The recent advancements of technology in remote sensing enable us to get very high-resolution images (VHR). To do scene classification in these images turned significant and become a challenging problem due to the lack of availability of adequate labelled data. We get over fitting problems by training a limited amount of labelled data. Considering the features obtained by deep learning convolutional nets as inputs we address this issue. Here we utilize the existing VGG16, Alex Net frameworks as a feature extractor to extricate informative features from the authentic VHR Images. Succeeding, discuss feature concatenation and classification framework based on Canonical Correlation Analysis (CCA). This strategy uses the correspondence of two feature vectors of discriminant information and eradicates redundant information within the features. This permits a more effective approach than conventional extraction techniques exclusively. The experimental results demonstrate that the feature concatenation strategy based on the CCA technique produces good informative features and accomplishes a higher accuracy with much dimension reduction than exclusively using the raw deep features. We use a 0.3 sub-meter resolution UC MERCED data set to explore our approach.
<p class="Abstract">Remote sensing images are obtained by electromagnetic measurement from the terrain of interest. In high-resolution remote sensing imageries extraction measurement technology plays a vital role. The scene classification is one of the interesting and challenging problems due to the similarity of image structure and the available HRRS image datasets are all small. Training new Convolutional Neural Networks (CNN) using small datasets is prone to overfitting and poor attainability. To overcome this situation using the features produced by pre-trained convolutional nets and using those features to train an image classifier. To retrieve informative features from these images we use the existing Alex Net, VGG16, and VGG19 frameworks as a feature extractor. To increase classification performance further makes an innovative contribution fusion of multilayer features obtained by using covariance. First, to extract multilayer features, a pre-trained CNN model is used. The features are then stacked, downsampling is used to stack features of different spatial dimensions together and the covariance for the stacked features is calculated. Finally, the resulting covariance matrices are employed as features in a support vector machine classification. The results of the experiments, which were conducted on two difficult data sets, UC Merced and SIRI-WHU. The proposed Staked Covariance method consistently outperforms and achieves better classification performance. Achieves accuracy by an average of 6 % and 4 %, respectively, when compared to corresponding pre-trained CNN scene classification methods.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.