skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Python and computer vision

Abstract

This paper discusses the use of Python in a computer vision (CV) project. We begin by providing background information on the specific approach to CV employed by the project. This includes a brief discussion of Constrained Delaunay Triangulation (CDT), the Chordal Axis Transform (CAT), shape feature extraction and syntactic characterization, and normalization of strings representing objects. (The terms 'object' and 'blob' are used interchangeably, both referring to an entity extracted from an image.) The rest of the paper focuses on the use of Python in three critical areas: (1) interactions with a MySQL database, (2) rapid prototyping of algorithms, and (3) gluing together all components of the project including existing C and C++ modules. For (l), we provide a schema definition and discuss how the various tables interact to represent objects in the database as tree structures. (2) focuses on an algorithm to create a hierarchical representation of an object, given its string representation, and an algorithm to match unknown objects against objects in a database. And finally, (3) discusses the use of Boost Python to interact with the pre-existing C and C++ code that creates the CDTs and CATS, performs shape feature extraction and syntactic characterization, and normalizes objectmore » strings. The paper concludes with a vision of the future use of Python for the CV project.« less

Authors:
 [1];
  1. (Justin E.)
Publication Date:
Research Org.:
Los Alamos National Laboratory
Sponsoring Org.:
USDOE
OSTI Identifier:
975972
Report Number(s):
LA-UR-02-0393; LA-UR-02-393
TRN: US201018%%1057
Resource Type:
Conference
Resource Relation:
Conference: Submitted to: Tenth International Python Conference, Alexandria, VA, February 3-7, 2002.
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICAL METHODS AND COMPUTING; ALGORITHMS; COMPUTERS; SHAPE; VISION

Citation Formats

Doak, J. E., and Prasad, Lakshman. Python and computer vision. United States: N. p., 2002. Web.
Doak, J. E., & Prasad, Lakshman. Python and computer vision. United States.
Doak, J. E., and Prasad, Lakshman. Tue . "Python and computer vision". United States. doi:. https://www.osti.gov/servlets/purl/975972.
@article{osti_975972,
title = {Python and computer vision},
author = {Doak, J. E. and Prasad, Lakshman},
abstractNote = {This paper discusses the use of Python in a computer vision (CV) project. We begin by providing background information on the specific approach to CV employed by the project. This includes a brief discussion of Constrained Delaunay Triangulation (CDT), the Chordal Axis Transform (CAT), shape feature extraction and syntactic characterization, and normalization of strings representing objects. (The terms 'object' and 'blob' are used interchangeably, both referring to an entity extracted from an image.) The rest of the paper focuses on the use of Python in three critical areas: (1) interactions with a MySQL database, (2) rapid prototyping of algorithms, and (3) gluing together all components of the project including existing C and C++ modules. For (l), we provide a schema definition and discuss how the various tables interact to represent objects in the database as tree structures. (2) focuses on an algorithm to create a hierarchical representation of an object, given its string representation, and an algorithm to match unknown objects against objects in a database. And finally, (3) discusses the use of Boost Python to interact with the pre-existing C and C++ code that creates the CDTs and CATS, performs shape feature extraction and syntactic characterization, and normalizes object strings. The paper concludes with a vision of the future use of Python for the CV project.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Jan 01 00:00:00 EST 2002},
month = {Tue Jan 01 00:00:00 EST 2002}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Topics addressed include active vision for intelligent robots, 3D vision methods, tracking in robotic and vision, visual servoing and egomotion in robotics, egomotion and time-sequential processing, and control and planning in robotics and vision. Particular attention is given to invariant in visual motion, generic target tracking using color, recognizing 3D articulated-line-drawing objects, range data acquisition from an encoded structured light pattern, and 3D edge orientation detection. Also discussed are acquisition of randomly moving objects by visual guidance, fundamental principles of robot vision, high-performance visual servoing for robot end-point control, a long sequence analysis of human motion using eigenvector decomposition, andmore » sequential computer algorithms for printed circuit board inspection.« less
  • This book presents the papers given at a conference on image processing and pattern recognition. Topics considered at the conference included the shape of three-dimensional objects, algorithms, map data, optimization, the segmentation of images, data analysis, computerized simulation, parallel processing, computer graphics, robot vision, parallel processors, satellite imagery, signal-to-noise ratio, vectors, image analysis in petrography, and real-time image processing.
  • This book presents the papers given at a conference on image processing and pattern recognition. Topics considered at the conference included stereovision, vision applications, parallel processing, algorithms, artificial intelligence, structure from motion, three-dimensional shape representation, array processors, industrial vision systems, computer architecture, homogeneous multiprocessors, shape from texture, and optimal likelihood generators for edge detection under Gaussian additive noise.
  • Various papers on intelligent robots and computer vision are presented. The general topics addressed include: pattern recognition, boundary description in pattern recognition, image processing, segmentation and feature extraction, color vision and multisensor processing, neural nets, image understanding, fuzzy logic in image processing and computer vision, robot vision software, parallel computer vision architectures, three-dimensional object recognition, and intelligent control of robot systems.
  • The algorithmic attributes of a representative sample of vision algorithms are discussed in the context of computer architecture. Major building blocks available to the designer of special purpose computers for computer vision are discussed. In addition, probable future improvements to these building blocks and new building blocks that might become available are also discussed. A proposal for combining these building blocks into a special purpose computer architecture for computer vision is presented. This architecture is a very straightforward multiprocessor design which can evolve in performance as improved versions of the common building blocks become available. 39 references.