Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ThesisVorhies2020.pdf (19.33 MB)

ETD Abstract Container

Abstract Header

Low-complexity Algorithms for Light Field Image Processing

Vorhies, John T
http://orcid.org/0000-0003-4879-747X
http://rave.ohiolink.edu/etdc/view?acc_num=akron1590771210097321

Abstract Details

2020, Master of Science in Engineering, University of Akron, Electrical Engineering.
Light fields present a new paradigm for capturing scene structure. While conventional 2-D images fail to convey depth, this information is inherent in 4-D light field images. Computer vision systems in the areas of robotics and autonomous vehicles can make use of this depth information for decision making processes. A camera gantry is designed and used for capturing light field images. Camera calibration techniques are explored and it is shown that calibration is required for determining the frequency domain region of support for a given depth. This thesis focuses on the use of light fields for extracting objects in a scene by their depth. A 4-D linear filter, presented in [Dansereau 2003, Dansereau 2007], is used for depth filtering, providing a constant runtime solution that is of low computational complexity. A novel method for determining the frequency content associated with a given depth is devised, and implemented in digital hardware. This method is capable of tuning a depth filter in real time as an object moves to different depths in a scene. Due to the large amount of data generated by light fields, data compression is required to effectively store and transmit light field images. This thesis explores 4-D light field compression, using the Karhunen-Loeve transform (KLT). The 4-D KLT is applied to light field images and compression performance is discussed.
Kye-shin Lee (Advisor)
Arjuna Madanayake (Committee Member)
Nghi Tran (Committee Member)
Alexander Hoover (Committee Member)
88 p.
Electrical Engineering
Light Fields, Image Processing, Depth Filters, Multi-dimensional Signal Processing

Recommended Citations

Citations

  • Vorhies, J. T. (2020). Low-complexity Algorithms for Light Field Image Processing [Master's thesis, University of Akron]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=akron1590771210097321

    APA Style (7th edition)

  • Vorhies, John. Low-complexity Algorithms for Light Field Image Processing. 2020. University of Akron, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=akron1590771210097321.

    MLA Style (8th edition)

  • Vorhies, John. "Low-complexity Algorithms for Light Field Image Processing." Master's thesis, University of Akron, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=akron1590771210097321

    Chicago Manual of Style (17th edition)

akron1590771210097321
74
© 2020, all rights reserved.
This open access ETD is published by University of Akron and OhioLINK.