Semantic Annotation of Satellite Images
نویسندگان
چکیده
We describe here a method to annotate satellite images with semantic concepts. The concepts used for the annotation are defined by the user. For each concept, the user provides a set of example images which will be used for learning. The method uses a first step of image processing which computes SIFT descriptors in the image. These feature vectors are quantized using an information theoretic criterion to define the optimal number of clusters. Then, a probabilistic modelization of the generation of this discrete collection of low-level features is set up. Each concept is associated to a mixture of models whoses parameters are learned by an Expectation-Maximization algorithm. The optimal complexity of the mixture is computed using a Minimum Description Length criterion. Finally, given a new image, a greedy algorithm is used to provide a segmentation whose regions are annotated with a concept. The performances of the system are evaluated on Quickbird images at 70cm resolution for a set of high-level concepts.
منابع مشابه
Automatic semantic network construction for multi-layer annotation of satellite images Construction de réseaux sémantiques pour l’annotation multi-couches d’images satellitaires
A novel method is presented for annotating satellite images. The labels used for annotation are given by a user with a set of example images. A learning step is then applied to learn the model. The originality of the method is to formulate the problem of semantic annotation to a further extent than a mere probabilistic classification task. The method takes into account the semantical relationsh...
متن کاملSemantic-Based Image Retrial in the VQ Compressed Domain using Image Annotation Statistical Models
متن کامل
Automatic Semantic Network Construction for Multi-Layer Annotation of Satellite Images
A novel method is presented for annotating satellite images. The labels used for annotation are given by a user with a set of example images. A learning step is then applied to learn the model. The originality of the method is to formulate the problem of semantic annotation to a further extent than a mere probabilistic classification task. The method takes into account the semantical relationsh...
متن کاملScalable Image Annotation by Summarizing Training Samples into Labeled Prototypes
By increasing the number of images, it is essential to provide fast search methods and intelligent filtering of images. To handle images in large datasets, some relevant tags are assigned to each image to for describing its content. Automatic Image Annotation (AIA) aims to automatically assign a group of keywords to an image based on visual content of the image. AIA frameworks have two main sta...
متن کاملTile-Level Annotation of Satellite Images Using Multi-Level Max-Margin Discriminative Random Field
This paper proposes a multi-level max-margin discriminative analysis (MDA) framework, which takes both coarse and fine semantics into consideration, for the annotation of high-resolution satellite images. In order to generate more discriminative topic-level features, the MDA uses the maximum entropy discrimination latent Dirichlet Allocation (MedLDA) model. Moreover, for improving the spatial c...
متن کاملA CAD System Framework for the Automatic Diagnosis and Annotation of Histological and Bone Marrow Images
Due to ever increasing of medical images data in the world’s medical centers and recent developments in hardware and technology of medical imaging, necessity of medical data software analysis is needed. Equipping medical science with intelligent tools in diagnosis and treatment of illnesses has resulted in reduction of physicians’ errors and physical and financial damages. In this article we pr...
متن کامل