Spatial pyramid matching has demonstrated its power for image recognition task by pooling features from spatially increasingly fine sub-regions. Motivated by the concept of feature pooling at multiple pyramid levels, we propose a novel spectral-spatial hyperspectral image classification approach using superpixel-based spatial pyramid representation. This technique first generates multiple superpixel maps by decreasing the superpixel number gradually along with the increased spatial regions for labelled samples. By using every superpixel map, sparse representation of pixels within every spatial region is then computed through local max pooling. Finally, features learned from training samples are aggregated and trained by a support vector machine (SVM) classifier. The proposed spectral-spatial hyperspectral image classification technique has been evaluated on two public hyperspectral datasets, including the Indian Pines image containing 16 different agricultural scene categories with a 20m resolution acquired by AVIRIS and the University of Pavia image containing 9 land-use categories with a 1.3m spatial resolution acquired by the ROSIS-03 sensor. Experimental results show significantly improved performance compared with the state-of-the-art works. The major contributions of this proposed technique include (1) a new spectral-spatial classification approach to generate feature representation for hyperspectral image, (2) a complementary yet effective feature pooling approach, i.e. the superpixel-based spatial pyramid representation that is used for the spatial correlation study, (3) evaluation on two public hyperspectral image datasets with superior image classification performance.
Processing and analysing large volume of remote sensing data is both labour intensive and time consuming. Therefore, there is a need to effectively and efficiently identify meaningful regions in these remote sensing data for timely resource management. In this paper, we propose a visual attention model for identifying regions-of-interest in remote sensing data. The proposed model incorporates both bottom-up spatial saliency and top-down objectness, by fusing a co-occurrence histogram saliency model with the BING objectness model. The co-occurrence histogram saliency model is constructed by first building a 2D co-occurrence histogram that captures co-occurrence and occurrence of image intensities, and then using the 2D co-occurrence histogram to model local and global saliency. On the other hand, the BING objectness model is constructed by resizing image intensities in variable-sized windows to 8x8 windows, and then using the norms of the gradients in the 8x8 windows as features to train a generic objectness measure. Our experimental results show that the proposed model can effectively and efficiently identify regions-of-interest in remote sensing data. The proposed model may be applied in various remote sensing applications such as anomaly detection, urban area detection, target detection, or land use classification.
With the rapid development of various satellite sensors, automatic and advanced scene classification technique is urgently needed to process a huge amount of satellite image data. Recently, a few of research works start to implant the sparse coding for feature learning in aerial scene classification. However, these previous research works use the single-layer sparse coding in their system and their performances are highly related with multiple low-level features, such as scale-invariant feature transform (SIFT) and saliency. Motivated by the importance of feature learning through multiple layers, we propose a new unsupervised feature learning approach for scene classification on very high resolution satellite imagery. The proposed unsupervised feature learning utilizes multipath sparse coding architecture in order to capture multiple aspects of discriminative structures within complex satellite scene images. In addition, the dense low-level features are extracted from the raw satellite data by using different image patches with varying size at different layers, and this approach is not limited to a particularly designed feature descriptors compared with the other related works. The proposed technique has been evaluated on two challenging high-resolution datasets, including the UC Merced dataset containing 21 different aerial scene categories with a 1 foot resolution and the Singapore dataset containing 5 land-use categories with a 0.5m spatial resolution. Experimental results show that it outperforms the state-of-the-art that uses the single-layer sparse coding. The major contributions of this proposed technique include (1) a new unsupervised feature learning approach to generate feature representation for very high-resolution satellite imagery, (2) the first multipath sparse coding that is used for scene classification in very high-resolution satellite imagery, (3) a simple low-level feature descriptor instead of many particularly designed low-level descriptor, such as SIFT descriptors and saliency, (4) evaluation on two satellite image datasets that come from different sensor sources.
Rooftop extraction from satellite/aerial imagery is an important geospatial problem with many practical applications. However, rooftop extraction remains a challenging problem due to the diverse characteristics and appearances of the buildings, as well as the quality of the satellite/aerial images. Many existing rooftop extraction methods use rooftop corners as a basic component. Nonetheless, existing rooftop corner detectors either suffer from high missed detection or introduce high false alarm. Based on the observation that rooftop corners are typically of L-shape, we propose an L-shaped corner detector for automatic rooftop extraction from high resolution satellite/aerial imagery. The proposed detector considers information in a spatial circle around each pixel to construct a feature map which captures the probability of L-shaped corner at every pixel. Our experimental results on a rooftop database of over 200 buildings demonstrate its effectiveness for detecting rooftop corners. Furthermore, our proposed detector is complementary to many existing rooftop extraction approaches which require reliable rooftop corners as their inputs. For instance, it can be used in the quadrilateral footprint extraction methods or in driving level-set-based segmentation techniques.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.