Please use this identifier to cite or link to this item:
Title: Super-precision colour CCD camera imaging through machine learning
Authors: Zhang, Ziyue
Keywords: Engineering::Electrical and electronic engineering
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Zhang, Z. (2022). Super-precision colour CCD camera imaging through machine learning. Final Year Project (FYP), Nanyang Technological University, Singapore.
Project: P2056-202
Abstract: Spectral imaging is an advanced technology that can be applied in many fields. Laser is a light source for spectral imaging and it can be captured by most CCD cameras. However, CCD is unable to read and analyse the wavelength of the light without additional instruments. Pictures captured by CCD can only be displayed in RGB, which is the most common used colour space nowadays. Artificial intelligence is booming these days. Machine learning is an useful tool for the analysis of spectral imaging. In this project, a data set of laser images were collected to train the machine learning models. RGB values, wavelength and luminous intensity of all images in the data set were extracted and analysed. A few machine learning methods include classification algorithms and convolutional neural network (CNN) were used to learn the relationship between RGB and wavelength of the images. The machine learning models is used to precision the wavelength of selected regions in laser image. Additional laser images including single-peak wavelength laser spots and single-peak wavelength laser spots were collected to verify prediction accuracy of the models in practical application.
Schools: School of Electrical and Electronic Engineering 
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP final report Zhang Ziyue.pdf
  Restricted Access
4.01 MBAdobe PDFView/Open

Page view(s)

Updated on Sep 26, 2023


Updated on Sep 26, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.