Cloud detection is a fundamental pre-processing task for high resolution satellite images, where the presence or the absence of the cloud plays an important role in making a decision for further processing. Existing techniques are based on per-pixel classification for region segmentation. However, due to the similarity of the features with other patterns like ice or snow, it may lead to misclassification. Moreover, cloud detection imposes the detection of cloud shadow as well since it also covers land areas. In order to come up with an efficient technique to tackle the complexity of pattern diversity, we exploit the recent advances in machine learning by designing and training a deep convolutional neural network model (ConvNet) based on multi-scale feature learning. Our proposed technique claims that different types of features can be learned at different scales to discriminate between image patterns. We chose two publicly available datasets for training. First, the 38-Cloud dataset was annotated as cloudy and non-cloudy classes. Second, the SPARCS (Spatial Procedures for Automated Removal of Cloud and Shadow) dataset with seven classes including cloud, ice/snow, and shadow. Both datasets contain images with four bands (R: Red, G: Green, B: Blue, Nir: Near Infrared), which we use as inputs of the ConvNet model for training and testing. The experimental results show that our proposed method can effectively detect clouds in complex scenes.
The geo-referencing information of the satellite imagery is obtained by the use of either attitudes and velocity/position provided by the satellite instrumentation , nevertheless the positioning using this approach depend directly on the quality of the ephemeris data. The image-to-image registration aims to find a geometric transformation relating two or more images in order to locate them in the same geographic reference. As the first Algerian high-resolution satellite ALSAT- 2A, has a nominal positional accuracy of around 300m, this last approach can be used to estimate and improve the raw image positional quality where these images are registered to orthoimages that are more accurate. Therefore, the resulting image have an enhanced geolocation quality than the one calculated using the estimation of the line of sight model. The purpose of this work is to provide a framework to automatically estimate the positional quality of Alsat2A image for cataloguing purpose, also to improve the geolocation accuracy of the level 1-A (geometrically raw) images by using other imagery that is already geo-localized with better accuracy. Landsat 8 orthoimagery coverage represents a data source that is worldwide available and its accuracy is sufficient for our work. Therefore, it is used as reference images to estimate and enhance the quality of raw ALSAT-2A images. This work is based on the use of open-source software such python, OpenCv and PostgreSQL and the open-data e.g. geo-referenced quick looks of landsat8 provided by EarthExplorer. Many experimentations has been conducted to decide on which algorithm is better to enhance the contrast and which spectral channel to use, also many methods for the homologue points extraction and description have been evaluated. Then, points of interest extracted from the reference orthoimagery are fed into a geographic database, allowing the use of these points as known ground points to estimate and enhance the quality of Alsat-2A imagery.
In this paper, a new pansharpening method, which uses nonnegative matrix factorization, is proposed to enhance the spatial resolution of remote sensing multispectral images. This method, based on the linear spectral unmixing concept and called joint spatial-spectral variables nonnegative matrix factorization, optimizes, by new iterative and multiplicative update rules, a joint-variables criterion that exploits spatial and spectral degradation models between the considered images. This criterion considers only two unknown high spatial-spectral resolutions variables. The proposed method is tested on synthetic and real datasets and its effectiveness, in spatial and spectral domains, is evaluated with established performance criteria. Results show the good performances of the proposed approach in comparison with other standard literature ones.
KEYWORDS: Satellites, Sensors, Systems modeling, Imaging systems, Cameras, 3D modeling, Data modeling, Mathematical modeling, Satellite imaging, Control systems
The use of the national very high resolution space system Alsat-2A is a primordial task having a significant technological and economical interest assuring the strengthening autonomy in terms of availability and coverage in the satellite data. Also it allows us to improve and update the base and thematic mapping throughout the national territory. Firstly, the characteristics of ALSAT-2A are presented, namely the images and the imaging system with a brief history of ALSAT program. Secondly, as a prerequisite, knowing the internal parameters is essentials to modelize the geometry of such imaging system. From metadata given by the images distributor and ground control points, several test are described and the results are presented. The test data are supplied by ASAL (Algerian Space Agency), the first dataset comprise a panchromatic image over the region of El Bayadh in the North West of Algeria equipped with nine GPS surveyed points. The second dataset is an along track stereoscopic panchromatic 1A level images over the town of sevilla in the south of Spain with 24 GCPs. Finally, a discussion on obtained results is dressed showing the geometric capability of ALSAT-2A.
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.