Semantic Annotation of Satellite Images Using Latent Dirichlet Allocation

By Marie Liénou, Henri Maître, Mihai Datcu. 2010

277 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Probabilistic & Bayesian Methods, Remote Sensing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Probabilistic & Bayesian Methods

In Remote Sensing

Citations in the corpus

Bag-of-Visual-Words Scene Classifier With Local and Global Features for High Spatial Resolution Remote Sensing Imagery (2016, unknown): “[18] and Bosch et al.”
Scene Classification Based on the Multifeature Fusion Probabilistic Topic Model for High Spatial Resolution Remote Sensing Imagery (2015, methodology): “In other words, the main problem in HSR image semantic interpretation is to bridge the semantic gap [34].”
The Fisher Kernel Coding Framework for High Spatial Resolution Scene Classification (2016, methodology): “In the experiments, the BOVW, LDA, and P-LDA scene classification methods employed the mean and standard deviation statistics as the low-level feature extractor, in the same way as the FK-O and FK-S scene classification methods.”
Semantic Annotation of High-Resolution Satellite Images via Weakly Supervised Learning (2016, methodology): “latent Dirichlet allocation (LDA) [19], and author topic model (ATM) [13], were proposed to annotate satellite images through representing images by mixtures of topics.”
AID: A Benchmark Data Set for Performance Evaluation of Aerial Scene Classification (2016, methodology): “In addition, to encode higher order spatial information between low-level local visual words for scene modeling, topic models along with the BoVW scheme are developed to take into account the semantic relationship among the visual words [11], [21], [28], [39], [56].”
Automatic Annotation of Satellite Images via Multifeature Joint Sparse Coding With Spatial Relation Constraint (2013, background): “We use the same settings as those in [8] for LDA and search “optimal” parameters for SRC and SVM with radial-basis-function kernels.”
Dirichlet-Derived Multiple Topic Scene Classification Model for High Spatial Resolution Remote Sensing Imagery (2016, methodology): “The scene classification methods based on the PTMs, such as PLSA and LDA, are able to reduce the dimension of the midlevel features [16], [18], [19], and they have been applied to the scene classification of HSR imagery [4], [13]–[15], [20], [22].”
Unsupervised Feature Learning Via Spectral Clustering of Multidimensional Patches for Remotely Sensed Scene Classification (2015, background): “In addition, [43] and [44] represent satellite image scenes as a finite mixture over some underlying semantic classes learnt by applying the latent Dirichlet allocation (LDA) model [45], which is a generative probabilistic model to the visual words histograms.”
Remote Sensing Image Classification: No Features, No Clustering (2015, background): “nity for image annotation [16], object classification [17], target detection [18], and land use classification [19] and it has already proven its discrimination power in image classification.”
MARTA GANs: Unsupervised Representation Learning for Remote Sensing Image Classification (2016, background): “The above-mentioned methods have comprised the state of the art for several years in the remote sensing community [3], but they are based on hand-crafted features, which are difficult, time-consuming, and require domain expertise to produce.”
Integrated visual vocabulary in latent Dirichlet allocation–based scene classification for IKONOS image (2014, methodology): “Let P 1⁄4 fpig be the topic proportion of the visual document and Q 1⁄4 fqig be the topic proportion of each semantic class; then the KLD can be computed by using the following formula: KLD 1⁄4 ðDPQþDQPÞ 2 ; (9)”
Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art (2016, background): “In the early years after 2010, based on the latent Dirichlet allocation (LDA) model [115], various supervised hierarchical feature-learning methods have been proposed in the RS community [116]–[120].”
Unsupervised Feature Learning for Aerial Scene Classification (2014, background): “In [16] simple image statistics such as the local mean and variance of pixel intensities were clustered to form the visual words.”