基于DBNMI模型的海洋遥感影像自动标注方法

doi:10.3969/j.issn.0253-2778.2017.07.001

中国科学技术大学学报 ›› 2017, Vol. 47 ›› Issue (7): 541-546.DOI: 10.3969/j.issn.0253-2778.2017.07.001

• 论著 • 下一篇

基于DBNMI模型的海洋遥感影像自动标注方法

黄冬梅，许琼琼，杜艳玲，贺琪

上海海洋大学信息学院，上海 201306）

收稿日期:2016-08-28 修回日期:2016-12-08 出版日期:2017-07-31 发布日期:2017-07-31
通讯作者: 贺琪
作者简介:黄冬梅，女，1964年生，博士/教授，研究方向：大数据和智能信息处理. E-mail：dmhuang@shou.edu.cn
基金资助:
国家重点基础研究发展计划(2012CB316206), 国家自然科学基金(61272098，61402282)资助.

Ocean remote sensing image auto-annotation based on DBNMI model

HUANG Dongmei, XU Qiongqiong, DU Yanling, HE Qi

College of Information and Technology, Shanghai Ocean University, Shanghai 201306, China

Received:2016-08-28 Revised:2016-12-08 Online:2017-07-31 Published:2017-07-31

摘要/Abstract

摘要： 研究大规模海洋遥感影像管理的关键是缩小影像低层视觉特征与高层语义之间的鸿沟.针对海洋遥感影像中不同区域对语义相似性度量的贡献程度不同，提出一种基于深度信念网络多示例（deep belief networks multi-instance, DBNMI）的遥感影像语义自动标注模型.模型对初始输入遥感影像进行自适应分割，粗粒度划分海洋遥感影像背景区域和对象区域；对影像对象区域的低层视觉特征和高层语义概念间关系，利用深度信念网络模型进行自动建模；定量计算标注词间共现和对立的语义关系，改善图像标注结果.在公开遥感影像数据集上进行验证，实验表明所提出方法在标注精度上取得了较好效果.

关键词: 深度信念网络, 自适应分割, 海洋遥感影像, 图像标注

Abstract: Bridge the semantic gap between low-level visual feature and high-level semantic concepts has been the subject of intensive investigation on large scale remote sensing image management for years in order to improve the accuracy of automatic image annotation. An ocean remote sensing image auto-annotation method based on DBNMI model was proposed for contributions of semantic similarity about different regions of ocean remote sensing images. Initial remote sensing images were adaptively segmented, ocean remote sensing images were divided into background and the object region by means of a coarse-grained method, the relationship between low-level visual feature and high-level semantics label of the object region was modeled automatically, using DBN model, and the co-occurrence relations and adversarial relations between semantic concepts for improving image annotation results were calculated. The proposed approach is evaluated on a public remote sensing image dataset. The experimental results show a satisfactory improvement on accuracy.

Key words: deep belief networks, adaptive segmentation, ocean remote sensing image, image annotation

中图分类号:

TP 391

黄冬梅，许琼琼，杜艳玲，贺琪. 基于DBNMI模型的海洋遥感影像自动标注方法[J]. 中国科学技术大学学报, 2017, 47(7): 541-546.

HUANG Dongmei, XU Qiongqiong, DU Yanling, HE Qi. Ocean remote sensing image auto-annotation based on DBNMI model[J]. Journal of University of Science and Technology of China, 2017, 47(7): 541-546.

参考文献

［1］
BRATASANU D, NEDELCU I, DATCU M. Bridging the semantic gap for satellite image annotation and automatic mapping applications[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011, 4(1): 193-204.
[2] 郑歆慰,胡岩峰,孙显,等. 基于空间约束多特征联合稀疏编码的遥感图像标注方法研究[J]. 电子与信息学报, 2014, 36(8): 1891-1898.
[3] LINOU M, MATRE H, DATCU M. Semantic annotation of satellite images using latent dirichlet allocation[J]. Geoscience and Remote Sensing Letters, 2010, 7(1): 28-32.
[4] LUO W, LI H L, LIU G H. Automatic annotation of multispectral satellite images using author-topic model[J]. Geoscience and Remote Sensing Letters, 2012, 9(4): 634-638.
[5] LUO W, LI H L, LIU G H, et al. Semantic annotation of satellite images using author-genre-topic model[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(2): 1356-1368.
[6] CHEN K M, JIAN P, ZHOU Z X, et al. Semantic annotation of high-resolution remote sensing images via Gaussian process multi-instance multilabel learning[J]. Geoscience and Remote Sensing Letters, 2013, 10(6): 1285-1289.
[7] WEI Y C, XIA W, HUANG J S, et al. CNN: Single-label to multi-label[J]. Computer Science, 2014: arXiv:1406.5726.
[8] 杨阳, 张文生. 基于深度学习的图像自动标注算法[J]. 数据采集与处理, 2015, 30(1): 88-98.
[9] 吕启, 窦勇, 牛新,等. 基于DBN模型的遥感图像分类[J]. 计算机研究与发展, 2014, 51(9): 1911-1918.
[10] 高常鑫, 桑农. 基于深度学习的高分辨率遥感影像目标检测[J]. 测绘通报, 2014, (S1):108-111.
[11] 黄志坚,黎湘,徐帆江. 基于视觉复杂度的自适应尺度遥感影像分割[J]. 电子与信息学报, 2013, 35(8): 1786-1792.
[12] 葛永,吴秀清,洪日昌. 基于多示例学习的遥感图像检索[J]. 中国科学技术大学学报, 2009, 39(2): 132-136.
GE Y, WU X Q, HONG R C. Remote sensing image retrieval based on multiple instance learning[J]. Journal of University of Science and Technology of China, 2009, 39(2): 132-136.

[13] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[14] YANG Y, NEWSAM S. Bag-of-visual-words and spatial extensions for land-use classification[C]// Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems. San Jose, USA: ACM Press, 2010: 270-279.
[15] CHANG C C, LIN C J. LIBSVM: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): 1-27.

基于DBNMI模型的海洋遥感影像自动标注方法

Ocean remote sensing image auto-annotation based on DBNMI model

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价