<\/div>Example<\/div>
<\/div><\/div>
<\/strong><\/p>\n\n\n\n
import numpy as np\nimport matplotlib.pyplot as plt\n\nx = np.linspace(0, 10, 100)\ny = np.sin(x)\n\nplt.plot(x, y)\nplt.title(\"Courbe du sinus\")\nplt.show()\n<\/code><\/pre>\n\n\n\n<\/div><\/div><\/strong><\/p>\n\n\n\n The graph will be displayed directly below the code.<\/p>\n\n\n\n<\/a><\/figure>\n\n\n\n \n\n\n\n<\/span>Backup<\/span><\/h3>\n\n\n\n\nYour notebooks are automatically backed up to your Google Drive.<\/li>\n\n\n\nYou can also import\/export .ipynb or .py files.<\/li>\n<\/ul>\n\n\n\n \n\n\n\n<\/span>S2DR3 (Sentinel-2 Deep Resolution 3.0<\/em>)<\/span><\/h2>\n\n\n\n \n\n\n\n<\/span>Overview<\/span><\/h3>\n\n\n\n\n<\/a><\/figure>\n\n\n\n<\/a><\/figure>\n\n\n\n<\/a><\/figure>\n<\/figure>\n\n\n\n\nS2DR3 is an artificial intelligence super-resolution model applied to images from the Sentinel-2 satellite (Copernicus program mission). (Medium<\/a>)<\/li>\n\n\n\n Its goal is to take Sentinel-2 multispectral bands (native resolutions of 10 m, 20 m, and sometimes 60 m) and \u201cenhance\u201d them to \u2248 1 m\/pixel for all bands. (LinkedIn<\/a>)<\/li>\n\n\n\n This provides access to much finer spatial details (roads, buildings, small objects, crop heterogeneity, etc.) than the standard Sentinel-2 resolution allows. (MalaGIS<\/a>)<\/li>\n<\/ul>\n\n\n\n \n\n\n\n<\/span>Why it’s interesting<\/span><\/h3>\n\n\n\nSome key points:<\/p>\n\n\n\n\nBetter spatial resolution: going from 10 m to 1 m is a leap in visible detail. This opens up uses that previously required very expensive commercial data. (data jungle adventures<\/a>)<\/li>\n\n\n\n Complete multispectral coverage: unlike some models that only focus on a few bands (e.g., RGB), S2DR3 processes all Sentinel-2 bands, which preserves spectral richness (useful for agriculture, vegetation, water, etc.). (Medium<\/a>)<\/li>\n\n\n\n Various applications: precision agriculture, detailed urban mapping, environmental monitoring, infrastructure. For example, a study shows that vegetation indices derived from S2DR3 correlate better with crop yields than those from the standard 10 m version. (MDPI<\/a>)<\/li>\n<\/ul>\n\n\n\n \n\n\n\n<\/span>Simplified operation<\/span><\/h2>\n\n\n\n\nWe start with a Sentinel-2 scene (L1C or L2A level) containing bands with resolutions of 10 m, 20 m, and 60 m. (MDPI<\/a>)<\/li>\n\n\n\n The model (deep neural network) learns to \u201csuper-resolve\u201d each band: that is, to predict a finer version of the image from the original, while trying to preserve the correct spectral signatures. (data jungle adventures<\/a>)<\/li>\n\n\n\n The result: a georeferenced multispectral image at ~1 m\/pixel, including all bands.<\/li>\n\n\n\n The script below uses this model via a Python package s2dr3.inferutils, to automate the call, scene retrieval, processing, export to Google Drive, etc.<\/li>\n<\/ul>\n\n\n\n \n\n\n\n<\/span>Practical uses in a script<\/span><\/h3>\n\n\n\nLet’s see how the code fits into this framework:<\/p>\n\n\n\n <\/div>Example de code S2DR3<\/div><\/div><\/div><\/strong><\/p>\n\n\n\n# Lien vers le dossier Drive\n!ln -s \/content\/drive\/MyDrive\/Sentinel2_1m \/content\/output\n\n# Installation du paquet S2DR3\n!pip -q install https:\/\/storage.googleapis.com\/...\/s2dr3-20250905.1-cp312-cp312-linux_x86_64.whl\n\nimport s2dr3.inferutils\n\n# Coordonn\u00e9es (Loja, \u00c9quateur)\nlonlat = (-79.203, -4.008)\n\n# Date de l\u2019image\ndate = '2024-08-23'\n\n# Traitement\ns2dr3.inferutils.test(lonlat, date)\n<\/code><\/pre>\n\n\n\n<\/div><\/div><\/strong><\/p>\n\n\n\n \nYou specify an area of interest (longitude\/latitude): this triggers the retrieval of the Sentinel-2 scene closest to the date, covering this area.<\/li>\n\n\n\n The model then generates the 1 m version of the band set from the original image.<\/li>\n\n\n\n The result is saved in \/content\/output (which is linked to your Google Drive) for access\/backup.<\/li>\n<\/ul>\n\n\n\n<\/a><\/figure>\n\n\n\n \n\n\n\n<\/span>Limitations & points to note<\/span><\/h2>\n\n\n\nIt is also important to be aware of the limitations:<\/p>\n\n\n\n \nVariable quality depending on area\/date: The model assumes that the image contains few clouds, little obscuring, and that surfaces are well represented. Under certain conditions (too much shade, clouds, very specific areas), super-resolution can produce artifacts.<\/li>\n\n\n\n Generated data, not \u201ctrue\u201d 1 m commercial data: Even if the resolution is ~1 m, this is still an estimate by a model, not a native 1 m acquisition. Care should therefore be taken with highly critical uses (e.g., legal or engineering).<\/li>\n\n\n\n Computing resources: Processing can be resource-intensive (GPU, memory). Free notebooks (e.g., Colab) have limitations. (See the following article on how to configure processing with a T4 GPU.)<\/li>\n\n\n\n Validations required: As with all AI, results must be validated according to the application. One study shows that for certain crops, the advantage is real but not automatic. (MDPI)<\/li>\n\n\n\n Cost\/licensing: Some commercial or large-scale versions may involve a cost or conditions of use. (data jungle adventures)<\/li>\n<\/ul>\n\n\n\n \n\n\n\n<\/span>Recommendations for use<\/span><\/h2>\n\n\n\nHere are some tips for getting the most out of S2DR3 in your scripts:<\/p>\n\n\n\n \nCheck that the chosen date has good coverage (few clouds) for the area of interest.<\/li>\n\n\n\n After processing, visually check the output (e.g., in QGIS or a notebook) for any artifacts.<\/li>\n\n\n\n Compare (if possible) with the original standard resolution version from Sentinel-2 to see what the model brings.<\/li>\n\n\n\n If you are performing analysis (e.g., vegetation, classification), test on a small batch before processing large areas.<\/li>\n\n\n\n Document the source clearly: \u201csuper-resolved image via S2DR3 on XX\/XX\/XXXX\u201d for transparency.<\/li>\n\n\n\n For production or commercial use, check the license terms for the model or generated products.<\/li>\n<\/ul>\n\n\n\n \n\n\n\nIn the next article, TUTORIAL: Using S2DR3 in Google Colab to study corals in Mauritius, we will look at a concrete example of how this procedure can be used.<\/p>\n\n\n\n \n","protected":false},"excerpt":{"rendered":"First of all, let’s give credit where credit is due! I discovered this technique thanks to Franz’s blog, which I invite you to explore if you speak Spanish. To process Sentinel 2 images, we will use…<\/p>\n","protected":false},"author":1,"featured_media":15855,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"give_campaign_id":0,"_bbp_topic_count":0,"_bbp_reply_count":0,"_bbp_total_topic_count":0,"_bbp_total_reply_count":0,"_bbp_voice_count":0,"_bbp_anonymous_reply_count":0,"_bbp_topic_count_hidden":0,"_bbp_reply_count_hidden":0,"_bbp_forum_subforum_count":0,"sfsi_plus_gutenberg_text_before_share":"","sfsi_plus_gutenberg_show_text_before_share":"","sfsi_plus_gutenberg_icon_type":"","sfsi_plus_gutenberg_icon_alignemt":"","sfsi_plus_gutenburg_max_per_row":"","_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[3833],"tags":[3835,3837,3839,3841,3843],"class_list":["post-15906","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-images-en","tag-google-colab-en","tag-images-en","tag-resolution-en","tag-s2dr3-en","tag-sentinel-2-en"],"aioseo_notices":[],"jetpack_featured_media_url":"https:\/\/i0.wp.com\/www.sigterritoires.fr\/wp-content\/uploads\/2025\/10\/article_sd3.jpg?fit=846%2C393&ssl=1","jetpack_shortlink":"https:\/\/wp.me\/p6XU0A-48y","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts\/15906","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/comments?post=15906"}],"version-history":[{"count":0,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts\/15906\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/media\/15855"}],"wp:attachment":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/media?parent=15906"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/categories?post=15906"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/tags?post=15906"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}