Electrical & Computer Engineering, Department of

 

First Advisor

Michael Hoffman

Second Advisor

Sina Balkir

Date of this Version

7-2018

Citation

D. Hong. Implementation of Vision Processing Tasks on a Smart Camera Platform Using Neighborhood Processors. Master’s thesis, University of Nebraska-Lincoln, Lincoln, Nebraska, July 2018.

Comments

A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Electrical Engineering, Under the Supervision of Professors Michael Ho man and Sina Balkir. Lincoln, Nebraska : July, 2018.

Copyright (c) 2018 Dingyi Hong

Abstract

Two applications of vision processing on a smart camera platform using neighborhood processors are presented. A smart camera application of real-time trajectory calculation for tracking objects and predicting their path of motion by using a single intelligent camera or a dual camera system are designed, simulated and implemented. Another application of object recognition with a key points algorithm for detecting simple shaped objects and locating the camera position within a 3-D coordinate system via imaging by using artificial neural network is applied to provide quick and accurate processing on a neighborhood-level parallel processor or other limited vision chip. For all applications an emphasis has been placed on simplifying algorithmic requirements to allow implementation on processors with limited computational resources.

Advisors: Michael Ho man and Sina Balkir

Share

COinS