Skip to Main content Skip to Navigation

Évaluation de l'interprétation d'images

Baptiste Hemery 1
1 Equipe Monétique & Biométrie - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : Image processing algorithms include a set of methods that process the image from its acquisition by a sensor to the extraction of useful information for a given application. Among these image interpretation algorithms, some are designed to detect, localize and identify one or more objects in an image. The problem addressed in this thesis is the evaluation of interpretation results of an image or a video, given the associated ground truth. Challenges are multiple such as the comparison of algorithms, evaluation of an algorithm during its development or its optimal setting. We propose in this thesis a formalization of desired properties for a localization metric. We make a rigorous comparative study of localization metric of the state of the art in view of these properties. We carry out similar work on recognition methods using a local representation of objects in order to quantify a recognition error. We have developed a method for evaluating an interpretation result of an image making use of the lessons from these studies. The advantage of the proposed method is to evaluate an image interpretation result by taking into account at the same time the quality of the localization, recognition and detection of objects of interest in the image. The behavior of this evaluation method has been tested on a database and is interesting. Several parameters allow us to change the behavior of this method for a given application.
Document type :
Complete list of metadatas
Contributor : Baptiste Hemery <>
Submitted on : Wednesday, April 6, 2011 - 2:53:10 PM
Last modification on : Tuesday, February 5, 2019 - 12:12:41 PM
Long-term archiving on: : Thursday, July 7, 2011 - 2:52:56 AM


  • HAL Id : tel-00583733, version 1


Baptiste Hemery. Évaluation de l'interprétation d'images. Interface homme-machine [cs.HC]. Université de Caen, 2009. Français. ⟨tel-00583733⟩



Record views


Files downloads