OAK

Design and implementation of functionalized artificial vision systems inspired by animal eyes

Metadata Downloads
Abstract
Modern cameras have transitioned from standalone systems to artificial vision, supporting the decision-making for next behaviors in machines or robots. This transition aims to emulate the functionalities of biological eyes, which process tremendous amounts of visual information economically for species survival in unpredictable natural habitats. Consequently, this has brought further requirements for artificial vision systems, including high compatibility with other electronics, lightweight and compact design, high-quality imaging under dynamic environments, low energy consumption, and highly efficient image pre-processing. However, traditional vision systems based on planar image sensors suffer from multiple and complex optical configurations, which eventually decrease the efficiency of the overall visual process. In this thesis, functionalized artificial vision systems tailored for the specific needs of machine vision and mobile robotics are explored. First, a high-contrast imaging system under uneven light conditions inspired by the cuttlefish-eye is explored with technical approaches developed in terms of shaped apertures and optical filters. Next, an artificial compound eye system for depth-sensing applications is explored; optics integration and imaging techniques are elaborated to construct a fully integrated camera module. Furthermore, potential application fields and remaining challenges in bio-inspired artificial vision systems will be discussed.
Author(s)
Sehui Chang
Issued Date
2024
Type
Thesis
URI
https://scholar.gist.ac.kr/handle/local/19079
공개 및 라이선스
  • 공개 구분공개
파일 목록
  • 관련 파일이 존재하지 않습니다.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.