus includes a camera configured to generate frames of the image. The apparatus also includes a computer processor. The computer processor calculates the average normalized brightness for each pixel in the image and adjusts the brightness of each pixel with the average normalized brightness.

Frame to frame persistent shadow reduction within an image

A method, data processing system, and computer program product for identifying abnormalities in data. A model representing a plurality of modes for an activity generated from training data is retrieved. The training data includes a first plurality of measurements of a first performance of the activity over a period of time. Each of the plurality of modes is identified as one of normal and abnormal. Activity data including a second plurality of measurements of a second performance of the activity is received. A portion of the activity data is compared with the plurality of modes in the model. A notification of an abnormality in the second performance of the activity is generated in response to an identification that the portion of the activity data matches a mode in the plurality of modes identified as abnormal. Confirmation of the abnormality is requested via a user interface.

Identifying abnormalities in resource usage

Techniques for segmenting an object at a self-checkout are provided. The techniques include capturing an image of an object at a self-checkout, dividing the image into one or more blocks, computing a confidence value for each of the one or more blocks, and eliminating one or more blocks from consideration based on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object.

A system, method, and computer program product for detecting anomalies in an image. In an example embodiment the method includes partitioning each image of a set of images into a plurality of image local units. The method further includes clustering all local units in the image set into clusters, and consequently assigning a class label to each local unit based on the clustering results. The local units with identical class labels having at least one substantially related image feature. Further, the method includes assigning a weight to each of the local units based on a variation of the class labels across all images in a set of images. The method further includes performing a clustering over all images in the set by using a distance metric that takes the learned weight of each local unit into account, then determining the images that belong to minorities of the clusters as anomalies.

Anomaly detection in images and videos

A method, system, and computer program product for automatically inspecting railroad tracks. The method includes assessing a configuration of rail components depicted in an image by comparing the configuration of the rail components to known hazards. The method also includes determining a severity of detected problems in the configuration of the rail components, using a computer processor.

Method and system of rail component detection using vision technology

A method, data processing system, and computer program product for managing a headlight on a vehicle are presented. An image of an area in front of the vehicle is received. A first set of features is identified in the received image. The first set of features in the received image is compared with a number of sets of features from a plurality of previous images. Each image in the plurality of previous images is associated with a headlight setting. A second set of features from a previous image in the plurality of previous images matching the first set of features in the received image is identified. A determination is made whether to change a setting for the headlight on the vehicle based on the headlight setting associated with the previous image.

Vehicle headlight management

Techniques are provided. The techniques include capturing an image of an object at a self-checkout, dividing the image into one or more blocks, computing one or more features of the image, computing a confidence value for each of the one or more blocks, wherein computing a confidence value for each of the one or more blocks comprises using a minimum feature distance from one or more reference background blocks, and eliminating one or more blocks from consideration via use of an adaptive threshold computed on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object.

The present invention provides a solution for controlling checkout throughput. Specifically, under the present invention a set of attributes of an item will be obtained. An analysis of the set of attributes will then be performed to determine whether an inconsistency exists based on a tolerance threshold. Such analysis can include, for example, comparing the weight of the item to its expected weight. This can help determine if a mistake has been made or fraud has been committed. Whether an inconsistency is observed is based on a tolerance threshold. For example, a certain deviation (i.e., the tolerance threshold) from an expected value could be allowed to could be allowed to reduce errors during the checkout process and thereby increase checkout throughput. As such, the present invention allows the tolerance threshold to be tuned based on a desired checkout throughput.

Growing community of inventors

Lansing, MI, United States of America

Charles A Otto