Abstract
Abstract. The application of hyperspectral image analysis for land cover classification is mainly executed in presence of manually labeled data. The ground truth represents the distribution of the actual classes and it is mostly derived from field recorded information. Its manual generation is ineffective, tedious and very time-consuming. The continuously increasing amount of proprietary and publicly available datasets makes it imperative to reduce these related costs. In addition, adequately equipped computer systems are more capable of identifying patterns and neighbourhood relationships than a human operator. Based on these facts, an unsupervised labeling approach is presented to automatically generate labeled images used during the training of a convolutional neural network (CNN) classifier. The proposed method begins with the segmentation stage where an adapted version of the simple linear iterative clustering (SLIC) algorithm for dealing with hyperspectral data is used. Consequently, the Hierarchical Agglomerative Clustering (HAC) and Fuzzy C-Means (FCM) algorithms are employed to efficiently group similar superpixels considering distances with respect to each other. The distinct utilization of these clustering techniques defines a complementary stage for overcoming class overlapping during image generation. Ultimately, a CNN classifier is trained using the computed image to pixel-wise predict classes on unseen datasets. The labeling results, obtained using two hyperspectral benchmark datasets, indicate that the current approach is able to detect objects boundaries, automatically assign class labels to the entire dataset and to classify new data with a prediction certainty of 90%. Additionally, this method is also capable of achieving better classification accuracy and visual correspondence with reality than the ground truth images.
Highlights
Hyperspectral data consists of a collection of scene radiance arranged in a spatial-spectral datacube
Their work resulted in classification outcomes which can perform better than Support Vector Machine (SVM) classifiers
To this formulation, the value of the used parameter m has been treated as an adjustable constant which ensures regular superpixels form and, most importantly, a high capability of the superpixels to adhere to image boundaries (Achanta et al, 2012)
Summary
Hyperspectral data consists of a collection of scene radiance arranged in a spatial-spectral datacube. The instruments in the first group enable users to collect sensor data anywhere at any time They are built with advanced technology and own the ability to acquire images with high spatial and spectral resolution. Among the most emblematic European missions is CHRIS, which was the prime instrument of the Proba-1 spacecraft launched on the 22 October 2001 This sensor setup aims to explore the capabilities of imaging spectrometers on agile small satellite platforms (ESA, 2020a). Due to the relevance of these decisions, it is beneficial to provide the means to swiftly and automatically generate land cover information instead of waiting for the preparation of the ground truth These days it is possible to adapt the calculations of the classification process to use Graphics Processing Units (GPUs) (Wuttke et al, 2018). The capability of this approach is demonstrated for the task of land cover classification on two benchmark datasets
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: 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.