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

By Fan Hu, Wen Yang, Jiayu Chen, Hong Sun. 2013

25 citations recorded by the source corpus.

This paper proposes a multi-level max-margin discriminative analysis (M3DA) 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 M3DA uses the maximum entropy discrimination latent Dirichlet Allocation (MedLDA) model. Moreover, for improving the spatial coherence of visual words neglected by M3DA, conditional random field (CRF) is employed to optimize the soft label field composed of multiple label posteriors. The framework of M3DA enables one to combine word-level features (generated by support vector machines) and topic-level features (generated by MedLDA) via the bag-of-words representation. The experimental results on high-resolution satellite images have demonstrated that, using the proposed method can not only obtain suitable semantic interpretation, but also improve the annotation performance by taking into account the multi-level semantics and the contextual information.

Subfields: Computer Vision and Pattern Recognition, Media Technology, 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

1. Junwei Han
2. Gong Cheng
3. Liangpei Zhang
4. Xiaoqiang Lu
5. Gui-Song Xia
6. Xiwen Yao
7. Yanfei Zhong
8. Lei Guo

In Media Technology

1. Gong Cheng
2. Junwei Han
3. Liangpei Zhang
4. Xiaoqiang Lu
5. Gui-Song Xia
6. Yanfei Zhong
7. Lei Guo
8. Xiwen Yao

In Remote Sensing

1. Gong Cheng
2. Junwei Han
3. Liangpei Zhang
4. Yanfei Zhong
5. Gui-Song Xia
6. Lei Guo
7. Jingwen Hu
8. Fan Hu

Citations in the corpus

• 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 [9,16,20,27,38,55].”
• Semantic Annotation of High-Resolution Satellite Images via Weakly Supervised Learning (2016, methodology): “In the topic model-based methods [13], [14], [17], [19], an image is treated as a collection of visual words, and a finite mixture of topics is used to describe the image content.”
• A Comparative Study of Sampling Analysis in the Scene Classification of Optical High-Spatial Resolution Remote Sensing Imagery (2015, background): “Observe that, in order to do scene classification for a large-scale HSR-RS image, one often requires to first construct the basic scene elements for classification, by using either non-overlapping tiles [2,5,13], overlapping tiles [10,14,16] or super-pixels [4,11].”
• A Color-Texture-Structure Descriptor for High-Resolution Satellite Image Classification (2016, background): “The second dataset contains two large HRS scenes: a large scene of Tongzhou (Scene-TZ) [56] and a large scene near the Tucson airport (Scene-AT) [57], which were both captured by the GeoEye-1 satellite sensor.”