Object detection is a fundamental problem faced in remote sensing images analysis. Most of object detection methods mainly focus on single-source image and utilize single spectral or spatial information. Therefore, they are easily affected by illumination angle, brightness and the structure similar to the object. To overcome these defects, a novel object detection framework is proposed using superpixel segmentation and multisource features in multispectral and panchromatic images. During multisource feature extraction stage, the local region spectral information and the spatial information are extracted from multispectral and panchromatic patches respectively. Then, we embed these spectral features into spatial features to construct the new multisource features. During the detection stage, superpixel segmentation method is applied to extract candidate patches based on the superpixel centers from multisource images, which makes detection more efficient. Then, multisource features are also extracted from these candidate patches, which are input to SVM for detection. Experiments are implemented using two groups of the panchromatic and multispectral images by WorldView 2. The results indicated that, compared with single-source detection result, the proposed method can effectively improve the detection performance both on precision and recall rate.
Classification of real-world remote sensing images is a challenging task because of complex spectral–spatial information with high-dimensional feature vectors. Most of the traditional classification approaches directly treat data as vectors, which usually results in a small sample size problem and abundant redundant information; thus, they inevitably degrade the performance of the classifier. To overcome the drawbacks, we take advantage of the benefits of local scatters and tensor representation and propose a framework for hyperspectral image (HSI) classification through combining local tensor discriminant analysis (LTDA) with spectral–spatial feature extraction. First, we use a well-known spectral–spatial feature extraction approach to extract abundant spectral–spatial features as feature tensors. Then, based on class label information, LTDA is used to eliminate redundant information and to extract discriminant feature tensors for the subsequent classification. Two real HSIs are used as experimental datasets. The obtained results indicate that the proposed method exhibits good performance, while using a small number of training samples.