Full metadata record
DC FieldValueLanguage
dc.contributorFaculty of Engineeringen_US
dc.contributor.advisorChi, Zheru (EIE)-
dc.creatorSun, Guoqing-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/10095-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic University-
dc.rightsAll rights reserveden_US
dc.titlePlant recognition through deep learningen_US
dcterms.abstractPlant is an important contributor of earth ecosystem. It is not only the largest source of oxygen but also the daily food of other creatures, therefore, plant research is extremely valuable for human. To study and protect flora diversities, pant classification is a primary step. However, classification of plants is very challenging, because different types of plants sometimes present similarity and the plants vary in different time. In this dissertation, I propose to classify the plant based on a multi-organ approach. The plant organs considered includes flower, leaf, fruit, and stem. Additional data provide more information that might help to distinguish between species, but the variability in shape and appearance in plant organs raises the degree of complexity of the problem when I train the network model by multi-organ images straightly. In order to tackle this problem, I use a CNN-based method to operate on one image from a single organ, simultaneously capturing one or more organs of the same plant species. Then the three combination methods, namely the majority principle (MAV), the majority appearance principle and the majority principle (MAP&MAV), and the majority principle of sum score for different organs (S-MAV), are used to classify plants. Experimental results show that by utilizing the mentioned techniques. The proposed multi-organ method is better than single organ in this work.en_US
dcterms.extentiv, 40 pages : color illustrationsen_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2019en_US
dcterms.educationalLevelM.Sc.en_US
dcterms.educationalLevelAll Masteren_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertationsen_US
dcterms.LCSHPlants -- Classificationen_US
dcterms.LCSHImage processingen_US
dcterms.LCSHMachine learningen_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
991022270856503411.pdfFor All Users (off-campus access for PolyU Staff & Students only)1.26 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/10095