Repository logo
 
Publication

PROCESSING OF MICROSCOPY IMAGES IN THE COMPRESSED DOMAIN

datacite.subject.fosEngenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
dc.contributor.advisorThomaz, Lucas Arrabal
dc.contributor.advisorTávora, Luís Miguel de Oliveira Pegado de Noronha e
dc.contributor.advisorGrilo, Carlos Fernando de Almeida
dc.contributor.advisorMiragaia, Rolando Lúcio Germano
dc.contributor.authorVasconcellos, Nicolas David Freire
dc.date.accessioned2025-09-02T11:17:11Z
dc.date.available2025-09-02T11:17:11Z
dc.date.issued2025-03-19
dc.description.abstractThis work presents a novel approach to the detection of mitochondria in microscopy images within the compressed domain, eliminating the need for full image decompression while maintaining high detection accuracy. The proposed architecture, Processing of Microscopy Images in the Compressed Domain (ProMIC), integrates a Domain Translator architecture, enabling object detection networks to operate directly on images’ latent representations. This approach significantly reduces computational costs while preserving essential visual information for detection tasks. The study also included a preliminary analysis of JPEG-AI compression efficiency against classical codecs (HEVC/H.265 and VVC/H.266) across multiple datasets. The proposed method builds upon a Base model inspired by a previous compresseddomain object detection approach. Key innovations in this work include: the use of a Lightweight Residual Block that enhances feature extraction from latent representations, improving detection robustness across different compression rates; a Pixel Domain Fine-Tuning process that enhances the Domain Adaptation of the model; and a Guided Domain Translator Training strategy that refines the Domain Translator to extract meaningful information directly from latent codes, minimising the gap between compressed and pixel-domain object detection models. Extensive experiments were conducted to compare the performance of both the Base model and the proposed architecture against a reference object detection model that operates on reconstructed images. Results on the Lucchi++ dataset show that ProMIC significantly outperforms the Base model. Notably, the proposed architecture exceeds the reference model in mean average precision, achieving a 2.31 percentage point improvement at the highest bitrate. This highlights the potential of compressed-domain image processing to capture meaningful features that may be lost in fully reconstructed images. Additionally, applying the ProMIC method reduced the complexity of the object detection system by 42.34% compared to the reference model. These findings validate the feasibility of direct object detection in latent codes, offering a computationally efficient alternative to traditional pixel-domain approaches. The proposed framework has promising applications in real-time processing, embedded systems, and large-scale AI-driven image analysis, paving the way for more efficient and scalable vision models that operate directly in the compressed domain.por
dc.identifier.tid203996577
dc.identifier.urihttp://hdl.handle.net/10400.8/13949
dc.language.isoeng
dc.relationCompression of Multimodal Biomedical Images using Neural Networks
dc.relationInstituto de Telecomunicações
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectEngenharia eletrónica
dc.subjectImagem
dc.subjectComputação
dc.subjectImagem por computação
dc.subjectImagem de Microscopia no domínio comprimido
dc.titlePROCESSING OF MICROSCOPY IMAGES IN THE COMPRESSED DOMAIN
dc.typemaster thesis
dspace.entity.typePublication
oaire.awardTitleCompression of Multimodal Biomedical Images using Neural Networks
oaire.awardTitleInstituto de Telecomunicações
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/3599-PPCDT/2022.09914.PTDC/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F50008%2F2020/PT
oaire.fundingStream3599-PPCDT
oaire.fundingStream6817 - DCRRNI ID
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
relation.isProjectOfPublication5fe22f63-2c55-4683-bae7-89d5dba06107
relation.isProjectOfPublication0836c6a6-afd0-499e-8a16-612dd27ec1dc
relation.isProjectOfPublication.latestForDiscovery5fe22f63-2c55-4683-bae7-89d5dba06107
thesis.degree.nameMaster in Electrical Engineering

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Dissertação_MEE_Nicolas_Vasconcellos_c f.pdf
Size:
7.8 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.32 KB
Format:
Item-specific license agreed upon to submission
Description: