No Cover Image

E-Thesis 939 views 494 downloads

Adaptive Learning for Segmentation and Detection / JINGJING DENG

Swansea University Author: JINGJING DENG

DOI (Published version): 10.23889/SUthesis.36297

Abstract

Segmentation and detection are two fundamental problems in computer vision and medical image analysis, they are intrinsically interlinked by the nature of machine learning based classification, especially supervised learning methods. Many automatic segmentation methods have been proposed which heavi...

Full description

Published: 2017
Institution: Swansea University
Degree level: Doctoral
Degree name: Ph.D
URI: https://cronfa.swan.ac.uk/Record/cronfa36297
first_indexed 2017-10-26T19:06:15Z
last_indexed 2020-09-02T02:57:58Z
id cronfa36297
recordtype RisThesis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2020-09-01T15:39:41.6046793</datestamp><bib-version>v2</bib-version><id>36297</id><entry>2017-10-26</entry><title>Adaptive Learning for Segmentation and Detection</title><swanseaauthors><author><sid>e4218e3d46c58fcc78db2ada87eacf63</sid><firstname>JINGJING</firstname><surname>DENG</surname><name>JINGJING DENG</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2017-10-26</date><abstract>Segmentation and detection are two fundamental problems in computer vision and medical image analysis, they are intrinsically interlinked by the nature of machine learning based classification, especially supervised learning methods. Many automatic segmentation methods have been proposed which heavily rely on hand-crafted discriminative features for specific geometry and powerful classifier for delinearating the foreground object and background region. The aimof this thesis is to investigate the adaptive schemes that can be used to derive efficient interactive segmentation methods for medical imaging applications, and adaptive detection methods for addressing generic computer vision problems. In this thesis, we consider adaptive learning as a progressive learning process that gradually builds the model given sequential supervision from user interactions. The learning process could be either adaptive re-training for smallscale models and datasets or adaptive fine-tuning for medium-large scale. In addition, adaptive learning is considered as a progressive learning process that gradually subdivides a big and difficult problem into a set of smaller but easier problems, where a final solution can be found via combining individual solvers consecutively. We first show that when discriminative features are readily available, the adaptive learning scheme can lead to an efficient interactive method for segmenting the coronary artery, where promising segmentation results can be achieved with limited user intervention. We then present a more general interactive segmentation method that integrates a CNN based cascade classifier and a parametric implicit shape representation. The features are self-learnt during the supervised training process, no hand-crafting is required.Then, the segmentation can be obtained via imposing a piecewise constant constraint to thedetection result through the proposed shape representation using region based deformation.Finally, we show the adaptive learning scheme can also be used to address the face detection problem in an unconstrained environment, where two CNN based cascade detectors are proposed. Qualitative and quantitative evaluations of proposed methods are reported, and show theefficiency of adaptive schemes for addressing segmentation and detection problems in general. Continues...</abstract><type>E-Thesis</type><journal/><publisher/><keywords>Segmentation, Image Analysis, Computer Science</keywords><publishedDay>31</publishedDay><publishedMonth>12</publishedMonth><publishedYear>2017</publishedYear><publishedDate>2017-12-31</publishedDate><doi>10.23889/SUthesis.36297</doi><url/><notes>A selection of third party content is redacted or is partially redacted from this thesis.Pages: 26, 31, 33, 35, 36, 38, 110, 111, 112, 113, 124, 126.</notes><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><degreelevel>Doctoral</degreelevel><degreename>Ph.D</degreename><apcterm/><lastEdited>2020-09-01T15:39:41.6046793</lastEdited><Created>2017-10-26T16:31:21.1688296</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>JINGJING</firstname><surname>DENG</surname><order>1</order></author></authors><documents><document><filename>0036297-26102017163355.pdf</filename><originalFilename>Deng_Jingjing_phd_thesis.redacted_content.pdf</originalFilename><uploaded>2017-10-26T16:33:55.7370000</uploaded><type>Output</type><contentLength>44404080</contentLength><contentType>application/pdf</contentType><version>Redacted version - open access</version><cronfaStatus>true</cronfaStatus><embargoDate>2017-10-26T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2020-09-01T15:39:41.6046793 v2 36297 2017-10-26 Adaptive Learning for Segmentation and Detection e4218e3d46c58fcc78db2ada87eacf63 JINGJING DENG JINGJING DENG true false 2017-10-26 Segmentation and detection are two fundamental problems in computer vision and medical image analysis, they are intrinsically interlinked by the nature of machine learning based classification, especially supervised learning methods. Many automatic segmentation methods have been proposed which heavily rely on hand-crafted discriminative features for specific geometry and powerful classifier for delinearating the foreground object and background region. The aimof this thesis is to investigate the adaptive schemes that can be used to derive efficient interactive segmentation methods for medical imaging applications, and adaptive detection methods for addressing generic computer vision problems. In this thesis, we consider adaptive learning as a progressive learning process that gradually builds the model given sequential supervision from user interactions. The learning process could be either adaptive re-training for smallscale models and datasets or adaptive fine-tuning for medium-large scale. In addition, adaptive learning is considered as a progressive learning process that gradually subdivides a big and difficult problem into a set of smaller but easier problems, where a final solution can be found via combining individual solvers consecutively. We first show that when discriminative features are readily available, the adaptive learning scheme can lead to an efficient interactive method for segmenting the coronary artery, where promising segmentation results can be achieved with limited user intervention. We then present a more general interactive segmentation method that integrates a CNN based cascade classifier and a parametric implicit shape representation. The features are self-learnt during the supervised training process, no hand-crafting is required.Then, the segmentation can be obtained via imposing a piecewise constant constraint to thedetection result through the proposed shape representation using region based deformation.Finally, we show the adaptive learning scheme can also be used to address the face detection problem in an unconstrained environment, where two CNN based cascade detectors are proposed. Qualitative and quantitative evaluations of proposed methods are reported, and show theefficiency of adaptive schemes for addressing segmentation and detection problems in general. Continues... E-Thesis Segmentation, Image Analysis, Computer Science 31 12 2017 2017-12-31 10.23889/SUthesis.36297 A selection of third party content is redacted or is partially redacted from this thesis.Pages: 26, 31, 33, 35, 36, 38, 110, 111, 112, 113, 124, 126. COLLEGE NANME COLLEGE CODE Swansea University Doctoral Ph.D 2020-09-01T15:39:41.6046793 2017-10-26T16:31:21.1688296 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science JINGJING DENG 1 0036297-26102017163355.pdf Deng_Jingjing_phd_thesis.redacted_content.pdf 2017-10-26T16:33:55.7370000 Output 44404080 application/pdf Redacted version - open access true 2017-10-26T00:00:00.0000000 true
title Adaptive Learning for Segmentation and Detection
spellingShingle Adaptive Learning for Segmentation and Detection
JINGJING DENG
title_short Adaptive Learning for Segmentation and Detection
title_full Adaptive Learning for Segmentation and Detection
title_fullStr Adaptive Learning for Segmentation and Detection
title_full_unstemmed Adaptive Learning for Segmentation and Detection
title_sort Adaptive Learning for Segmentation and Detection
author_id_str_mv e4218e3d46c58fcc78db2ada87eacf63
author_id_fullname_str_mv e4218e3d46c58fcc78db2ada87eacf63_***_JINGJING DENG
author JINGJING DENG
author2 JINGJING DENG
format E-Thesis
publishDate 2017
institution Swansea University
doi_str_mv 10.23889/SUthesis.36297
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Segmentation and detection are two fundamental problems in computer vision and medical image analysis, they are intrinsically interlinked by the nature of machine learning based classification, especially supervised learning methods. Many automatic segmentation methods have been proposed which heavily rely on hand-crafted discriminative features for specific geometry and powerful classifier for delinearating the foreground object and background region. The aimof this thesis is to investigate the adaptive schemes that can be used to derive efficient interactive segmentation methods for medical imaging applications, and adaptive detection methods for addressing generic computer vision problems. In this thesis, we consider adaptive learning as a progressive learning process that gradually builds the model given sequential supervision from user interactions. The learning process could be either adaptive re-training for smallscale models and datasets or adaptive fine-tuning for medium-large scale. In addition, adaptive learning is considered as a progressive learning process that gradually subdivides a big and difficult problem into a set of smaller but easier problems, where a final solution can be found via combining individual solvers consecutively. We first show that when discriminative features are readily available, the adaptive learning scheme can lead to an efficient interactive method for segmenting the coronary artery, where promising segmentation results can be achieved with limited user intervention. We then present a more general interactive segmentation method that integrates a CNN based cascade classifier and a parametric implicit shape representation. The features are self-learnt during the supervised training process, no hand-crafting is required.Then, the segmentation can be obtained via imposing a piecewise constant constraint to thedetection result through the proposed shape representation using region based deformation.Finally, we show the adaptive learning scheme can also be used to address the face detection problem in an unconstrained environment, where two CNN based cascade detectors are proposed. Qualitative and quantitative evaluations of proposed methods are reported, and show theefficiency of adaptive schemes for addressing segmentation and detection problems in general. Continues...
published_date 2017-12-31T07:07:28Z
_version_ 1827277109528100864
score 11.05492

Similar Items