Abstract
With the large number of high-resolution images now being acquired, high spatial resolution (HSR) remote sensing imagery scene classification has drawn great attention but is still a challenging task due to the complex arrangements of the ground objects in HSR imagery, which leads to the semantic gap between low-level features and high-level semantic concepts. As a feature representation method for automatically learning essential features from image data, convolutional neural networks (CNNs) have been introduced for HSR remote sensing image scene classification due to their excellent performance in natural image classification. However, some scene classes of remote sensing images are object-centered, i.e., the scene class of an image is decided by the objects it contains. Although previous methods based on CNNs have achieved comparatively high classification accuracies compared with the traditional methods with handcrafted features, they do not consider the scale variation of the objects in the scenes. This makes it difficult to directly utilize CNNs on those remote sensing images belonging to object-centered classes to extract features that are robust to scale variation, leading to wrongly classified scene images. To solve this problem, scene classification based on a deep random-scale stretched convolutional neural network (SRSCNN) for HSR remote sensing imagery is proposed in this paper. In the proposed method, patches with a random scale are cropped from the image and stretched to the specified scale as the input to train the CNN. This forces the CNN to extract features that are robust to the scale variation. Furthermore, to further improve the performance of the CNN, a robust scene classification strategy is adopted, i.e., multi-perspective fusion. The experimental results obtained using three datasets—the UC Merced dataset, the Google dataset of SIRI-WHU, and the Wuhan IKONOS dataset—confirm that the proposed method performs better than the traditional scene classification methods.
Highlights
Remote sensing image scene classification, which involves dividing images into different categories without semantic overlap [1], has recently drawn great attention due to the increasing availability of high-resolution remote sensing data
It is difficult to achieve satisfactory results in scene classification directly based on low-level features, such as the spectral, textural, and geometrical attributes, because of the so-called “semantic gap” between low-level features and high-level semantic concepts, which is caused by the object category diversity and the distribution complexity in the scene
The semantic allocation level (SAL) multifeature fusion strategy based on the probabilistic topic model (PTM) (SAL-PTM) [10] was proposed to combine three complementary features, i.e., the spectral, texture, and scale-invariant feature transform (SIFT) features
Summary
Remote sensing image scene classification, which involves dividing images into different categories without semantic overlap [1], has recently drawn great attention due to the increasing availability of high-resolution remote sensing data. The features extracted by these methods are not robust to the object scale, leading to unsatisfactory remote sensing image scene classification results. In order to solve the object scale variation problem, scene classification based on a deep randomscale stretched convolutional neural network (SRSCNN) is proposed in this paper. In SRSCNN, patches with a random scale are cropped from the image, and patch stretching is applied, which simulates the object scale variation to ensure that the CNN learns robust features from the remote sensing images. Scene Classification for HSR Imagery Based on a Deep Random-Scale Stretched Convolutional Neural Network To solve this object scale problem, the main idea in this paper is to generate multiple-scale samples to train the CNN model, forcing the trained CNN model to correctly classify the same image with different scales. Each experiment on each dataset was repeated ffiive ttiimmeess,,ananddthtehaevaevraegraegcelascslaifiscsiaftiicoantiaocncuarcaccuyrawcays wreacosrrdeecdo.rTdoedk.eeTpothkienegps sthiminpglse,sCimCNplNe, dCeCnoNteNs dtheencootemsmthoencCoNmNmwonithCoNuNt rawnidthoomu-tscraalnedsotrmet-cshcianleg satnrdetmchuinltgi-paenrdspmecutlitvie-pfeurssipoenc,tCivNeNfuVsidoenn,oCteNsNthVe dcoemnomteosnthCeNcNomwmitohnmCuNltNi-pweritshpemcutilvtei-pfuesrisopnecbtuivtenfoutsriaonndboumt -nsoctalreanstdroetmch-sincagl,eSsRtrSeCtcNhNin-gN, SVRdSeCnNotNesNthVe CdNenNotwesitthheraCnNdoNmw-sicthalerasntrdeotcmh-isncgalbeusttrneottchminugltib-upternsoptemctiuvleti-fpuesirospn.ective fusion
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
