SAR-Based Terrain Classification Using Weakly Supervised Hierarchical Markov Aspect Models

Abstract

We introduce the hierarchical Markov aspect model (HMAM), a computationally efficient graphical model for densely labeling large remote sensing images with their underlying terrain classes. HMAM resolves local ambiguities efficiently by combining the benefits of quadtree representations and aspect models-the former incorporate multiscale visual features and hierarchical smoothing to provide improved local label consistency, while the latter sharpen the labelings by focusing them on the classes that are most relevant for the broader local image context. The full HMAM model takes a grid of local hierarchical Markov quadtrees over image patches and augments it by incorporating a probabilistic latent semantic analysis aspect model over a larger local image tile at each level of the quadtree forest. Bag-of-word visual features are extracted for each level and patch, and given these, the parent-child transition probabilities from the quadtree and the label probabilities from the tile-level aspect models, an efficient forwards-backwards inference pass allows local posteriors for the class labels to be obtained for each patch. Variational expectation-maximization is then used to train the complete model from either pixel-level or tile-keyword-level labelings. Experiments on a complete TerraSAR-X synthetic aperture radar terrain map with pixel-level ground truth show that HMAM is both accurate and efficient, providing significantly better results than comparable single-scale aspect models with only a modest increase in training and test complexity. Keyword-level training greatly reduces the cost of providing training data with little loss of accuracy relative to pixel-level training.

Tile-Level Annotation of Satellite Images Using Multi-Level Max-Margin Discriminative Random Field

“…The most related work to ours is detailed in [16], where the authors only exploit different types of feature representation and do not make full use of the contextual information that may be beneficial to annotation tasks. Some other related studies [1,5,9] have investigated the application of topic model in satellite images annotation task. These studies did not apply multi-level features into classification framework [5] and introduced spatial information by means of cutting large image into small patches with an overlap and [9] employed Markov random field for the sake of utilizing the contextual information in satellite images.…”

“…Some other related studies [1,5,9] have investigated the application of topic model in satellite images annotation task. These studies did not apply multi-level features into classification framework [5] and introduced spatial information by means of cutting large image into small patches with an overlap and [9] employed Markov random field for the sake of utilizing the contextual information in satellite images. However we suggest that the CRF model is more suitable for discriminant tasks like image annotation or scene classification.…”

“…In another method, the random field models such as the Markov Random Field (MRF), Conditional Random Field (CRF) have been employed for improving the spatial coherence of aspect models as well. Particularly, for the sake of describing the spatial relationship of latent topics, an MRF prior has been defined over hidden topic labels, which has been obtained by PLSA, and the experimental results of supervised and weakly supervised manner have demonstrated that the segmentation and recognition accuracy is obviously enhanced by the two complementary models detailed in [8,9].…”

Change Detection Using High Resolution Remote Sensing Images Based on Active Learning and Markov Random Fields

“…To integrate the contextual information and attribute information, we formulate the change detection problem in a Bayesian framework [46], wherein the detection process is usually conducted by maximizing the posterior distribution as follows:…”

Land use and land cover classification for rural residential areas in China using soft-probability cascading of multifeatures

“…[3][4][5][6][7] In processing high-resolution remote sensing images, numerous classification algorithms, such as the object-oriented approach, [8][9][10] based on the classification of a support vector machine (SVM) [11][12][13] and Markov random fields (MRF) [14][15][16][17][18] are being developed. Local features [19][20][21][22][23] have been successfully applied to image retrieval, semantic segmentation, and scene understanding.…”