Patents by Inventor Ali Mosleh
Ali Mosleh has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11922609Abstract: End to end differentiable machine vision systems, training methods, and processor-readable media are disclosed. A differentiable image signal processor (ISP) is disclosed that can be trained, using machine learning techniques, to adapt raw images received from a new sensor into an adapted images of the same type (i.e. in the same visual domain) as the images previously used to train a perception module, without fine-tuning the perception module itself. The differentiable ISP may include functional blocks for performing specific image enhancement operations using a relatively small number of learned parameters corresponding to meaningful characteristics of the image enhancement operations.Type: GrantFiled: March 17, 2021Date of Patent: March 5, 2024Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Ali Mosleh, Vahid Partovi Nia
-
Patent number: 11849212Abstract: Image Signal Processing (ISP) optimization framework for computer vision applications is disclosed. The tuning of the ISP is performed automatically and presented as a nonlinear multi-objective optimization problem, followed by solving the problem using an evolutionary stochastic solver. An improved ISP of the embodiments of the invention includes at least features of search space reduction for reducing a number of ISP configurations, remapping the generated population to the reduced search space via mirroring, and global optimization function processing, which allow tuning all the blocks of the ISP at the same time instead of the prior art tuning of each ISP block separately. Also shown that an ISP tuned for image quality performs inferior compared with an ISP trained for a specific downstream image recognition task.Type: GrantFiled: February 22, 2022Date of Patent: December 19, 2023Assignee: TORC CND ROBOTICS, INC.Inventors: Avinash Sharma, Emmanuel Luc Julien Onzon, Nicolas Joseph Paul Robidoux, Ali Mosleh
-
Publication number: 20220301123Abstract: End to end differentiable machine vision systems, training methods, and processor-readable media are disclosed. A differentiable image signal processor (ISP) is disclosed that can be trained, using machine learning techniques, to adapt raw images received from a new sensor into an adapted images of the same type (i.e. in the same visual domain) as the images previously used to train a perception module, without fine-tuning the perception module itself. The differentiable ISP may include functional blocks for performing specific image enhancement operations using a relatively small number of learned parameters corresponding to meaningful characteristics of the image enhancement operations.Type: ApplicationFiled: March 17, 2021Publication date: September 22, 2022Inventors: Ali MOSLEH, Vahid PARTOVI NIA
-
Publication number: 20220182536Abstract: Image Signal Processing (ISP) optimization framework for computer vision applications is disclosed. The tuning of the ISP is performed automatically and presented as a nonlinear multi-objective optimization problem, followed by solving the problem using an evolutionary stochastic solver. An improved ISP of the embodiments of the invention includes at least features of search space reduction for reducing a number of ISP configurations, remapping the generated population to the reduced search space via mirroring, and global optimization function processing, which allow tuning all the blocks of the ISP at the same time instead of the prior art tuning of each ISP block separately. Also shown that an ISP tuned for image quality performs inferior compared with an ISP trained for a specific downstream image recognition task.Type: ApplicationFiled: February 22, 2022Publication date: June 9, 2022Inventors: Avinash SHARMA, Emmanuel Luc Julien ONZON, Nicolas Joseph Paul ROBIDOUX, Ali MOSLEH
-
Patent number: 11283991Abstract: Image Signal Processing (ISP) optimization framework for computer vision applications is disclosed. The tuning of the ISP is performed automatically and presented as a nonlinear multi-objective optimization problem, followed by solving the problem using an evolutionary stochastic solver. An improved ISP of the embodiments of the invention includes at least features of search space reduction for reducing a number of ISP configurations, remapping the generated population to the reduced search space via mirroring, and global optimization function processing, which allow tuning all the blocks of the ISP at the same time instead of the prior art tuning of each ISP block separately. Also shown that an ISP tuned for image quality performs inferior compared with an ISP trained for a specific downstream image recognition task.Type: GrantFiled: June 4, 2020Date of Patent: March 22, 2022Assignee: ALGOLUX INC.Inventors: Avinash Sharma, Emmanuel Luc Julien Onzon, Nicolas Joseph Paul Robidoux, Ali Mosleh
-
Publication number: 20200389588Abstract: Image Signal Processing (ISP) optimization framework for computer vision applications is disclosed. The tuning of the ISP is performed automatically and presented as a nonlinear multi-objective optimization problem, followed by solving the problem using an evolutionary stochastic solver. An improved ISP of the embodiments of the invention includes at least features of search space reduction for reducing a number of ISP configurations, remapping the generated population to the reduced search space via mirroring, and global optimization function processing, which allow tuning all the blocks of the ISP at the same time instead of the prior art tuning of each ISP block separately. Also shown that an ISP tuned for image quality performs inferior compared with an ISP trained for a specific downstream image recognition task.Type: ApplicationFiled: June 4, 2020Publication date: December 10, 2020Inventors: Avinash Sharma, Emmanuel Luc Julien Onzon, Nicolas Joseph Paul Robidoux, Ali Mosleh
-
Patent number: 9800866Abstract: A method for determining a point spread function (PSF) of a camera, comprising: displaying on a display at least two images comprising visual features, elements having a different optical intensity, and a test pattern; acquiring the images using the camera; determining a coordinate transformation using the image comprising the visual features and its corresponding acquired image, the coordinate transformation for aligning together a coordinate system of the camera and a coordinate system of the display; aligning the acquired image comprising the elements having a different optical intensity using the determined coordinate system and determining a difference of intensity between the displayed image comprising the elements having a different optical intensity and the corresponding acquired image; aligning the acquired image comprising the test pattern and correcting an intensity of the acquired image comprising the test pattern, thereby obtaining a corrected image; and estimating a PSF using the corrected imageType: GrantFiled: April 7, 2016Date of Patent: October 24, 2017Assignee: Algolux Inc.Inventors: Paul Green, Ali Mosleh, Emmanuel Onzon
-
Publication number: 20160301922Abstract: A method for determining a point spread function (PSF) of a camera, comprising: displaying on a display at least two images comprising visual features, elements having a different optical intensity, and a test pattern; acquiring the images using the camera; determining a coordinate transformation using the image comprising the visual features and its corresponding acquired image, the coordinate transformation for aligning together a coordinate system of the camera and a coordinate system of the display; aligning the acquired image comprising the elements having a different optical intensity using the determined coordinate system and determining a difference of intensity between the displayed image comprising the elements having a different optical intensity and the corresponding acquired image; aligning the acquired image comprising the test pattern and correcting an intensity of the acquired image comprising the test pattern, thereby obtaining a corrected image; and estimating a PSF using the corrected imageType: ApplicationFiled: April 7, 2016Publication date: October 13, 2016Inventors: Paul Green, Ali Mosleh, Emmanuel Onzon
-
Patent number: 8443089Abstract: A method and apparatus for providing an integrated communication services (ICS) network that permits data subscribers to share information on a hospital campus using wireless or wireline local area networks, or to seamlessly tap into the facility's central database while in transit away from the hospital. The central database, referred to as the integrated communications clearinghouse (ICC) is adapted to store a communication profile for every network server on the ICS network. The ICC also maintains a Global Patient Profile (GPP), consisting of a patient profile for every patient on the network. The GPP provides the ICC with the capability of providing aggregated data and demographic information, creating “live” public health statistics through a single query to a central server. The actual selection of a particular channel to access the ICS will be service-dependent and a function of user whereabouts, desired quality of the connection, and costs of the connection.Type: GrantFiled: February 28, 2011Date of Patent: May 14, 2013Assignee: Verizon Laboratories Inc.Inventors: Ali Mosleh, Richard A. Stanley
-
Patent number: 8301191Abstract: A method and apparatus is disclosed for providing an integrated communication services (ICS) network that permits wireless and wireline data subscribers to share information on a hospital campus using wireless or wireline local area networks, or to seamlessly tap into the facility's central database while in transit away from the hospital. Users interface with the ICS network through the use of a point of care terminal that resembles a clipboard and accepts data entry from a stylus on an touch screen. The central database, referred to as the integrated communications clearinghouse (ICC) is adapted to store a communication profile for every network server on the ICS network. The ICC also maintains a Global Patient Profile (GPP), consisting of a patient profile for every patient ever added to the network.Type: GrantFiled: December 20, 2006Date of Patent: October 30, 2012Assignee: Verizon Laboratories Inc.Inventors: Ali Mosleh, Richard A. Stanley
-
Publication number: 20110208773Abstract: A method and apparatus is disclosed for providing an integrated communication services (ICS) network that permits wireless and wireline data subscribers to share information on a hospital campus using wireless or wireline local area networks, or to seamlessly tap into the facility's central database while in transit away from the hospital. Users interface with the ICS network through the use of a point of care terminal that resembles a clipboard and accepts data entry from a stylus on an touch screen. The central database, referred to as the integrated communications clearinghouse (ICC) is adapted to store a communication profile for every network server on the ICS network. The ICC also maintains a Global Patient Profile (GPP), consisting of a patient profile for every patient ever added to the network.Type: ApplicationFiled: February 28, 2011Publication date: August 25, 2011Inventors: Ali Mosleh, Richard A. Stanley
-
Patent number: 7899910Abstract: A method and system of accessing an information system using a portable access device is provided. The method includes identifying a communication profile associated with a first network server, attempting, by the portable access device, to establish a communication link between the portable access device and the first network server using communication channel that is selected by the portable access device based on the communication profile, a location of the portable access device with respect to the first network server, a desired quality of the communication link, and a cost of the communication link, attempting to establish the communication link includes using at least a portion of the description information, and capturing data received by the portable access device in a memory located in the portable access device in accordance with a failed attempt to establish the communication link.Type: GrantFiled: August 30, 1999Date of Patent: March 1, 2011Assignee: Verizon Laboratories Inc.Inventors: Ali Mosleh, Richard A. Stanley
-
Patent number: 7774293Abstract: A hybrid causal framework applies properties of probabilistic models, such as Bayesian belief networks, to causal logic models, such as fault trees and event sequence diagrams. The probabilistic model establishes a joint probability distribution of causal relationships between events and conditions in the logic models. The probability of the events and conditions are found by propagating probabilities from the probabilistic model through the logic models.Type: GrantFiled: March 17, 2006Date of Patent: August 10, 2010Assignee: University of MarylandInventors: Ali Mosleh, Chengdong Wang, Franciscus J. Groen
-
Publication number: 20080167068Abstract: A method and apparatus is disclosed for providing an integrated communication services (ICS) network that permits wireless and wireline data subscribers to share information on a hospital campus using wireless or wireline local area networks, or to seamlessly tap into the facility's central database while in transit away from the hospital. Users interface with the ICS network through the use of a point of care terminal that resembles a clipboard and accepts data entry from a stylus on an touch screen. The central database, referred to as the integrated communications clearinghouse (ICC) is adapted to store a communication profile for every network server on the ICS network. The ICC also maintains a Global Patient Profile (GPP), consisting of a patient profile for every patient ever added to the network.Type: ApplicationFiled: December 20, 2006Publication date: July 10, 2008Applicant: VERIZON LABORATORIES INC.Inventors: Ali Mosleh, Richard A. Stanley
-
Publication number: 20070011113Abstract: A hybrid causal framework applies properties of probabilistic models, such as Bayesian belief networks, to causal logic models, such as fault trees and event sequence diagrams. The probabilistic model establishes a joint probability distribution of causal relationships between events and conditions in the logic models. The probability of the events and conditions are found by propagating probabilities from the probabilistic model through the logic models.Type: ApplicationFiled: March 17, 2006Publication date: January 11, 2007Inventors: Ali Mosleh, Chengdong Wang, Franciscus Groen
-
Patent number: 6820044Abstract: A computer readable medium includes one or more sequences of one or more instructions for execution by one or more processors. The instructions when executed by the one or more processors cause the one or more processors to automatically insert common cause basic events into multiple tree structures in a risk model stored in a computer memory. The multiple tree structures are based on a single definition of the common cause failure group. The processors also replace regular basic events within the multiple structures with logigates including all common cause basic events associated with the common cause failure group.Type: GrantFiled: October 9, 2001Date of Patent: November 16, 2004Assignee: University of MarylandInventors: Franciscus J. Groen, Carol S. Smidts, Ali Mosleh
-
Publication number: 20030070108Abstract: A computer readable medium includes one or more sequences of one or more instructions for execution by one or more processors. The instructions when executed by the one or more processors cause the one or more processors to automatically insert common cause basic events into multiple tree structures in a risk model stored in a computer memory. The multiple tree structures are based on a single definition of the common cause failure group. The processors also replace regular basic events within the multiple structures with logigates including all common cause basic events associated with the common cause failure group.Type: ApplicationFiled: October 9, 2001Publication date: April 10, 2003Inventors: Franciscus J. Groen, Carol S. Smidts, Ali Mosleh
-
Patent number: 6223143Abstract: A quantitative risk assessment system (QRAS) builds a risk model of a system for which risk of failure is being assessed, then analyzes the risk of the system corresponding to the risk model. The QRAS performs sensitivity analysis of the risk model by altering fundamental components and quantifications built into the risk model, then re-analyzes the risk of the system using the modifications. More particularly, the risk model is built by building a hierarchy, creating a mission timeline, quantifying failure modes, and building/editing event sequence diagrams. Multiplicities, dependencies, and redundancies of the system are included in the risk model. For analysis runs, a fixed baseline is first constructed and stored. This baseline contains the lowest level scenarios, preserved in event tree structure. The analysis runs, at any level of the hierarchy and below, access this baseline for risk quantitative computation as well as ranking of particular risks.Type: GrantFiled: August 31, 1998Date of Patent: April 24, 2001Assignee: The United States Government as represented by the Administrator of the National Aeronautics and Space AdministrationInventors: Robert M Weinstock, Carol S Smidts, Ali Mosleh, Yung-Hsien Chang, Sankaran Swaminathan, Francisco J Groen, Zhibin Tan