Classifying land use from postearthquake very high-resolution (VHR) images is challenging due to the complexity of objects in Earth surface after an earthquake. Convolutional neural network (CNN) exhibits satisfied performance in differentiating complex postearthquake objects, thanks to its automatic extraction of high-level features and accurate identification of target geo-objects. Nevertheless, in view of the scale variance of natural objects, the fact that CNN suffers from the fixed receptive field, the reduced feature resolution, and the insufficient training sample has severely contributed to its limitation in the rapid damage mapping. Multiscale segmentation technique is considered as a promising solution as it can generate the homogenous regions and provide the boundary information. Therefore, we propose a combined multiscale segmentation convolutional neural network (CMSCNN) method for postearthquake VHR image classification. First, multiscale training samples are selected based on segments derived from the multiscale segmentation. Then, CNN is directly trained to classify the original image to further produce the preliminary classification maps. To enhance the localization accuracy, the output of CNN is further refined using multiscale segmentations from fine to coarse iteratively to obtain the multiscale classification maps. As a result, the combination strategy is able to capture objects and image context simultaneously. Experimental results show that the proposed CMSCNN method can reflect the multiscale information of complex scenes and obtain satisfied classification results for mapping postearthquake damage using VHR remote sensing images.
Seismic image classification is of vital importance for extracting damage information and evaluating disaster losses. With the increasing availability of high resolution remote sensing images, automatic image classification offers a unique opportunity to accommodate the rapid damage mapping requirements. However, the diversity of disaster types and the lack of uniform statistical characteristics in seismic images increase the complexity of automated image classification. This paper presents a novel automatic seismic image classification approach by integrating an adaptive spectral-textural descriptor into gravitational self-organizing map (gSOM). In this approach, seismic image is first segmented into several objects based on mean shift (MS) method. These objects are then characterized explicitly by spectral and textural feature quantization histograms. To objectify the image object delineation adapt to various disaster types, an adaptive spectral-textural descriptor is developed by integrating the histograms automatically. Subsequently, these objects as classification units are represented by neurons in a self-organizing map and clustered by adjacency gravitation. By moving the neurons around the gravitational space and merging them according to the gravitation, the object-based gSOM is able to find arbitrary shape and determine the class number automatically. Taking advantage of the diversity of gSOM results, consensus function is then conducted to discover the most suitable classification result. To confirm the validity of the presented approach, three aerial seismic images in Wenchuan covering several disaster types are utilized. The obtained quantitative and qualitative experimental results demonstrated the feasibility and accuracy of the proposed seismic image classification method.
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.