Mihai Datcu

German Aerospace Center (DLR), Köln, North Rhine-Westphalia, Germany

Are you Mihai Datcu?

Claim your profile

Publications (336)93.48 Total impact

  • Nazli Deniz Cagatay, Mihai Datcu
    IEEE International Conference on Image Processing (ICIP), Paris, France; 10/2014
  • Nazli Deniz Cagatay, Mihai Datcu
    ICIP; 10/2014
  • Dusan Gleich, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents an information extraction and image enhancement technique using single-look complex (SLC) synthetic aperture radar data. The novelty of this method is the proposed complex-domain despeckling stage. Tikhonov-like optimization is used for minimizing the cost function, which consists of a Gauss-Markov random field (GMRF) prior. The GMRF model is used for texture modeling. The texture parameters of the GMRF are estimated using the evidence maximization framework. The experimental results showed that despeckled SLC images have well-preserved textural features, structures, and point scatterers. The phase of the reconstructed image is well preserved and provides good-quality interferograms of high-resolution spotlight images.
    IEEE Transactions on Geoscience and Remote Sensing 07/2014; 52(8). · 3.47 Impact Factor
  • Nazli Deniz Cagatay, Mihai Datcu
    EUSAR, Berlin; 06/2014
  • Nazli Deniz Cagatay, Mihai Datcu
    European Conference on Synthetic Aperture Radar (EUSAR), Berlin, Germany; 06/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we focus on the extraction and analysis of long-term satellite image time series, and on applications in urban development monitoring. A Satellite Image Time Series (SITS) comprised of more than 100 data sets, covering a time span of over 25 years is extracted from the current Landsat data archives, Then we create stacks of intra-annual images, which are analyzed by applying a series of direct multidate classifications in order to extract the annual built-up surface. The evolution of the built-up areas is then compared against population dynamics for the studied area.
    2014 10th International Conference on Communications (COMM); 05/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this letter, we carry out a comparative study of statistical models for multilook synthetic aperture radar amplitude images. Ten state-of-the-art statistical models are selected for comparison. To achieve a fair evaluation, we estimate all model parameters using the method of log-cumulants and apply the method to an image pyramid with varying pixel spacing (and resolution). The pyramid is created by different image product generation options. In addition to pixel spacing and resolution, we also consider the homogeneity of a scene for performance evaluation and we apply three performance measures. Through this study, it was found out that some models perform well for all resolutions, while the performance of other models depends heavily on the image content.
    IEEE Geoscience and Remote Sensing Letters 04/2014; 11(10):1752-1756. · 1.82 Impact Factor
  • Mihai Datcu
    Copernicus Big Data workshop; 04/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Information extraction from SAR images can be based on Gibbs random fields and wavelet-based methods
    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 01/2014; · 2.87 Impact Factor
  • Corneliu Dumitru, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: The amount of collected Earth Observation (EO) data is increasing immensely with a rate of several Terabytes of data per day. Simultaneously with this increasing of data, new trends for exploration and information retrieval are highly needed. In the last year, the proposed method tries to explore the EO data using Image Information Mining (IIM) approach in which primitive feature extraction and classification are the main steps, developing a new process chain and a new taxonomy for the retrieved categories, mainly based on human interaction, can be a good solution. This paper proposes to explore the content of images and to identify the number of objects and land cover categories that can be retrieved from high resolution TerraSAR-X data. We need to mention that is for the first time when for remote sensing a large data set (e.g., TerraSAR-X images) covering different cities over the world is annotated and for each category a taxonomy for high resolution data is defined. Applications that may result from this study can be a semantic catalogue for TerraSAR-X, urban crisis, disasters, etc. First, we strongly need an automatic or semi-automatic searching tool capable to find in large EO data set similar sub-images (i.e. patches) and to group them in categories. Secondary, we need to define a taxonomy that can be used to semantically annotate each category using the human interaction. The data set consists of 109 scenes that cover different areas over the world: 5 scenes from Africa, 27 scenes from Asia, 44 scenes from Europe, 11 scenes from Middle East, and 22 scenes from North and South of America. These scenes are grouped in collections based on three criteria in order: (1) to get an idea about how many categories can be retrieved for each city/country/continent, (2) to see whether the same urban categories belong to two different scenes, and (3) to help us to annotate large data. Data set is grouped in more collections using previous three criteria and each collection is process separately using an enhanced methodology that take the scene/scenes and tile them into patches. Gabor filters is used as a primitive features method and is applied to each patch. Support vector machine with relevance feedback is implemented in order to group the features in categories of relevance. Finally, these categories are semantically annotated using as ground truth the Google Earth. In our investigation more than 850 categories were retrieved with their specific taxonomy. The novelty of this paper lies in the fact that this is the first time when a semantic annotation was made on a large number of scenes containing high resolution synthetic aperture radar images. This investigation has an important impact in applications such as classification of urban areas, infrastructure (e.g., airport, port, etc.), geography images (e.g., mountains, etc.), industrial sites, military sites, vegetation, and agriculture. The proposed taxonomies can be the basis for building a semantic catalogue for EO images. Finally, four types of query are defined and are planned to be integrated into the new system developed at DLR. The query provides opportunities to EO users to search into the database for some specific parameters or semantic of the existing data set.
    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 01/2014; · 2.87 Impact Factor
  • Shiyong Cui, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent years have witnessed a fruitful development of image representation. The most prevalent one is the Bag-of-Words (BoW) method, which gives state-of-the-art performance in many applications. It has four steps: local feature extraction, dictionary learning, feature coding, and feature pooling. In this paper, we focus on feature coding. On the basis of an analysis of currently popular feature coding methods, we propose a supervised incremental coding method. The most different characteristic of this method is that coding of a new image relies on the coding of the previous image from the same class. Therefore, we need to know the label of one image before coding. This point can be argued as a drawback of this method. However, we demonstrate that it could give much better feature for image classification. This finding gives some hints about further development of feature coding method. We believe that the entire class should be considered when coding the local features.
    ESA-EUSC-JRC 2014; 01/2014
  • Shiyong Cui, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: Changes in image time series have to be quantified by information similarity measures.
    IEEE Transactions on Geoscience and Remote Sensing 01/2014; · 3.47 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes to perform authorship analysis using the Fast Compression Distance (FCD), a similarity measure based on compression with dictionaries directly extracted from the written texts. The FCD computes a similarity between two documents through an effective binary search on the intersection set between the two related dictionaries. In the reported experiments the proposed method is applied to documents which are heterogeneous in style, written in five different languages and coming from different historical periods. Results are comparable to the state of the art and outperform traditional compression-based methods.
    Pattern Recognition Letters 01/2014; · 1.27 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Advances in the image retrieval (IR) field have contributed to the elaboration of tools for interactive exploration and extraction of the images from huge archives associating the content of the images with semantic meaning. This paper presents an Earth-observation (EO) IR system based on enriched metadata, semantic annotations, and image content called EO retrieval. EO retrieval generates an EO-data model by using automatic feature extraction, processing the EO product metadata, and defining semantics, which later is fully exploited for supporting complex queries. In order to demonstrate the functionality of the system, we have created a semantic catalog of TerraSAR-X as application scenario. The database is composed of 39 high-resolution TerraSAR-X scenes comprising about 50 000 image patches (160 times 160 pixels) with their feature descriptors, 100 of metadata entries for each scene, and about 330 semantic annotations. Many query examples combining semantics, metadata, and image content for full exploitation of the image database are presented.
    IEEE Transactions on Geoscience and Remote Sensing 11/2013; 51(2013-11-11):5145-5159. · 3.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: As the data acquisition capabilities of Earth Observation (EO) satellites have been improved significantly, a large amount of high resolution images are downlinked continuously to ground stations. The data volume increases rapidly beyond the users' capability to access the information content of the data. Thus, interactive systems that allow fast indexing of high resolution images based on image content are urgently needed. In this paper, we present an interactive learning system for semantic annotation and content mining at patch level. It mainly comprises four components: primitive feature extraction including both spatial and temporal features, relevance feedback based on active learning, a Human Machine Communication (HMC) interface, and data visualization. To overcome the shortage of training samples and to speed up the convergence, active learning is employed in this system. Two core components of active learning are the classifier training using already labeled image patches, and the sample selection strategy which selects the most informative samples for manual labeling. These two components work alternatively, significantly reducing the labeling effort and achieving fast indexing. In addition, our data visualization is particularly designed for multi-temporal and multi-sensor image indexing, where efficient visualization plays a critical role. The system is applicable to both optical and SAR images. It can index patches and it can also discover temporal patterns in satellite image time series. Three typical study cases are included to show its wide variety of use in EO applications.
    International Journal of Image and Data Fusion. 11/2013; 5(2013-11-1):1-23.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Content-Based Image Retrieval (CBIR) systems are powerful search tools in image databases that have been little applied to hyperspectral images. Relevance feedback (RF) is an iterative process that uses machine learning techniques and user's feedback to improve the CBIR systems performance. We pursued to expand previous research in hyperspectral CBIR systems built on dissimilarity functions defined either on spectral and spatial features extracted by spectral unmixing techniques, or on dictionaries extracted by dictionary-based compressors. These dissimilarity functions were not suitable for direct application in common machine learning techniques. We propose to use a RF general approach based on dissimilarity spaces which is more appropriate for the application of machine learning algorithms to the hyperspectral RF-CBIR. We validate the proposed RF method for hyperspectral CBIR systems over a real hyperspectral dataset.
    Pattern Recognition Letters 10/2013; · 1.27 Impact Factor
  • Source
    N. D. Cagatay, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work, a Markov random field based phase locked loop is proposed for phase unwrapping. The neighboring pixels are used to update the phase estimate of the centering pixel. The performance of the proposed method is evaluated for both synthetic and real interferometric phase. For terrains with relatively low slopes, the phase unwrapping is done successfully. However, in case of high fringe frequency, the method fails to unwrap the whole phase gradient. Nevertheless, the noise suppression capability of phase locked loop is remarkable.
    Proc SPIE 10/2013;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: With the advent of very high resolution (VHR) synthetic aperture radar (SAR) images, local content description is becoming a critical issue for indexing. Conventional SAR image analysis techniques, like segmentation and pixel-level classification, are likely to fail as high-level semantic description should be considered for better discrimination. Therefore, we propose to use image-patch-based analysis method for SAR image interpretation. Inspired by ratio edge detector, in this letter, a new feature extraction method represented by the mean ratios in different directions is proposed for VHR SAR image content characterization. Based on the mean ratio, two simple yet powerful and robust features are proposed for SAR image patch indexing. One is the bag-of-word model using not only the basic statistics, i.e., local mean and variance, but also the mean ratios in different directions. The second one is an adaptation of the Weber local descriptor to SAR images by substituting the gradient with the ratio of mean differences in vertical and horizontal directions. To evaluate the proposed features, image patch indexing based on active learning using a SAR image database consisting of high-resolution TerraSAR-X patches is performed. Comparison with the state-of-the-art features, particularly texture features, has shown improved performance for SAR image categorization.
    IEEE Geoscience and Remote Sensing Letters 09/2013; 10(5):1175-1179. · 1.82 Impact Factor
  • Source
    Octavian Dumitru, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: Very high resolution SAR images have a peculiar information content. Feature extraction and imaging parameters can help to desribe the information content.
    IEEE Transactions on Geoscience and Remote Sensing 07/2013; 51(2013-07-22-8):4591-4610. · 3.47 Impact Factor
  • Source
    Octavian Dumitru, Mihai Datcu
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose to identify the number of categories that can be retrieved from a very high resolution SAR data. The evaluation is done on TerraSAR-X high resolution Spotlight data and the retrieved categories are semi-automatically annotated using as feature vector the Gabor filters; as a classifier the Support Vector Machine, and for ranking the suggested images the relevance feedback. The visualization of the tool was enhanced compared with our previous implementation in order to support the users in his/her approach to search the patches of interest in a large repository. Our dataset consist in 43 scenes that cover as much as possible all the regions over the world. A total of 352 categories are identified that contain urban and non-urban categories.
    IGARSS 2013; 07/2013

Publication Stats

322 Citations
93.48 Total Impact Points

Institutions

  • 1997–2013
    • German Aerospace Center (DLR)
      • • Earth Observation Center
      • • Remote Sensing Technology Institute (IMF)
      Köln, North Rhine-Westphalia, Germany
  • 2012
    • National and Kapodistrian University of Athens
      Athínai, Attica, Greece
  • 2011
    • Polytechnic University of Bucharest
      Bucureşti, Bucureşti, Romania
  • 2009–2010
    • University of Maribor
      • Faculty of Electrical Engineering and Computer Science (FERI)
      Maribor, Mestna Obcina Maribor, Slovenia
  • 2007–2010
    • Maison de la Télédétection en Languedoc-Roussillon
      Montpelhièr, Languedoc-Roussillon, France
    • France Télécom
      Lutetia Parisorum, Île-de-France, France
  • 2008
    • University of Bucharest
      Bucureşti, Bucureşti, Romania
    • ParisTech
      Lutetia Parisorum, Île-de-France, France
  • 2005
    • Institut de Recherche en Informatique de Toulouse
      Tolosa de Llenguadoc, Midi-Pyrénées, France