Unlock instant, AI-driven research and patent intelligence for your innovation.

Machine Vision with Dimensional Data Reduction

A machine vision and data technology, applied in digital video signal modification, instruments, computer components, etc., can solve problems such as utility limitations

Active Publication Date: 2021-01-22
CORNELL UNIVERSITY
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Several approaches have been proposed for developing image data preprocessing schemes for machine vision based on abstract models of the operation of the vision system, but their utility is limited

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Machine Vision with Dimensional Data Reduction
  • Machine Vision with Dimensional Data Reduction
  • Machine Vision with Dimensional Data Reduction

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0069] Example 1 - Navigation Handler

[0070] Machine Vision Applications Describes Navigators Effective in Virtual Reality Environments. However, according to one example of the present disclosure, the navigator is also effective in real world environments. Accordingly, various embodiments of the present disclosure may be used in a variety of environments and in a variety of applications.

[0071] For example, a mobile robot called Turtlebot can be used. Turtlebot is an open source personal robot designed for robotics development and testing. Turtlebot runs on the Robot Operating System (ROS), which facilitates hardware and communication mechanisms, and brings data from sensors and hardware components on the robot together into a single software framework. The robot includes a 3-wheeled circular locomotive base from a Yujin robot called iClebo Kobuki, a Microsoft Kinect sensor including a camera, and an onboard factory-calibrated gyroscope for better sensor input and stat...

example 2

[0089] Example 2 - Vision Task

[0090] As mentioned above, the methods described in machine vision applications allow machine learning algorithms to learn features of the visual world efficiently and in a generalized manner. Such methods do this by reducing the dimensionality of the visual input (eg, using retinal encoding). This application is concerned with applying one or more additional dimensionality reduction methods to the encoded data in such a way that a machine learning algorithm (such as a convolutional neural network, or CNN) when searching the parameter space (such as discovering weights in a CNN), finds a general solution instead of falling into a local solution (e.g. due to a local minimum in the parameter space).

[0091] For example, in various embodiments, the solution for a given training data set may be a set of weights that capture transformations (or calculations or mappings). Reducing the dimensionality of the training set allows the algorithm to find...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A method is described comprising receiving raw image data corresponding to a series of raw images, and processing the raw image data with an encoder of a processing device to generate encoded data. The encoder is characterized by an input / output transition that substantially mimics the input / output transition of at least one retinal cell of the vertebrate retina. The method also includes processing the encoded data by applying a dimensionality reduction algorithm to the encoded data to generate reduced dimensionality encoded data. The dimensionality reduction algorithm is configured to compress an amount of information contained in the encoded data. Also described are devices and systems that can be used with such methods.

Description

[0001] Cross References to Related Applications [0002] This application claims priority under 35 U.S.C. §119 to U.S. Provisional Patent Application No. 62 / 150,068, filed April 20, 2015, the entire contents of which are incorporated herein by reference. This application is further incorporated by reference into U.S. Provisional Application Nos. 61 / 527,493 (filed August 25, 2011), 61 / 657,406 (filed June 8, 2012), 61 / 308,681 (filed 2010 26 February), 61 / 359,188 (filed 28 June 2010), 61 / 378,793 (filed 31 August 2010), 61 / 382,280 (filed 13 September 2010) ) and No. 13 / 230,488 (filed 12 September 2011). This application further incorporates by reference International Patent Application Nos. PCT / US2011 / 026526 (filed February 28, 2011) and PCT / US2011 / 049188, referred to herein as "Prosthesis Applications" International Patent Application No. PCT / US2012 / 052348 (filed Aug. 24, 2012) (filed Aug. 25, 2011), referred to herein as "Machine Vision Application." The content of each of the...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): A61F9/08A61N5/06G06K9/46G06N3/04G06K9/00H04N19/60H04N19/85G06V10/20
CPCG06N3/008H04N19/85G06N3/049A61F9/08H04N19/60G06N3/08G06V40/103G06V40/19G06V40/193G06V10/449G06V10/454G06V10/20G06N3/045G06F18/214
Inventor S·尼伦伯格
Owner CORNELL UNIVERSITY