Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 15 of 15 for “"Binarisation"”.
-
Extracting generic text information from images
… system. TIE consists of text detection, text binarisation and text recognition. For different applications or projects, one or more of these three TIE components may be embedded. Although many efforts have been made to extract text from images and videos, this problem is far from being solved …
-
Recherche de l’information dans les réseaux de neurones convolutifs pré-entraînés
… sortie du réseau sont moins précises (98%). La binarisation améliore la précision de la classification par sexe (InceptionV3 avec binarisation, AUC=0,981) par rapport à (InceptionV3 sans binarisation, AUC=0,966), ce qui est supérieur à la SIFT 2D (AUC=0,926). Pour les cas de cerveau humain …
-
Mikroskopijos vaizdų analizė /
… is compared. A comparison of two commonly used binarisation algorithms, Otsu and Sauvola, shows that Sauvola’s algorithm performs better, while image averaging-based blurring is the most optimal when using Sauvola’s algorithm. An additional visual review of the results identified cases where the …
-
Évaluation d'une technique de compensation du mouvement des artères coronaires à partir d'une séquence angiographique
… suivi par des filtres de débruitage et de binarisation. Finalement, la transformée de distance modifiée est calculée. Un algorithme est aussi présenté afin de segmenter les bifurcations des artères coronaires tout en minimisant le bruit. Les particules sont échantillonnées en favorisant …
-
Visual feature learning with application to medical image classification
… (including gLTP) lose information due to the binarisation step involved in their construction. Therefore, I then propose a descriptor called the Extended Multi-Resolution Local Patterns (xMRLP), which is real-valued and reduces information loss. I propose unsupervised and weakly-supervised …
-
Automatic Gender Binary Beliefs and their Role in Gender Inequality
… Chapter Three provided strong evidence that the binarisation of gender underpins discrimination in occupational contexts, effects in the remaining chapters were comparably weaker. Lastly, this thesis took the novel step of gathering a sufficiently large IRAP dataset for a set of pooled analyses. …
-
Décomposition autonome et interprétable d’images multispectrales de documents par apprentissage contraint
… encre/fond de +29.5 points F-score contre la binarisation de Howe et dépasse ACE v2 de +1.32 points (état-de-l’art). De plus, pour décomposition d’images MS non-supervisée, PRISM reste jusqu’à 7.4× plus rapide que VBONMF, la meilleure approche NMF concurrente. Des tests sur des scènes …
-
Automatic Analysis of Holographic Interferograms
… analysis system was devised. This involved the binarisation of the fringe field, followed by detection and tracing of the fringe edges. The extracted fringe co-ordinates are then ordered using a semi-automatic fringe ordering system. In this way a map of the complete density distribution over …
-
Scattering Properties of Suspended Particles
… an apparent particle break-up due to incorrect binarisation. Despite this, the holographic method provides a unique and powerful mechanism that enables images of particles to be analysed within the context of their in-situ environment.
-
Automatic detection and identification of cells in digital images of day 2 IVF embryos
… to this approach was made. This involved binarisation of the images and templates, and the creation of two new edge-detection algorithms, one of which was based on the convolution technique while the other was based on the difference of the grey level between the border of the cell and its …
-
Visual unsupervised deep learning model design for historical document image analysis
Historical documents are one of the most crucial influences that drive scientific and historical development. Some historical documents are present and can be used through classical models to be analyzed. Other documents do not meet the quality and minimum visibility required by information …
-
Visual image processing in various representation spaces for documentary heritage preservation
This thesis establishes an advanced image processing framework for the enhancement and restoration of historical document images (HDI) in both intensity (gray-scale or color) and multispectral (MS) representation spaces. It provides three major contributions: 1) the binarization of gray-scale HDI; …
-
From edges to pages: boundary-aware binarization and two-stage reconstruction of historical documents
Historical documents often suffer from severe degradations such as bleed-through, stains, fading, and physical losses, which compromise both human readability and machine analysis. Historical document restoration must therefore recover legible text and faithful backgrounds while ensuring structural …