Exploración de características geológicas de la superficie lunar de Encélado (Saturno) utilizando técnicas de aprendizaje automático para la clasificación de imágenes satelitales

dc.contributor.advisorMontenegro Diaz, Alvaro Mauriciospa
dc.contributor.authorForero Larrotta, Juliana Paolaspa
dc.date.accessioned2024-09-06T15:01:46Z
dc.date.available2024-09-06T15:01:46Z
dc.date.issued2024
dc.descriptionilustraciones, diagramas, fotografías, tablasspa
dc.description.abstractEl análisis de imágenes satelitales brinda mucha información valiosa que puede ser aplicada en diferentes contextos. En el caso de los cuerpos planetarios, el análisis de imágenes tomadas por sondas espaciales es útil para determinar el origen, evolución, distribución y comportamiento geológico de un cuerpo del sistema solar (planetas, lunas, meteoritos). Gracias a estos datos podemos determinar la edad geológica relativa de un cuerpo con base en impactos de meteoritos observados que como consecuencia dejan cráteres y deforman superficies planetarias y lunares, incluso podemos determinar distintas propiedades fisicoquímicas que nos pueden dar indicio de la existencia de fuentes de agua, composiciones atmosféricas, abundancia de elementos y minerales de interés y así comprender mejor la mecánica interna y externa del cuerpo. Este tipo de aplicación requiere la identificación manual de particularidades y características morfológicas y composicionales en cientos de imágenes tomadas por diferentes instrumentos en diferentes longitudes de onda, con diferentes características de resolución, ángulo de captura de la imagen, tiempo de exposición, longitud de onda captada, posición del cuerpo planetario respecto a su estrella más cercana, ente otros factores. El presente trabajo es una aplicación de modelos de aprendizaje automático de tipo no supervisado como el clustering para el procesamiento de imágenes de la luna de Encelado del planeta Saturno tomadas por la sonda Cassini-Huygens entre los años 2005 y 2017 y que se pueden encontrar en el siguiente repositorio del proyecto PILOT de la NASA (Planetary Image Locator Tool) (USGS/NASA, 2015, https://pilot.wr.usgs.gov), el cual es el archivo más completo de imágenes tomadas por sondas enviadas al espacio hasta la fecha. La clasificación de las imágenes tomadas por la sonda Cassini-Huygens permite ampliar la comprensión de los diferentes procesos que dieron lugar a una gran variedad de características morfológicas y tectónicas de su superficie, ya que basta con observar y clasificar distintos tipos de geoformas como lo son cráteres de impacto, fracturas, fallas, surcos, elevaciones, montañas, distribución y tamaño de partículas, para entender la dinámica geológica de la luna y su dinámica criovolcánica. Se plantea un marco de trabajo para la aplicación de modelos de aprendizaje automático no supervisado como el k-means, MeanShift, DBSCAN y Mixtura Gaussiana para abordar el problema de segmentación de la imagen y detección de particularidades en la clasificación de morfologías, ya que este tipo de algoritmos permite dividir un conjunto de imágenes en grupos basados en sus características o propiedades identificadas, adicionalmente se entrena un modelo de red neuronal convolucional que toma las imágenes etiquetadas con k-means y busca predecir la clase sobre nuevas imágenes. Se prueban distintas combinaciones de técnicas de preprocesamiento y extracción de características y se aplica la técnica de transferencia de aprendizaje en modelos de redes neuronales preentrenadas tanto para poder extraer las características de una imagen, como para poder entrenar un clasificador que permita agrupar nuevas imágenes lunares en las categorías identificadas. Para Encélado, la sonda Cassini Huygens cuenta con dos tipos de instrumentos para la toma de datos: ISS (Cassini Imaging Science Subsystem) y VIMS (Visual and Infrared Mapping Spectrometer), los cuales producen imágenes de alta resolución. Se usaron 5167 imágenes mapeadas mediante un lente NA (Narrow Angle), es decir, un ángulo de imagen normal y no más amplio, del instrumento ISS que cuenta con imágenes tanto en el canal visible como en el infrarrojo cercano, estas imágenes fueron tomadas a distintas distancias y capturan distintas regiones de la luna (Texto tomado de la fuente).spa
dc.description.abstractThe analysis of satellite images provides valuable information that can be applied in different contexts. In the case of planetary bodies, the analysis of images taken by space probes is useful to determine the origin, evolution, distribution and geological behavior of a solar system body (planets, moons, meteorites). Thanks to these data we can determine the relative geological age of a body based on observed meteorite impacts that as a consequence leave craters and deform planetary and lunar surfaces, we can even determine different physicochemical properties that can give us an indication of the existence of water sources, atmospheric compositions, abundance of elements and minerals of interest and thus better understand the internal and external mechanics of the body. This type of application requires the manual identification of morphological and compositional features and characteristics in hundreds of images taken by different instruments at different wavelengths, with different resolution characteristics, image capture angle, exposure time, wavelength captured, position of the planetary body respect to its nearest star, among other factors. The present work is an application of unsupervised machine learning models such as clustering for the processing of images of the Enceladus moon of the planet Saturn taken by the Cassini-Huygens probe between the years 2005 and 2017 and that can be found in the following repository of NASA’s PILOT (Planetary Image Locator Tool) project (USGS/NASA, 2015, https://pilot.wr.usgs.gov), which is the most complete archive of images taken by probes sent to space to date. The classification of the images taken by the Cassini-Huygens probe allows to understand the different processes that gave rise to a great variety of morphological and tectonic characteristics of its surface, since it is enough to observe and classify different types of geoforms such as impact craters, fractures, faults, grooves, elevations, mountains, distribution and size of particles, to understand the geological dynamics of the moon and its cryovolcanic dynamics. A framework is proposed for the application of unsupervised machine learning models such as k-means, MeanShift, DBSCAN and Gaussian Mixture to address the problem of image segmentation and detection of particularities in the classification of morphologies, since this type of algorithms allows dividing a set of images into groups or clusters based on their identified characteristics or properties. In addition, a convolutional neural network model is trained that takes the images labeled with k-means and seeks to predict the class on new images. Different combinations of preprocessing and feature extraction techniques are tested and the transfer learning technique is applied to pre-trained neural network models both to extract features from an image and to train a classifier to group new lunar images into the identified categories. For Enceladus, the Cassini Huygens probe has two types of instruments for data acquisition: ISS (Cassini Imaging Science Subsystem) and VIMS (Visual and Infrared Mapping Spectrometer), which produce high-resolution images. We used 5167 images mapped by means of a NA (Narrow Angle) lens, that is, a normal image angle and not wider, of the ISS instrument that has images in both the visible and near infrared channels, these images were taken at different distances and capture different regions of the moon.eng
dc.description.degreelevelMaestríaspa
dc.description.degreenameMagíster en Ciencias - Estadísticaspa
dc.format.extentviii, 162 páginasspa
dc.format.mimetypeapplication/pdfspa
dc.identifier.instnameUniversidad Nacional de Colombiaspa
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombiaspa
dc.identifier.repourlhttps://repositorio.unal.edu.co/spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/86800
dc.language.isospaspa
dc.publisherUniversidad Nacional de Colombiaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Bogotáspa
dc.publisher.facultyFacultad de Cienciasspa
dc.publisher.placeBogotá, Colombiaspa
dc.publisher.programBogotá - Ciencias - Maestría en Ciencias - Estadísticaspa
dc.relation.referencesSmoothing Images. In: Docs OpenCV. Enhanced by Google https://docs.opencv.org/4.x/d4/d13/tutorial_py_filtering.htmlspa
dc.relation.referencesADDISON, Automatic: Pros and Cons of Gaussian Smoothing. In: Automatic Addison (2019). https://automaticaddison.com/pros-and-cons-of-gaussian-smoothing/spa
dc.relation.referencesALIYARI GHASSABEH, Youness: On the convergence of the mean shift algorithm in the one-dimensional space. In: Journal of Multivariate Analysis (2013), Nr. arXiv:1407.2961. Bibcode:2013PaReL..34.1423A. doi:10.1016/j.patrec.2013.05.004. S2CID 10233475, S. 1423–1427spa
dc.relation.referencesALIYARI GHASSABEH, Youness: A sufficient condition for the convergence of the meanshift algorithm with Gaussian kernel. In: Journal of Multivariate Analysis (2015), Nr. doi:10.1016/j.jmva.2014.11.009, S. 1–10spa
dc.relation.referencesBARGHOUT, Lawrence W. L. Lauren: Perceptual information processing system. In: Paravue Inc. U.S. Patent Application 10 (2003), S. 618,543spa
dc.relation.referencesBEER A., Hohma E. Jahn P. Frey C. Assent I. Draganov A. A. Draganov A.: Connecting the Dots – Density-Connectivity Distance unifies DBSCAN, k-Center and Spectral Clustering. In: KDD ’23, Long Beach, CA, USA (2023), Nr. 10.1145/3580305.3599283, 80-92. https://web.archive.org/web/20100421170848/http://academic.research.microsoft.com/CSDirectory/paper_category_7.htmspa
dc.relation.referencesBRADSKI, Adrian Gary; K. Gary; Kaehler: Learning OpenCV: Computer vision with the OpenCV library. In: O’Reilly Media, Inc. (2008), S. 6spa
dc.relation.referencesC., Menor-Salván: Qué sabemos realmente de la vida y la habitabilidad en Encelado, la luna de Saturno. In: The Conversation (2023).spa
dc.relation.referencesC., Piech: Stanford CS221. In: stanford.edu (2013). https://stanford.edu/~cpiech/cs221/handouts/kmeans.htmlspa
dc.relation.referencesC., Rebecca: K-means Clustering. In: Steorts, Duke University (2018), Nr. STA 325, Chapter10 ISL. http://www2.stat.duke.edu/~rcs46/lectures_2017/10-unsupervise/10-kmeans_v2.pdfspa
dc.relation.referencesC., Yizong: Mean Shift, Mode Seeking, and Clustering. In: IEEE Transactions on Pattern Analysis and Machine Intelligence 8 (1995), Nr. 10.1109/34.400568, S. 790–799spa
dc.relation.referencesCHANG, Kenneth: Cassini Vanishes Into Saturn, Its Mission Celebrated and Mourned. In: The New York Times (2017). https://www.nytimes.com/2017/09/14/science/cassini-grand-finale-saturn.htmlspa
dc.relation.referencesCHANG, Kenneth: The ’Sounds’ of Space as NASA’s Cassini Dives by Saturn. In: The New York Times (2017). https://www.nytimes.com/2017/05/03/science/nasa-cassini-sound-recording-saturn.htmlspa
dc.relation.referencesCHENG, Yizong: Mean Shift, Mode Seeking, and Clustering. In: HIPRIEEE Transactions on Pattern Analysis and Machine Intelligence2 (1995), Nr. doi:10.1109/34.400568, S. 790–799spa
dc.relation.referencesCOATES A., Ng A. Lee H. H. Lee H.: An Analysis of Single-Layer Networks in Unsupervised Feature Learning. In: Stanford University, University of Michigan (2011). https: //proceedings.mlr.press/v15/coates11a/coates11a.pdfspa
dc.relation.referencesCOHN R., Holm E.: Unsupervised machine learning via transfer learning and k-means clustering to classify materials image data. In: arXiv (2020), Nr. arXiv:2007.08361v1 [cond-mat.mtrl-sci]. https://arxiv.org/pdf/2007.08361v1.pdfspa
dc.relation.referencesCORUM, Jonathan: Mapping Saturn’s Moons. In: The New York Times (2015). https://www.nytimes.com/interactive/2015/12/18/science/space/nasa-cassini-maps-saturns-moons.htmlspa
dc.relation.referencesCORUM, Jonathan: Saturn Plunge Nears for Cassini Spacecraft. In: NASA - National Aeronautics and Space Administration (2017). https://www.nasa.gov/feature/jpl/saturn-plunge-nears-for-cassini-spacecraftspa
dc.relation.referencesD., Bolles: Cassini: FAQ. In: Science NASA (2023). https://science.nasa.gov/mission/cassini/faq/spa
dc.relation.referencesD., Marin: Encélado tiene un océano global subterráneo. In: Eureka (2015)spa
dc.relation.referencesDOMÍNGUEZ, Nuño: Una sonda de la NASA confirma que puede haber vida en Encélado. In: El País (2017)spa
dc.relation.referencesDYCHES, D.; Mullins S. P.; Brown B. P.; Brown: Cassini Spacecraft Reveals 101 Geysers and More on Icy Saturn Moon. (2014)spa
dc.relation.referencesE., Guzmán L.: Métricas para la validación de Clustering / Universidad Nacional de Colombia Ingeniería de Sistemas y Computación. – Forschungsberichtspa
dc.relation.referencesESTER, Hans-Peter; Sander Jörg; Xu Xiaowei Simoudis Evangelos; Han Jiawei; Fayyad Usama M. Martin; Kriegel K. Martin; Kriegel: A density-based algorithm for discovering clusters in large spatial databases with noise. In: Proceedings of the Second International Conference on Knowledge Discovery and Data Mining (1996), Nr. 1-57735-004-9, S. 226–231spa
dc.relation.referencesESTER M., Sander J. Kriegel H.-P. ; X., Xu: A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. In: In Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining (KDD’96) (1996), 226-231. https://faratarjome.ir/u/media/shopping_files/store-EN-1486556195-4303.pdfspa
dc.relation.referencesF., Chollet: Transfer learning fine-tuning. Building powerful image classification models using very little data. In: The Keras Blog (2020). https://keras.io/guides/transfer_learning/spa
dc.relation.referencesFISHER, Walker W. Perkins: Spatial Filters - Laplacian of Gaussian. Image Processing Learning Resources. In: HIPR2 (2003). https://homepages.inf.ed.ac.uk/rbf/HIPR2/ log.htmspa
dc.relation.referencesFUKUNAGA, Keinosuke; Larry D. H.: The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition. In: IEEE Transactions on Information Theory (1975), Nr. doi:10.1109/TIT.1975.1055330, S. 32–40spa
dc.relation.referencesGOODFELLOW, Bengio Y. Courville A. I.: Deep Learning. MIT Press, 20161spa
dc.relation.referencesH., Rodríguez: Saturno, el famoso planeta de los anillos. In: National Geographic. (2023). https://www.nationalgeographic.com.es/ciencia/ saturno-famoso-planeta-anillos_18640spa
dc.relation.referencesHE, Zhang X. Ren S. Sun J. K.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)spa
dc.relation.referencesHOSNA A., Gyalmo J. Alom Z. Aung Z. Azim M. Merry E. E. Merry E.: Transfer learning: a friendly introduction. In: Journal of Big Data (2022). https://journalofbigdata.springeropen.com/articles/10.1186/s40537-022-00652-wspa
dc.relation.referencesI., Dabbura: K-means Clustering: Algorithm, Applications, Evaluation Methods, and Drawbacks. In: Towards Data Science (2018).spa
dc.relation.referencesJ., Lara: Diplomado DEEP LEARNING: INTRODUCCIÓN AL APRENDIZAJE PROFUNDO CON PYTHON. Módulo 5, Clase Redes Neuronales Convolucionales / Universidad Nacional de Colombia. Departamento de Ingenieria Industrial y de Sistemas. 2022. – Forschungsberichtspa
dc.relation.referencesJ. ZHANG, et al. et a.: Wireless Channel Propagation Scenarios Identification: A Perspective of Machine Learning. In: IEEE Acess 4 (2016), Nr. 10.1109/ACCESS.2020.2979220spa
dc.relation.referencesK., Nayar S.: Introduction to Computer Vision. Monograph: FPCV-0-1 / Computer Science Department, School of Engineering and Applied Sciences, Columbia University. 2022. – Forschungsberichtspa
dc.relation.referencesKE T., Guo Y. Wang X. Yu S. Hwang J. J. Hwang J.: Unsupervised Hierarchical Semantic Segmentation with Multiview Cosegmentation and Clustering Transformers. In: UC Berkeley / ICSI (2022), Nr. arXiv:2204.11432v1spa
dc.relation.referencesKRIZHEVSKY, Sutskever I. Hinton G. E. A.: Imagenet classification with deep convolutional neural networks. In: Neural Computation (2012)spa
dc.relation.referencesKRIZHEVSKY, Sutskever I. Hinton G. E. A.: Very deep convolutional networks for large-scale image recognition. In: arXiv (2014)spa
dc.relation.referencesKURT DEMAAGD, Nathan Oostendorp Katherine S. Anthony Oliver O. Anthony Oliver: Practical Computer Vision with SimpleCV. O’Reilly Media, Inc., 2012spa
dc.relation.referencesL., Forero: Distribución de crioclastos en la región damascus sulcus de la luna Encélado (Saturno) usando imágenes vims. (2016)spa
dc.relation.referencesLECUN, Bengio Y. Hinton G. Y.: Deep learning. In: Nature 521 (2015), S. 436–444spa
dc.relation.referencesLECUN, Boser B. Denker J. S. Henderson D. Howard R. E. Hubbard W. Jackel L. D. Y.: Backpropagation Applied to Handwritten Zip Code Recognition. In: Neural Computation (1989)spa
dc.relation.referencesLI, Zhanyi; Wu F. Xiangru; Hu H. Xiangru; Hu: A note on the convergence of the mean shift. In: Journal of Multivariate Analysis (2007), Nr. Bibcode:2007PatRe..40.1756L. doi:10.1016/j.patcog.2006.10.016, S. 1756–1762spa
dc.relation.referencesLIU Y., Xiong H. Gao X. Wu J. Li Z. Z. Li Z.: Understanding of Internal Clustering Validation Measures. In: IEEE International Conference on Data Mining (2010). https://medium.com/@haataa/how-to-measure-clustering-performances-when-there-are-no-ground-truth-db027e9a871cspa
dc.relation.referencesMONGODB: What is Unstructured Data? https://www.mongodb.com/resources/basics/unstructured-data. Version: 2024. – Accessed: 2024-07-27spa
dc.relation.referencesMOSHER, Dave: NASA will destroy a 3.26 billion Saturn probe this summer to protect an alien water world. In: Business Insider (2017). https://www.businessinsider.com/cassini-death-grand-finale-reason-2017-4spa
dc.relation.referencesNIELSEN F., Nock R.: On region merging: The statistical soundness of fast sorting, with applications. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2 (2003), Nr. doi:10.1109/CVPR.2003.1211447. ISBN 0-7695-1900-8, S. II:19–26spa
dc.relation.referencesOVERBYE, Dennis: Cassini Flies Toward a Fiery Death on Saturn. In: The New York Times (2017). https://www.nytimes.com/2017/09/08/science/cassini-saturn-nasa.htmlspa
dc.relation.referencesP., Getreuer: A Survey of Gaussian Convolution Algorithms. In: Image Processing On Line -IPOL (2013). https://homepages.inf.ed.ac.uk/rbf/HIPR2/log.htmspa
dc.relation.referencesPAPERS, Microsoft Academic S.: Most cited data mining articles according to Microsoft academic search; DBSCAN is on rank 24. In: Microsoft Academic Search: Papers (2010). https://web.archive.org/web/20100421170848/http://academic.research.microsoft.com/CSDirectory/paper_category_7.htmspa
dc.relation.referencesPEDREGOSA, Varoquaux G. Gramfort A. Michel V. Thirion B. Grisel O. ... Duchesnay E. F.: Scikit-learn: Machine Learning in Python. In: Journal of Machine Learning Research 12 (2011), S. 2825–2830spa
dc.relation.referencesPINHEIRO P., Collobert R.: From Image-level to Pixel-level Labeling with Convolutional Networks. In: Computer Vision Foundation 2 (2015). https://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Pinheiro_From_Image-Level_to_2015_CVPR_paper.pdfspa
dc.relation.referencesPLATT, B. J.; B. J.; Bell: NASA Space Assets Detect Ocean inside Saturn Moon. (2014)spa
dc.relation.referencesPRITT M, Chern G.: Satellite Image Classification with Deep Learning. In: Research Gate (2017)spa
dc.relation.referencesPULLI, Anatoly; Kornyakov Kirill; Eruhimov V. Kari; Baksheev B. Kari; Baksheev: Realtime Computer Vision with OpenCV. In: Queue 10 (2012), Nr. doi:10.1145/2181796.2206309, S. 40:40–40:56spa
dc.relation.referencesPULLI, Anatoly; Kornyakov Kirill; Eruhimov V. Kari; Baksheev B. Kari; Baksheev: Intel acquires Itseez. In: Wayback Machine (2019)spa
dc.relation.referencesR. A., P. G.; Bordi J. J.; Criddle K. E.; Ionasescu R.; Jones J. B.; MacKenzie R. A.; Meek M. C. et a. Jacobson; Antreasian A. Jacobson; Antreasian: The Gravity Field of the Saturnian System from Satellite Observations and Spacecraft Tracking Data. In: The Astronomical Journal 132 (2006), S. 2520–2526spa
dc.relation.referencesR. FISHER, A. W. S. Perkins P. S. Perkins ; WOLFART., E.: A to Z of Common Image Processing Concepts. Edge Detectors. In: Image Processing Learning Resources, HIPR2 (2003). https:// homepages.inf.ed.ac.uk/rbf/HIPR2/edgdetct.htmspa
dc.relation.referencesR. FISHER, A. W. S. Perkins P. S. Perkins ; WOLFART., E.: A to Z of Common Image Processing Concepts. Grayscale Images. In: Image Processing Learning Resources, HIPR2 (2003). https://homepages.inf.ed.ac.uk/rbf/HIPR2/gryimage.htmspa
dc.relation.referencesR. FISHER, A. W. S. Perkins P. S. Perkins ; WOLFART., E.: A to Z of Common Image Processing Concepts. Kernel. In: Image Processing Learning Resources, HIPR2 (2003). https://homepages.inf.ed.ac.uk/rbf/HIPR2/kernel.htmspa
dc.relation.referencesR. FISHER, A. W. S. Perkins P. S. Perkins ; WOLFART., E.: A to Z of Common Image Processing Concepts. Pixel Values. In: Image Processing Learning Resources, HIPR2 (2003). https://homepages.inf.ed.ac.uk/rbf/HIPR2/value.htmspa
dc.relation.referencesRADFORD A., Salimans T. Sutskever I. Narasimham K. K. Narasimham K.: Improving Language Understanding by Generative Pre-Training. In: OpenAI (2018). https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdfspa
dc.relation.referencesSAAVEDRA, Sanchez J. F.: 48. CRIOVOLCANISMO, LOS MECANISMOS TÉRMICOS Y DE FUERZAS DE MAREA EN EL POLO SUR DE ENCÉLADO: ANÁLOGOS EN LA TIERRA Y TITÁN. In: XV CONGRESO COLOMBIANO DE GEOLOGÍA, 2015 ”Innovar en Sinergia con el Medio Ambiente” Bucaramanga, Colombia (2015). https://www.researchgate.net/publication/ 282253101_Criovolcanismo_los_mecanismos_termicos_y_de_fuerzas_de_marea_en_el_polo_sur_de_Encelado_Analogos_en_la_Tierra_y_Titanspa
dc.relation.referencesSANDER, Martin; Kriegel Hans-Peter; Xu X. Jörg; Ester E. Jörg; Ester: Density-Based Clustering in Spatial Databases: The Algorithm GDBSCAN and Its Applications. In: Data Mining and Knowledge Discovery. Berlin: Springer-Verlag. 2 (1998), Nr. doi:10.1023/A:1009745219419, S. 169–194spa
dc.relation.referencesSCHMIDHUBER, J.: Deep learning in neural networks: An overview. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition 61 (2015), S. 85–117spa
dc.relation.referencesSCHUBERT, Jörg; Ester Martin; Kriegel Hans Peter; Xu X. Erich; Sander S. Erich; Sander: DBSCAN Revisited, Revisited: Why and How You Should (Still) Use DBSCAN. In: ACM Trans. Database Syst. 42 (2017), Nr. 10.1145/3068335, S. 226–231spa
dc.relation.referencesSHAPIRO, G L. G. S. L. G. Stockman: Computer Vision. Prentice Hall, 2001. – 137, 150 S.spa
dc.relation.referencesSHAPIRO L. G., Stockman G. C.: Computer Vision. In: New Jersey, Prentice-Hall ISBN 0-13-030796-3 (2001), S. 279–325spa
dc.relation.referencesSPENCER, F. J. R.; N. J. R.; Nimmo: Enceladus: An Active Ice World in the Saturn System. In: Annual Review of Earth and Planetary Sciences 41 (2013), S. 693–717spa
dc.relation.referencesSZELISKI, Richard: Computer Vision, Algorithms and Applications. Springer, 2011spa
dc.relation.referencesTEAM, Great L.: OpenCV Tutorial: A Guide to Learn OpenCV in Python. In: Great Learning (2023). https://www.mygreatlearning.com/blog/opencv-tutorial-in-python/spa
dc.relation.referencesTEAM, OpenCV: CUDA. In: opencv.org (2020)spa
dc.relation.referencesTEAM, OpenCV: OpenCV Consulting Site. In: OpenCV Consulting Site (2024)spa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-NoComercial-SinDerivadas 4.0 Internacionalspa
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/spa
dc.subject.ddc310 - Colecciones de estadística generalspa
dc.subject.ddc520 - Astronomía y ciencias afinesspa
dc.subject.ddc000 - Ciencias de la computación, información y obras generalesspa
dc.subject.lembGEOLOGÍA LUNARspa
dc.subject.lembLunar geologyeng
dc.subject.lembSONDAS ESPACIALESspa
dc.subject.lembSpace probeseng
dc.subject.lembCRONOLOGÍA GEOLÓGICAspa
dc.subject.lembGeological timeeng
dc.subject.lembMETEORITOSspa
dc.subject.lembMeteoriteseng
dc.subject.lembCRATERES METEORICOSspa
dc.subject.lembMeteorite craterseng
dc.subject.lembPROPIEDADES FISICOQUÍMICASspa
dc.subject.lembChemicophysical propertieseng
dc.subject.proposalAprendizaje automáticospa
dc.subject.proposalAprendizaje profundospa
dc.subject.proposalClusteringeng
dc.subject.proposalRedes neuronalesspa
dc.subject.proposalTransferencia de aprendizajespa
dc.subject.proposalImágenes satelitalesspa
dc.subject.proposalAprendizaje no supervisadospa
dc.subject.proposalMachine learningeng
dc.subject.proposalDeep learningeng
dc.subject.proposalNeural networkseng
dc.subject.proposalUnsupervised learningeng
dc.subject.proposalSatellite imageseng
dc.subject.proposalTransfer learningeng
dc.titleExploración de características geológicas de la superficie lunar de Encélado (Saturno) utilizando técnicas de aprendizaje automático para la clasificación de imágenes satelitalesspa
dc.title.translatedExploration of geological features of the lunar surface of Enceladus (Saturn) using machine learning and deep learning techniques for image classificationeng
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.redcolhttp://purl.org/redcol/resource_type/TMspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
dcterms.audience.professionaldevelopmentEstudiantesspa
dcterms.audience.professionaldevelopmentInvestigadoresspa
dcterms.audience.professionaldevelopmentMaestrosspa
dcterms.audience.professionaldevelopmentMedios de comunicaciónspa
dcterms.audience.professionaldevelopmentPadres y familiasspa
dcterms.audience.professionaldevelopmentPersonal de apoyo escolarspa
dcterms.audience.professionaldevelopmentProveedores de ayuda financiera para estudiantesspa
dcterms.audience.professionaldevelopmentPúblico generalspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
1026277472.2024.pdf
Tamaño:
24.29 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ciencias Estadística

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
5.74 KB
Formato:
Item-specific license agreed upon to submission
Descripción: