Saliency and Gist Features for Target Detection in Satellite Images
Saliency and Gist Features for Target Detection in Satellite Images
复制标题
DOI：
10.1109/tip.2010.2099128
复制
发表时间：
2011-07
影响因子：
10.6
通讯作者：
Zhicheng Li;L. Itti
中科院分区：
文献类型：
--
作者：
Zhicheng Li;L. Itti
随着图像采集能力的快速增长，可靠地检测大范围头顶或卫星图像中的物体已成为日益紧迫的需求。在存在较大类内变异的情况下，这个问题尤其困难，例如，寻找“船”或“建筑物”，基于模型的方法往往会失败，因为无法为高度可变的目标定义良好的模型或模板。本文探索了一种在高分辨率广域卫星图像中检测和分类目标的自动方法，该方法依赖于根据一组受生物学启发的低级视觉特征检测目标的统计特征。大区域图像被切成小图像芯片，以两种互补的方式进行分析：“注意力/显着性”分析利用局部特征及其跨空间的相互作用，而“主旨”分析则侧重于全局非空间特征及其统计数据。两个特征集都用于使用支持向量机将每个芯片分类为包含目标或不包含目标。进行了四个实验来寻找“船”（实验 1 和 2）、“建筑物”（实验 3）和“飞机”（实验 4）。实验1中，将14 416个图像芯片随机分为训练集（300个船，300个非船）和测试集（13 816），并对测试集进行分类（ROC面积：0.977±0.003）。在实验2中，对来自另一幅广域图像的11 385个芯片的另一个测试集进行分类，保持与实验1中相同的训练集（ROC面积：0.952±0.006）。实验3中，从108 885个芯片中随机选取600个训练芯片（每种类型300个）进行分类（ROC面积：0.922±0.005）。实验4中，随机选择20个训练芯片（每种类型10个）对剩余2581个芯片进行分类（ROC面积：0.976±0.003）。在所有四个实验中，所提出的算法都优于最先进的 SIFT、HMAX 和隐藏尺度显着结构方法以及之前的仅要点特征。这项研究表明，所提出的目标搜索方法可以可靠有效地检测大型图像数据集中的高度可变的目标对象。
Reliably detecting objects in broad-area overhead or satellite images has become an increasingly pressing need, as the capabilities for image acquisition are growing rapidly. The problem is particularly difficult in the presence of large intraclass variability, e.g., finding “boats” or “buildings,” where model-based approaches tend to fail because no good model or template can be defined for the highly variable targets. This paper explores an automatic approach to detect and classify targets in high-resolution broad-area satellite images, which relies on detecting statistical signatures of targets, in terms of a set of biologically-inspired low-level visual features. Broad-area images are cut into small image chips, analyzed in two complementary ways: “attention/saliency” analysis exploits local features and their interactions across space, while “gist” analysis focuses on global nonspatial features and their statistics. Both feature sets are used to classify each chip as containing target(s) or not, using a support vector machine. Four experiments were performed to find “boats” (Experiments 1 and 2), “buildings” (Experiment 3) and “airplanes” (Experiment 4). In experiment 1, 14 416 image chips were randomly divided into training (300 boat, 300 nonboat) and test sets (13 816), and classification was performed on the test set (ROC area: 0.977 ±0.003). In experiment 2, classification was performed on another test set of 11 385 chips from another broad-area image, keeping the same training set as in experiment 1 (ROC area: 0.952 ±0.006). In experiment 3, 600 training chips (300 for each type) were randomly selected from 108 885 chips, and classification was conducted (ROC area: 0.922 ±0.005). In experiment 4, 20 training chips (10 for each type) were randomly selected to classify the remaining 2581 chips (ROC area: 0.976 ±0.003). The proposed algorithm outperformed the state-of-the-art SIFT, HMAX, and hidden-scale salient structure methods, and previous gist-only features in all four experiments. This study shows that the proposed target search method can reliably and effectively detect highly variable target objects in large image datasets.