Preprint Article Version 1 NOT YET PEER-REVIEWED

High Resolution Satellite Image Classification Using Multi-Task Joint Sparse and Low-Rank Representation

Kunlun Qi 1,2 , Wenxuan Liu 3,4 , Chao Yang 1,2 , Qingfeng Guan 1,2,* , Huayi Wu 3,4
  1. National Engineering Research Center of Geographic Information System, China University of Geosciences (Wuhan), Wuhan 430074, China
  2. Faculty of Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, China
  3. State Key Laboratory for Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan 430079, China
  4. Collaborative Innovation Center of Geospatial Technology, Wuhan University, Wuhan 430079, China
Version 1 : Received: 6 November 2016 / Approved: 7 November 2016 / Online: 7 November 2016 (05:25:11 CET)

How to cite: Qi, K.; Liu, W.; Yang, C.; Guan, Q.; Wu, H. High Resolution Satellite Image Classification Using Multi-Task Joint Sparse and Low-Rank Representation. Preprints 2016, 2016110036 (doi: 10.20944/preprints201611.0036.v1). Qi, K.; Liu, W.; Yang, C.; Guan, Q.; Wu, H. High Resolution Satellite Image Classification Using Multi-Task Joint Sparse and Low-Rank Representation. Preprints 2016, 2016110036 (doi: 10.20944/preprints201611.0036.v1).

Abstract

Scene classification plays an important role in the intelligent processing of high-resolution satellite (HRS) remotely sensed image. In HRS image classification, multiple features, e.g. shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS image. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structure across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.

Subject Areas

multi-task learning; feature fusion; sparse representation; low-rank representation; scene classification

Readers' Comments and Ratings (0)

Discuss and rate this article
Views 102
Downloads 107
Comments 0
Metrics 0
Discuss and rate this article

×
Alerts
Notify me about updates to this article or when a peer-reviewed version is published.