{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-11-02T02:00:36.453Z","headline":"Google DeepMind 用卫星数据和 AI 绘制全球森林类型地图 ForTy","description":"Google DeepMind 与世界资源研究所联合开发了 ForTy，这是首个全球森林类型地图，以 10 米分辨率将 2020 年全球土地划分为 6 类，包括原始森林、天然再生林、人工林、种植园、油棕橡胶可可等树木作物及其他土地。该地图主要为欧盟无毁林产品法规（EUDR）提供 2020 年基准，在 8190 个随机抽样点的专家复核中整体准确率为 90.2%，但对树木作物和人工林仅识别出约一半。","url":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","mainEntityOfPage":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","datePublished":"2026-10-31T10:15:45.000Z","dateModified":"2026-10-31T10:15:45.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://the-decoder.com/google-deepmind-creates-a-global-map-of-forests-with-satellite-data-and-ai","https://aihot.news/items/cbmtamj96m73qxjf1nnhteiv9"],"canonicalUrl":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","directAnswer":{"@type":"Answer","text":"Google DeepMind 与世界资源研究所联合开发 ForTy，将 2020 年全球土地以 10 米分辨率划分为六类森林、种植和其他土地类型，旨在提供与欧盟无毁林产品法规相关的基准地图。","url":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","dateCreated":"2026-10-31T10:15:45.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"The Decoder：AI News source article","url":"https://the-decoder.com/google-deepmind-creates-a-global-map-of-forests-with-satellite-data-and-ai","datePublished":"2026-10-31T10:15:45.000Z","provider":{"@type":"Organization","name":"The Decoder：AI News","url":"https://the-decoder.com/google-deepmind-creates-a-global-map-of-forests-with-satellite-data-and-ai"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.news/items/cbmtamj96m73qxjf1nnhteiv9","datePublished":"2026-10-31T10:15:45.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.news/items/cbmtamj96m73qxjf1nnhteiv9"}}],"aggregationSource":"The Decoder：AI News","originalPublisher":{"name":"The Decoder：AI News","url":"https://the-decoder.com/google-deepmind-creates-a-global-map-of-forests-with-satellite-data-and-ai"},"geoDeepAnswer":null,"article":{"id":"cbmtamj96m73qxjf1nnhteiv9","slug":"cbmtamj96m73qxjf1nnhteiv9","url":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","title":"Google DeepMind 用卫星数据和 AI 绘制全球森林类型地图 ForTy","title_en":"","summary":"Google DeepMind 与世界资源研究所联合开发了 ForTy，这是首个全球森林类型地图，以 10 米分辨率将 2020 年全球土地划分为 6 类，包括原始森林、天然再生林、人工林、种植园、油棕橡胶可可等树木作物及其他土地。该地图主要为欧盟无毁林产品法规（EUDR）提供 2020 年基准，在 8190 个随机抽样点的专家复核中整体准确率为 90.2%，但对树木作物和人工林仅识别出约一半。","source":"The Decoder：AI News","sourceUrl":"https://the-decoder.com/google-deepmind-creates-a-global-map-of-forests-with-satellite-data-and-ai","aiHotUrl":"https://aihot.news/items/cbmtamj96m73qxjf1nnhteiv9","publishedAt":"2026-10-31T10:15:45.000Z","category":"行业动态","score":58,"selected":false,"articleBody":["Google Deepmind and the World Resources Institute have developed ForTy, an AI map that distinguishes primary forests, timber plantations, and tree crop farms worldwide.","Satellite maps usually show whether trees grow in a given area without identifying the forest or crop type. ForTy, short for Forest Typology, goes further by assigning the world's land area in 2020 to 6 classes at 10-meter resolution. These cover primary forest, naturally regenerating forest, planted forest, plantation forest, tree crops such as oil palm, rubber, or cocoa, and other land. The authors say ForTy is the first global map of its kind. Earlier datasets, including one from the European Commission, cover only parts of the world.","The main driver behind the project is the EU Regulation on Deforestation-free Products (EUDR)：https://www.ble.de/DE/Themen/Wald-Holz/Entwaldungsfreie-Produkte/Was-ist-die-EUDR/EUDR-Info_node.html, which prohibits goods linked to forests cleared or degraded after December 31, 2020, including goods produced after primary forest was converted to a plantation. Oil palm and cocoa farms count as agricultural land, so companies assessing supply chain risks need to know how a site was used in 2020. Ad","ForTy gives them that 2020 reference point and could also feed into other initiatives that depend on the same kind of data, including the Tropical Forest Forever Facility (TFFF) and the UN goal of protecting 30 percent of the world's land by 2030. Ad","Building that reference point is hard because a forest's type depends on its history, not just how it looks, DeepMind says. An old planted mixed forest can look like a natural one from space, and a recently cleared plantation still counts as forest by definition even though satellite images show bare ground.","To distinguish these categories, ForTy runs on Google DeepMind's AlphaEarth Foundations model, which compresses data from multiple satellites into a compact description of each pixel. Several image recognition models build on those descriptions, starting with one that learns from verified examples. Ad","That first model's predictions, combined with outside data on tree heights, buildings, and forest loss, become expanded training data for a second model. A final step analyzes several years of data together to smooth outliers. Google's Gemini language model also helped adapt older training data to the EU's definitions.","During those correction rounds, experts added about 43,000 examples to address errors such as vineyards in Europe classified as tree crops, natural pine forests in Central America labeled as planted forests, and oil palms missed in Africa and Indonesia. Ad","Even with those corrections, some distinctions remain out of reach for satellites alone. Satellite images rarely show how land is managed, so the researchers use a 40-year rotation period to separate plantations from other planted forests, and that cutoff does not fit every region. Ad","To check the map, 2 experts reviewed each of 8,190 randomly selected sites, finding it correct in 90.2 percent of cases. The authors warn that this overall score gets a large boost from the easy-to-identify \"other land\" class.","Results vary sharply by forest type. The map correctly identified 89 percent of primary forests that experts confirmed on the ground but found only about half of tree crops and planted forests and 60 percent of plantations. Flipping the lens, roughly a quarter to a third of areas mapped as plantations or tree crops turned out to be misclassified, and the map was wrong for a third of areas labeled as planted forest.","These gaps hit the boundaries that matter most for the EUDR, and uncertainty in the evaluation may also make the results look slightly better than they are. Experts were most often unsure about managed areas, and the team excluded those uncertain cases from its assessment.","ForTy still outperforms European Commission maps made specifically for the EUDR on the same test data, scoring 95.5 percent compared with 93.6 percent for forest versus nonforest classification, and 90.9 percent compared with 87.8 percent for forest types.","The model also inherited a weakness from its training data, whose annotators rarely classified forests near roads or settlements as primary forest, even when the forests appeared untouched.","The researchers estimate that the world has about 4.1 billion hectares of forest, including nearly 3.9 billion hectares of natural forest, of which about 2 billion hectares are primary forest. Planted forests cover about 250 million hectares, and tree crops cover nearly 200 million hectares, mostly in Asia.","Regional errors complicate that picture. The map overestimates primary forest in Asia by about 200 million hectares, largely in Russia's conifer forests. In Africa, it instead misses about 94 million hectares of forest, likely because of open dry forests and savannas.","About two-thirds of Sweden's forest area is planted, and natural forest covers only around one-fifth of the country. The authors say overall tree cover there may be adequate, but the share of natural forest, which matters for biodiversity, could be critically low. The Democratic Republic of the Congo, Indonesia, and Brazil have the highest shares of natural forest.","Beyond these regional errors, some land uses remain difficult to classify. In some regions, the map still labels shaded cocoa or coffee grown beneath a natural canopy as forest, and selectively logged or burned forests also pose problems.","ForTy does not yet have a separate class for open woodlands, which the EUDR treats separately, though the researchers say future versions should add one. Users can make their own classification choices using the dataset's confidence scores for each class, which let them set their own thresholds.","The map：https://doi.org/10.25452/figshare.plus.32288616, classification layer：https://doi.org/10.25452/figshare.plus.32301177, and test data：https://doi.org/10.25452/figshare.plus.33093404 are freely available under a CC BY 4.0 license, including in Google Earth Engine. Google DeepMind has released the code in the GeeFlow：https://github.com/google-deepmind/geeflow and JEO：https://github.com/google-deepmind/jeo repositories.","Deepmind introduced the underlying AlphaEarth Foundations model in July 2025 as a kind of virtual satellite：https://the-decoder.com/google-deepmind-unveils-an-ai-model-that-acts-as-a-virtual-satellite-for-mapping-the-entire-planet/. It combines optical satellite images, radar, laser scans, and climate simulations into a compact description of each 10-by-10-meter area, with annual embeddings covering years since 2017 freely available in Google Earth Engine.","ForTy is one of the first large mapping projects built on those embeddings and, Google says, is part of Earth AI. That connection places the project within a broader effort, as Google connected Earth AI to Gemini：https://the-decoder.com/google-integrates-earth-ai-and-gemini-language-models-for-what-it-calls-geospatial-reasoning/ in fall 2025 to help detect floods, droughts, and algal blooms faster.","Stay in the loop on AI. Clear, useful, no fluff.","Follow The Decoder for AI news, background stories and expert analyses.","The Decoder：https://the-decoder.com/"],"articleImages":[{"sourceUrl":"https://the-decoder.com/wp-content/uploads/2026/10/forest-typology-01-hero.jpg","alt":"Image description","afterParagraph":0,"url":"/media/articles/cbmtamj96m73qxjf1nnhteiv9/bacb644da73ab56a.jpg"},{"sourceUrl":"https://the-decoder.com/wp-content/uploads/2026/10/forest-typology-02-world-map.jpg","alt":"The map shows the most common forest or land use class in each square kilometer, while the underlying data has a resolution of 10 meters.","afterParagraph":1,"url":"/media/articles/cbmtamj96m73qxjf1nnhteiv9/7ef23eb698154efd.jpg"},{"sourceUrl":"https://the-decoder.com/wp-content/uploads/2026/10/forest-typology-03-forest-classes.jpg","alt":"The model distinguishes the 5 visually similar forest and tree crop classes using planting patterns, growth cycles, and harvest schedules.","afterParagraph":4,"url":"/media/articles/cbmtamj96m73qxjf1nnhteiv9/7982802d599d2885.jpg"},{"sourceUrl":"https://the-decoder.com/wp-content/uploads/2026/10/forest-typology-05-training-pipeline.jpg","alt":"ForTy trains 3 models in sequence and improves its training data through several rounds of active learning focused on areas where the map made mistakes.","afterParagraph":6,"url":"/media/articles/cbmtamj96m73qxjf1nnhteiv9/abb4d3b47cb91bb6.jpg"},{"sourceUrl":"https://the-decoder.com/wp-content/uploads/2026/10/forest-typology-04-regional-hotspots.jpg","alt":"Tree crops and plantations dominate in Southeast Asia, while planted forests account for most of the forest area in Sweden and Finland.","afterParagraph":15,"url":"/media/articles/cbmtamj96m73qxjf1nnhteiv9/69fe63f9ade86b86.jpg"}],"mediaStatus":"ok","articleBodyZh":["谷歌DeepMind和世界资源研究所开发了ForTy，一种能够区分全球原始森林、木材种植园和树木作物农场的人工智能地图。","卫星地图通常只能显示某个地区是否有树木生长，但无法识别森林或作物类型。ForTy，全称Forest Typology（森林类型学），更进一步，将2020年的全球土地划分为6类，分辨率为10米。这些类别包括原始森林、自然再生林、人工林、种植园森林、如油棕、橡胶或可可等树木作物及其他土地。作者表示，ForTy是首个此类全球地图。早期的数据集，包括欧盟委员会的数据，仅覆盖世界的部分地区。","该项目背后的主要推动力是欧盟《无毁林产品条例》(EUDR)：https://www.ble.de/DE/Themen/Wald-Holz/Entwaldungsfreie-Produkte/Was-ist-die-EUDR/EUDR-Info_node.html，该条例禁止与2020年12月31日后被清理或退化的森林有关的商品，包括在原始森林转为种植园后生产的商品。油棕和可可农场被视为农业用地，因此评估供应链风险的公司需要了解地块在2020年的使用情况。","ForTy为他们提供了2020年的基准参考点，并且也可以为依赖相同类型数据的其他项目提供支持，包括热带森林永续基金（Tropical Forest Forever Facility, TFFF）以及联合国到2030年保护全球30%土地的目标。","DeepMind表示，建立这一基准点非常困难，因为森林类型取决于其历史，而不仅仅是外观。一片历史悠久的人工混交林从卫星上看可能像自然林，而一片新近清理的种植园即使卫星图像显示裸地，按定义仍然算作森林。","为了区分这些类别，ForTy运行在谷歌DeepMind的AlphaEarth Foundations模型上，该模型将来自多颗卫星的数据压缩为每个像素的紧凑描述。若干图像识别模型在这些描述的基础上构建，从第一个以已验证示例进行学习的模型开始。","第一个模型的预测结果，加上有关树木高度、建筑物和森林损失的外部数据，成为第二个模型的扩展训练数据。最后一步是将数年的数据一起分析，以平滑异常值。谷歌的Gemini语言模型也帮助将旧的训练数据适配到欧盟的定义。","在这些纠正轮次中，专家增加了约43,000个示例，以纠正诸如欧洲的葡萄园被归类为树木作物、中美洲的天然松树林被标记为人工林、非洲和印度尼西亚的油棕被遗漏等错误。","即使有这些修正，一些区分仍然仅凭卫星是无法达到的。卫星图像很少显示土地的管理方式，因此研究人员使用40年的轮作周期来区分种植园和其他人工林，而这个界限并不适用于每个地区。","为了检查地图，2名专家审核了随机选择的8,190个地点，发现其中90.2%的情况是正确的。作者警告说，这一总体得分很大程度上因容易识别的“其他土地”类别而被提升。","结果因森林类型而差异很大。地图正确识别了专家在实地确认的89%的原始森林，但仅发现大约一半的树木作物和人工林，以及60%的种植园。从另一个角度看，大约四分之一到三分之一的标记为种植园或树木作物的区域实际上被错误分类，而标记为人工林的区域中有三分之一是错误的。","这些差距涉及欧盟尽职调查条例（EUDR）最关心的边界，而且评估中的不确定性也可能使结果看起来略好于实际。专家最常对管理区域感到不确定，而团队在评估中排除了这些不确定情况。","ForTy在同一测试数据上仍然优于专门为EUDR制作的欧盟委员会地图，在森林与非森林分类上得分为95.5%，而欧盟委员会地图得分为93.6%；在森林类型上得分为90.9%，而欧盟委员会地图为87.8%。","该模型还继承了其训练数据的一个弱点，因为其标注员很少将靠近道路或居民点的森林分类为原始森林，即使这些森林看起来未受破坏。","研究人员估计，全球大约有41亿公顷森林，其中近39亿公顷为天然森林，大约20亿公顷为原始森林。人工造林覆盖约2.5亿公顷，树木作物覆盖近2亿公顷，主要集中在亚洲。","地区性误差使情况更复杂。该地图高估了亚洲的原始森林面积约2亿公顷，这主要发生在俄罗斯的针叶林。在非洲，地图反而遗漏了大约9400万公顷的森林，这可能是由于开阔干旱林和稀树草原的原因。","瑞典约三分之二的森林是人工种植的，天然森林仅覆盖全国约五分之一。作者表示，总体而言，该国树木覆盖可能足够，但天然森林的比例——对生物多样性至关重要——可能严重偏低。刚果民主共和国、印度尼西亚和巴西的天然森林比例最高。","除了这些地区性误差外，一些土地使用类别仍然难以分类。在某些地区，地图仍将生长在天然林冠下的遮荫可可或咖啡标记为森林，选择性砍伐或烧毁的森林也同样存在问题。","ForTy 还没有为开阔林地设置单独类别，而欧盟森林数据注册（EUDR）是单独分类的，尽管研究人员表示未来版本应增加这一类别。用户可以使用数据集提供的各类置信度分数自行进行分类选择，并设定自己的阈值。","地图：https://doi.org/10.25452/figshare.plus.32288616，分类图层：https://doi.org/10.25452/figshare.plus.32301177，测试数据：https://doi.org/10.25452/figshare.plus.33093404 可在 CC BY 4.0 许可下自由使用，包括在 Google Earth Engine 中。谷歌 DeepMind 已在 GeeFlow：https://github.com/google-deepmind/geeflow 和 JEO：https://github.com/google-deepmind/jeo 代码库中发布了代码。","Deepmind 在 2025 年 7 月推出了基础 AlphaEarth Foundations 模型，作为一种虚拟卫星：https://the-decoder.com/google-deepmind-unveils-an-ai-model-that-acts-as-a-virtual-satellite-for-mapping-the-entire-planet/。它将光学卫星图像、雷达、激光扫描和气候模拟结合起来，对每块 10×10 米的区域进行紧凑描述，并且从 2017 年开始的年度嵌入可以在 Google Earth Engine 免费获取。","ForTy 是基于这些嵌入构建的首批大型绘图项目之一，Google 表示，它是 Earth AI 的一部分。这种关联将该项目纳入更广泛的努力中，因为 Google 在 2025 年秋季将 Earth AI 与 Gemini 连接：https://the-decoder.com/google-integrates-earth-ai-and-gemini-language-models-for-what-it-calls-geospatial-reasoning/，以便更快地检测洪水、干旱和藻类暴发。","保持对 AI 的关注。清晰、有用，无废话。","关注 The Decoder 获取 AI 新闻、背景故事和专家分析。","解码器：https://the-decoder.com/"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Google DeepMind 与世界资源研究所联合开发 ForTy，将 2020 年全球土地以 10 米分辨率划分为六类森林、种植和其他土地类型，旨在提供与欧盟无毁林产品法规相关的基准地图。","background":"ForTy 试图区分原始森林、天然再生林、人工林、种植园、油棕橡胶可可等树木作物及其他土地。材料称其在 8190 个随机抽样点的专家复核中整体准确率为 90.2%，但树木作物和人工林的识别约为一半。","viewpoint":"Aioga 判断：ForTy 的价值不只在于提高地图分辨率，更在于把森林类型与 2020 年这一法规基准联系起来。不过，部分类别识别能力有限，意味着其更适合作为风险评估参考，而非单独结论。","implications":"可能影响：ForTy 可能为供应链地点核查、热带森林相关计划和土地保护目标提供基础数据，但类别误判仍可能影响判断；使用者需要结合专家复核及其他数据，不代表地图结果足以独立证明合规或违规。","nextStep":"后续观察：应关注 ForTy 在不同地区、树木作物和人工林类别上的验证结果，以及其与欧盟定义、企业供应链核查流程和其他土地数据的衔接方式。","evidenceRefs":["title","summary","articleBody","source"],"status":"published","aiGenerated":true,"autoApproved":true,"generatedBy":"aioga-editorial:gpt-5.6-sol","reviewedBy":"aioga-editorial-review:gpt-5.6-sol","generatedAt":"2026-10-31T12:06:16.522Z","sourceHash":"31041ab90df58a64","review":{"approved":true,"groundedness":91,"clarity":89,"duplicationRisk":35,"blockingIssues":[],"notes":["“Aioga 判断”是候选内容中的署名观点，来源材料未提供该署名；如需来源可追溯，建议确认署名。","“更适合作为风险评估参考”及“需要结合专家复核及其他数据”属于基于识别局限提出的建议，不是来源直接结论；当前已用判断性措辞表达。"]},"validation":{"passed":true,"mode":"ai-auto","revisions":0,"checks":["schema","length","source-attribution","editorial-labels","inference-boundary","low-source-overlap","no-html","independent-ai-review"]}},"tags":["行业动态","The Decoder：AI News"],"translations":{"zh-CN":{"title":"Google DeepMind 用卫星数据和 AI 绘制全球森林类型地图 ForTy","summary":"Google DeepMind 与世界资源研究所联合开发了 ForTy，这是首个全球森林类型地图，以 10 米分辨率将 2020 年全球土地划分为 6 类，包括原始森林、天然再生林、人工林、种植园、油棕橡胶可可等树木作物及其他土地。该地图主要为欧盟无毁林产品法规（EUDR）提供 2020 年基准，在 8190 个随机抽样点的专家复核中整体准确率为 90.2%，但对树木作物和人工林仅识别出约一半。","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind 用卫星数据和 AI 绘制全球森林类型地图 ForTy - Aioga AI资讯","description":"Google DeepMind 与世界资源研究所联合开发了 ForTy，这是首个全球森林类型地图，以 10 米分辨率将 2020 年全球土地划分为 6 类，包括原始森林、天然再生林、人工林、种植园、油棕橡胶可可等树木作物及其他土地。该地图主要为欧盟无毁林产品法规（EUDR）提供 2020 年基准，在 8190 个随机抽样点的专家复核中整体准确率为 90.2%...","url":"https://www.aioga.com/news/cbmtamj96m73qxjf1nnhteiv9/","articleBody":["谷歌DeepMind和世界资源研究所开发了ForTy，一种能够区分全球原始森林、木材种植园和树木作物农场的人工智能地图。","卫星地图通常只能显示某个地区是否有树木生长，但无法识别森林或作物类型。ForTy，全称Forest Typology（森林类型学），更进一步，将2020年的全球土地划分为6类，分辨率为10米。这些类别包括原始森林、自然再生林、人工林、种植园森林、如油棕、橡胶或可可等树木作物及其他土地。作者表示，ForTy是首个此类全球地图。早期的数据集，包括欧盟委员会的数据，仅覆盖世界的部分地区。","该项目背后的主要推动力是欧盟《无毁林产品条例》(EUDR)：https://www.ble.de/DE/Themen/Wald-Holz/Entwaldungsfreie-Produkte/Was-ist-die-EUDR/EUDR-Info_node.html，该条例禁止与2020年12月31日后被清理或退化的森林有关的商品，包括在原始森林转为种植园后生产的商品。油棕和可可农场被视为农业用地，因此评估供应链风险的公司需要了解地块在2020年的使用情况。","ForTy为他们提供了2020年的基准参考点，并且也可以为依赖相同类型数据的其他项目提供支持，包括热带森林永续基金（Tropical Forest Forever Facility, TFFF）以及联合国到2030年保护全球30%土地的目标。","DeepMind表示，建立这一基准点非常困难，因为森林类型取决于其历史，而不仅仅是外观。一片历史悠久的人工混交林从卫星上看可能像自然林，而一片新近清理的种植园即使卫星图像显示裸地，按定义仍然算作森林。","为了区分这些类别，ForTy运行在谷歌DeepMind的AlphaEarth Foundations模型上，该模型将来自多颗卫星的数据压缩为每个像素的紧凑描述。若干图像识别模型在这些描述的基础上构建，从第一个以已验证示例进行学习的模型开始。","第一个模型的预测结果，加上有关树木高度、建筑物和森林损失的外部数据，成为第二个模型的扩展训练数据。最后一步是将数年的数据一起分析，以平滑异常值。谷歌的Gemini语言模型也帮助将旧的训练数据适配到欧盟的定义。","在这些纠正轮次中，专家增加了约43,000个示例，以纠正诸如欧洲的葡萄园被归类为树木作物、中美洲的天然松树林被标记为人工林、非洲和印度尼西亚的油棕被遗漏等错误。","即使有这些修正，一些区分仍然仅凭卫星是无法达到的。卫星图像很少显示土地的管理方式，因此研究人员使用40年的轮作周期来区分种植园和其他人工林，而这个界限并不适用于每个地区。","为了检查地图，2名专家审核了随机选择的8,190个地点，发现其中90.2%的情况是正确的。作者警告说，这一总体得分很大程度上因容易识别的“其他土地”类别而被提升。","结果因森林类型而差异很大。地图正确识别了专家在实地确认的89%的原始森林，但仅发现大约一半的树木作物和人工林，以及60%的种植园。从另一个角度看，大约四分之一到三分之一的标记为种植园或树木作物的区域实际上被错误分类，而标记为人工林的区域中有三分之一是错误的。","这些差距涉及欧盟尽职调查条例（EUDR）最关心的边界，而且评估中的不确定性也可能使结果看起来略好于实际。专家最常对管理区域感到不确定，而团队在评估中排除了这些不确定情况。","ForTy在同一测试数据上仍然优于专门为EUDR制作的欧盟委员会地图，在森林与非森林分类上得分为95.5%，而欧盟委员会地图得分为93.6%；在森林类型上得分为90.9%，而欧盟委员会地图为87.8%。","该模型还继承了其训练数据的一个弱点，因为其标注员很少将靠近道路或居民点的森林分类为原始森林，即使这些森林看起来未受破坏。","研究人员估计，全球大约有41亿公顷森林，其中近39亿公顷为天然森林，大约20亿公顷为原始森林。人工造林覆盖约2.5亿公顷，树木作物覆盖近2亿公顷，主要集中在亚洲。","地区性误差使情况更复杂。该地图高估了亚洲的原始森林面积约2亿公顷，这主要发生在俄罗斯的针叶林。在非洲，地图反而遗漏了大约9400万公顷的森林，这可能是由于开阔干旱林和稀树草原的原因。","瑞典约三分之二的森林是人工种植的，天然森林仅覆盖全国约五分之一。作者表示，总体而言，该国树木覆盖可能足够，但天然森林的比例——对生物多样性至关重要——可能严重偏低。刚果民主共和国、印度尼西亚和巴西的天然森林比例最高。","除了这些地区性误差外，一些土地使用类别仍然难以分类。在某些地区，地图仍将生长在天然林冠下的遮荫可可或咖啡标记为森林，选择性砍伐或烧毁的森林也同样存在问题。","ForTy 还没有为开阔林地设置单独类别，而欧盟森林数据注册（EUDR）是单独分类的，尽管研究人员表示未来版本应增加这一类别。用户可以使用数据集提供的各类置信度分数自行进行分类选择，并设定自己的阈值。","地图：https://doi.org/10.25452/figshare.plus.32288616，分类图层：https://doi.org/10.25452/figshare.plus.32301177，测试数据：https://doi.org/10.25452/figshare.plus.33093404 可在 CC BY 4.0 许可下自由使用，包括在 Google Earth Engine 中。谷歌 DeepMind 已在 GeeFlow：https://github.com/google-deepmind/geeflow 和 JEO：https://github.com/google-deepmind/jeo 代码库中发布了代码。","Deepmind 在 2025 年 7 月推出了基础 AlphaEarth Foundations 模型，作为一种虚拟卫星：https://the-decoder.com/google-deepmind-unveils-an-ai-model-that-acts-as-a-virtual-satellite-for-mapping-the-entire-planet/。它将光学卫星图像、雷达、激光扫描和气候模拟结合起来，对每块 10×10 米的区域进行紧凑描述，并且从 2017 年开始的年度嵌入可以在 Google Earth Engine 免费获取。","ForTy 是基于这些嵌入构建的首批大型绘图项目之一，Google 表示，它是 Earth AI 的一部分。这种关联将该项目纳入更广泛的努力中，因为 Google 在 2025 年秋季将 Earth AI 与 Gemini 连接：https://the-decoder.com/google-integrates-earth-ai-and-gemini-language-models-for-what-it-calls-geospatial-reasoning/，以便更快地检测洪水、干旱和藻类暴发。","保持对 AI 的关注。清晰、有用，无废话。","关注 The Decoder 获取 AI 新闻、背景故事和专家分析。","解码器：https://the-decoder.com/"]},"en":{"title":"Google DeepMind Uses Satellite Data and AI to Map Global Forest Types with ForTy","summary":"Google DeepMind and the World Resources Institute jointly developed ForTy, the first global forest-type map. At a resolution of 10 meters, it classifies global land in 2020 into six categories, including primary forests, naturally regenerating forests, managed forests, plantations, tree crops such as oil palm, rubber, and cocoa, and other land. The map was primarily created to provide a 2020 baseline for the European Union Deforestation Regulation (EUDR). In expert reviews of 8,190 randomly sampled points, its overall accuracy was 90.2%, but it identified only about half of tree crops and plantations.","category":"Industry","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind Uses Satellite Data and AI to Map Global Forest Types with ForTy - Aioga AI News","description":"Google DeepMind and the World Resources Institute jointly developed ForTy, the first global forest-type map. At a resolution of 10 meters, it classifies global land in 2020 into si...","url":"https://www.aioga.com/en/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:02.510Z"},"ja":{"title":"Google DeepMind、衛星データとAIで世界の森林タイプマップ「ForTy」を作成","summary":"Google DeepMindは世界資源研究所と共同でForTyを開発しました。これは初の世界森林タイプマップで、10メートル解像度で2020年の世界の陸地を原生林、天然再生林、人工林、プランテーション、油パーム・ゴム・カカオなどの樹木作物、その他の土地の6種類に分類しています。このマップは主にEUの違法伐採製品規制（EUDR）のための2020年基準を提供し、8190のランダムサンプリングポイントでの専門家レビューで全体精度は90.2％ですが、樹木作物と人工林の識別は半分程度にとどまります。","category":"業界動向","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind、衛星データとAIで世界の森林タイプマップ「ForTy」を作成 - Aioga AIニュース","description":"Google DeepMindは世界資源研究所と共同でForTyを開発しました。これは初の世界森林タイプマップで、10メートル解像度で2020年の世界の陸地を原生林、天然再生林、人工林、プランテーション、油パーム・ゴム・カカオなどの樹木作物、その他の土地の6種類に分類しています。このマップは主にEUの違法伐採製品規制（EUDR）のための2020年基準を提供し...","url":"https://www.aioga.com/ja/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:17.395Z"},"ko":{"title":"Google DeepMind, 위성 데이터와 AI로 글로벌 산림 유형 지도 ForTy 제작","summary":"Google DeepMind가 세계자원연구소와 공동 개발한 ForTy는 최초의 글로벌 산림 유형 지도이며, 10m 해상도로 2020년 전 세계 토지를 6개 유형으로 분류합니다. 원시림, 자연 재생림, 인공림, 조림지, 야자·고무·코코아 등 나무 작물 및 기타 토지가 포함됩니다. 본 지도는 유럽연합 탈림 제품 규정(EUDR) 2020년 기준 제공을 목적으로 하며, 8190개의 무작위 샘플 지점 전문가 검토에서 전체 정확도는 90.2%였지만, 나무 작물과 인공림은 약 절반만 식별되었습니다.","category":"업계 동향","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind, 위성 데이터와 AI로 글로벌 산림 유형 지도 ForTy 제작 - Aioga AI 뉴스","description":"Google DeepMind가 세계자원연구소와 공동 개발한 ForTy는 최초의 글로벌 산림 유형 지도이며, 10m 해상도로 2020년 전 세계 토지를 6개 유형으로 분류합니다. 원시림, 자연 재생림, 인공림, 조림지, 야자·고무·코코아 등 나무 작물 및 기타 토지가 포함됩니다. 본 지도는 유럽연합 탈림 제품 규정(EUDR...","url":"https://www.aioga.com/ko/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:16.107Z"},"es":{"title":"Google DeepMind crea el mapa de tipos de bosques globales ForTy usando datos satelitales y IA","summary":"Google DeepMind y el World Resources Institute desarrollaron conjuntamente ForTy, el primer mapa global de tipos de bosques, que divide la tierra a nivel mundial en 2020 en 6 categorías con resolución de 10 metros, incluyendo bosques primarios, bosque regenerado natural, bosques plantados, plantaciones, cultivos arbóreos de palma, caucho y cacao, y otros tipos de tierra. Este mapa principalmente proporciona la referencia de 2020 para la regulación de productos sin deforestación de la UE (EUDR); en 8,190 puntos de muestreo aleatorio revisados por expertos, la precisión global fue del 90,2 %, aunque solo identificó aproximadamente la mitad de los cultivos arbóreos y bosques plantados.","category":"Industria","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind crea el mapa de tipos de bosques globales ForTy usando datos satelitales y IA - Aioga Noticias de IA","description":"Google DeepMind y el World Resources Institute desarrollaron conjuntamente ForTy, el primer mapa global de tipos de bosques, que divide la tierra a nivel mundial en 2020 en 6 categ...","url":"https://www.aioga.com/es/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:39.030Z"},"fr":{"title":"Google DeepMind crée la carte mondiale des types de forêts ForTy en utilisant des données satellites et l'IA","summary":"Google DeepMind et le World Resources Institute ont développé ForTy, la première carte mondiale des types de forêts, qui divise les terres mondiales de 2020 en 6 catégories à une résolution de 10 mètres, y compris forêt primaire, forêt naturelle régénérée, forêt artificielle, plantations, cultures d’arbres comme palmier à huile, caoutchouc, cacao et autres terres. Cette carte fournit principalement la référence 2020 pour le règlement européen sur les produits sans déforestation (EUDR). Lors de la vérification par des experts sur 8 190 points échantillonnés aléatoirement, la précision globale est de 90,2 %, mais seulement environ la moitié des cultures arboricoles et des forêts artificielles sont correctement identifiées.","category":"Industrie","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind crée la carte mondiale des types de forêts ForTy en utilisant des données satellites et l'IA - Aioga Actualités IA","description":"Google DeepMind et le World Resources Institute ont développé ForTy, la première carte mondiale des types de forêts, qui divise les terres mondiales de 2020 en 6 catégories à une r...","url":"https://www.aioga.com/fr/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:39.265Z"},"de":{"title":"Google DeepMind erstellt globale Waldtyp-Karten \"ForTy\" mit Satellitendaten und KI","summary":"Google DeepMind hat zusammen mit dem World Resources Institute „ForTy“ entwickelt, die erste globale Waldtyp-Karte. Sie unterteilt das globale Land von 2020 mit 10-Meter-Auflösung in sechs Kategorien: Primärwald, natürlich wiederkehrender Wald, Plantagenwald, Plantagen, Baumkulturen für Palmöl, Kautschuk, Kakao und andere Nutzpflanzen sowie sonstiges Land. Die Karte dient hauptsächlich als 2020-Basis für die EU-Verordnung zu entwaldungsfreien Produkten (EUDR). Bei der Expertenkontrolle von 8190 zufällig gewählten Stichprobenpunkten wurde insgesamt eine Genauigkeit von 90,2 % erreicht, bei Baumkulturen und Plantagenwald jedoch nur etwa die Hälfte erkannt.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind erstellt globale Waldtyp-Karten \"ForTy\" mit Satellitendaten und KI - Aioga KI-News","description":"Google DeepMind hat zusammen mit dem World Resources Institute „ForTy“ entwickelt, die erste globale Waldtyp-Karte. Sie unterteilt das globale Land von 2020 mit 10-Meter-Auflösung...","url":"https://www.aioga.com/de/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:37.598Z"},"pt-BR":{"title":"Google DeepMind cria mapa global de tipos de florestas ForTy usando dados de satélite e IA","summary":"Google DeepMind, em parceria com o World Resources Institute, desenvolveu o ForTy, o primeiro mapa global de tipos de floresta, que divide a terra global de 2020 em 6 categorias com resolução de 10 metros, incluindo florestas primárias, florestas regeneradas naturalmente, plantações artificiais, plantações, culturas arbóreas como óleo de palma, borracha e cacau, e outras terras. O mapa serve principalmente como base para o Regulamento da UE sobre Produtos sem Desflorestação (EUDR) de 2020. A acurácia geral, verificada em 8190 pontos amostrais aleatórios por especialistas, é de 90,2%, mas apenas cerca de metade das plantações arbóreas e florestas artificiais foram corretamente identificadas.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind cria mapa global de tipos de florestas ForTy usando dados de satélite e IA - Aioga Notícias de IA","description":"Google DeepMind, em parceria com o World Resources Institute, desenvolveu o ForTy, o primeiro mapa global de tipos de floresta, que divide a terra global de 2020 em 6 categorias co...","url":"https://www.aioga.com/pt-BR/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:00.202Z"},"ru":{"title":"Google DeepMind с помощью спутниковых данных и AI создала глобальную карту типов лесов ForTy","summary":"Google DeepMind совместно с Институтом мировых ресурсов разработала ForTy — первую глобальную карту типов лесов, которая с разрешением 10 метров классифицирует земную поверхность планеты в 2020 году по 6 категориям, включая первичные леса, естественно возобновляющиеся леса, искусственные леса, плантации, древесные культуры, такие как масличная пальма, каучук и какао, а также другие земли. Карта в первую очередь предназначена для формирования базового уровня за 2020 год в рамках Регламента ЕС о продукции, не связанной с вырубкой лесов (EUDR). Экспертная проверка 8190 случайно выбранных точек показала общую точность 90.2%, однако для древесных культур и искусственных лесов удалось распознать лишь около половины объектов.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind с помощью спутниковых данных и AI создала глобальную карту типов лесов ForTy - Aioga Новости ИИ","description":"Google DeepMind совместно с Институтом мировых ресурсов разработала ForTy — первую глобальную карту типов лесов, которая с разрешением 10 метров классифицирует земную поверхность п...","url":"https://www.aioga.com/ru/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:49.564Z"},"ar":{"title":"Google DeepMind ترسم خريطة عالمية لأنواع الغابات باستخدام بيانات الأقمار الصناعية والذكاء الاصطناعي: ForTy","summary":"طورت Google DeepMind بالتعاون مع معهد الموارد العالمية ForTy، وهي أول خريطة عالمية لأنواع الغابات، تقسم الأراضي العالمية في عام 2020 إلى 6 فئات بدقة 10 أمتار، تشمل الغابات الأولية، والغابات المتجددة طبيعيًا، والغابات المزروعة، والمزارع، والمحاصيل الشجرية مثل زيت النخيل والمطاط والكاكاو، والأراضي الأخرى. وتهدف الخريطة أساسًا إلى توفير خط أساس لعام 2020 للائحة الاتحاد الأوروبي للمنتجات الخالية من إزالة الغابات (EUDR). وفي مراجعة خبراء شملت 8190 نقطة عشوائية، بلغت الدقة الإجمالية 90.2%، لكنها لم تتعرف إلا على نحو نصف المحاصيل الشجرية والغابات المزروعة.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind ترسم خريطة عالمية لأنواع الغابات باستخدام بيانات الأقمار الصناعية والذكاء الاصطناعي: ForTy - Aioga أخبار الذكاء الاصطناعي","description":"طورت Google DeepMind بالتعاون مع معهد الموارد العالمية ForTy، وهي أول خريطة عالمية لأنواع الغابات، تقسم الأراضي العالمية في عام 2020 إلى 6 فئات بدقة 10 أمتار، تشمل الغابات الأولية،...","url":"https://www.aioga.com/ar/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:02:48.229Z"},"hi":{"title":"Google DeepMind ने उपग्रह डेटा और AI का उपयोग करके वैश्विक वन प्रकार का मानचित्र ForTy बनाया","summary":"Google DeepMind ने विश्व संसाधन संस्थान के साथ मिलकर ForTy विकसित किया, जो पहला वैश्विक वन प्रकार का मानचित्र है, जो 10 मीटर रिज़ॉल्यूशन में 2020 की वैश्विक भूमि को 6 श्रेणियों में विभाजित करता है, जिसमें मूल जंगल, प्राकृतिक पुनर्जनित जंगल, कृत्रिम जंगल, प्लांटेशन, पाम ऑयल, रबर, कोको जैसे वृक्षीय फसल और अन्य भूमि शामिल हैं। यह मानचित्र मुख्य रूप से यूरोपीय संघ के वन-नाश उत्पादन विनियम (EUDR) के लिए 2020 की आधार रेखा प्रदान करता है, और 8190 यादृच्छिक नमूना बिंदुओं पर विशेषज्ञ समीक्षा में कुल मिलाकर 90.2% सटीकता है, लेकिन वृक्षीय फसल और कृत्रिम जंगल में केवल लगभग आधी पहचान कर पाता है।","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind ने उपग्रह डेटा और AI का उपयोग करके वैश्विक वन प्रकार का मानचित्र ForTy बनाया - Aioga AI समाचार","description":"Google DeepMind ने विश्व संसाधन संस्थान के साथ मिलकर ForTy विकसित किया, जो पहला वैश्विक वन प्रकार का मानचित्र है, जो 10 मीटर रिज़ॉल्यूशन में 2020 की वैश्विक भूमि को 6 श्रेणियों में...","url":"https://www.aioga.com/hi/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:22.293Z"},"it":{"title":"Google DeepMind crea la mappa dei tipi di foresta globale ForTy con dati satellitari e AI","summary":"Google DeepMind, in collaborazione con il World Resources Institute, ha sviluppato ForTy, la prima mappa globale dei tipi di foresta, che suddivide la terra del 2020 in sei categorie a risoluzione di 10 metri, tra cui foreste primarie, foreste naturali rigenerate, foreste artificiali, piantagioni, colture arboree come palma da olio, gomma, cacao e altre terre. La mappa fornisce principalmente il riferimento 2020 per la normativa europea sui prodotti senza deforestazione (EUDR), con una precisione complessiva del 90,2% su 8190 punti di campionamento casuali verificati da esperti, ma identifica solo circa metà delle colture arboree e delle foreste artificiali.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind crea la mappa dei tipi di foresta globale ForTy con dati satellitari e AI - Aioga Notizie IA","description":"Google DeepMind, in collaborazione con il World Resources Institute, ha sviluppato ForTy, la prima mappa globale dei tipi di foresta, che suddivide la terra del 2020 in sei categor...","url":"https://www.aioga.com/it/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:21.969Z"},"nl":{"title":"Google DeepMind brengt met satellietgegevens en AI de wereldwijde bostypen in kaart met ForTy","summary":"Google DeepMind heeft samen met het World Resources Institute ForTy ontwikkeld, de eerste wereldwijde kaart van bostypen, die het wereldwijde landoppervlak van 2020 met een resolutie van 10 meter in 6 categorieën indeelt, waaronder oerbossen, natuurlijk regenererende bossen, aangeplante bossen, plantages, boomgewassen zoals oliepalmen, rubber en cacao, en overige gronden. De kaart is voornamelijk bedoeld als referentie voor 2020 in het kader van de EU-verordening inzake ontbossingsvrije producten (EUDR). Bij deskundige controle van 8190 willekeurig geselecteerde steekproefpunten bedroeg de algemene nauwkeurigheid 90,2%, maar van boomgewassen en aangeplante bossen werd slechts ongeveer de helft correct herkend.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind brengt met satellietgegevens en AI de wereldwijde bostypen in kaart met ForTy - Aioga AI-nieuws","description":"Google DeepMind heeft samen met het World Resources Institute ForTy ontwikkeld, de eerste wereldwijde kaart van bostypen, die het wereldwijde landoppervlak van 2020 met een resolut...","url":"https://www.aioga.com/nl/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:09.030Z"},"tr":{"title":"Google DeepMind, uydu verileri ve AI kullanarak küresel orman türleri haritası ForTy'yi oluşturdu","summary":"Google DeepMind, Dünya Kaynakları Enstitüsü ile iş birliği yaparak ForTy'yi geliştirdi; bu, dünya ölçeğinde orman türleri haritası olup, 2020 yılındaki küresel araziyi 10 metre çözünürlükle 6 kategoriye ayırır: doğal orman, doğal yeniden büyüyen orman, yapay orman, plantasyon, palmiye, kauçuk, kakao gibi ağaç ürünleri ve diğer arazi türleri. Bu harita özellikle AB Kesintisiz Ormansız Ürün Yönetmeliği (EUDR) için 2020 temel veri sağlar; 8190 rastgele örnek noktasında uzman denetiminde genel doğruluk oranı %90,2 iken, ağaç ürünleri ve yapay ormanlarda yalnızca yaklaşık yarısı doğru tanımlanmıştır.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind, uydu verileri ve AI kullanarak küresel orman türleri haritası ForTy'yi oluşturdu - Aioga AI Haberleri","description":"Google DeepMind, Dünya Kaynakları Enstitüsü ile iş birliği yaparak ForTy'yi geliştirdi; bu, dünya ölçeğinde orman türleri haritası olup, 2020 yılındaki küresel araziyi 10 metre çöz...","url":"https://www.aioga.com/tr/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:42.343Z"},"vi":{"title":"Google DeepMind sử dụng dữ liệu vệ tinh và AI để lập bản đồ các loại rừng toàn cầu ForTy","summary":"Google DeepMind phối hợp với Viện Tài nguyên Thế giới phát triển ForTy, bản đồ các loại rừng toàn cầu đầu tiên, phân loại đất đai trên toàn thế giới vào năm 2020 thành 6 loại với độ phân giải 10 mét, gồm rừng nguyên sinh, rừng tái sinh tự nhiên, rừng trồng, đồn điền, cây trồng thân gỗ như cọ dầu, cao su và ca cao, cùng các loại đất khác. Bản đồ này chủ yếu cung cấp đường cơ sở năm 2020 cho Quy định của EU về các sản phẩm không gây mất rừng (EUDR). Trong quá trình chuyên gia đánh giá lại 8190 điểm lấy mẫu ngẫu nhiên, độ chính xác tổng thể đạt 90.2%, nhưng bản đồ chỉ nhận diện được khoảng một nửa số cây trồng thân gỗ và rừng trồng.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind sử dụng dữ liệu vệ tinh và AI để lập bản đồ các loại rừng toàn cầu ForTy - Tin tức AI Aioga","description":"Google DeepMind phối hợp với Viện Tài nguyên Thế giới phát triển ForTy, bản đồ các loại rừng toàn cầu đầu tiên, phân loại đất đai trên toàn thế giới vào năm 2020 thành 6 loại với đ...","url":"https://www.aioga.com/vi/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:36.090Z"},"id":{"title":"Google DeepMind menggunakan data satelit dan AI untuk membuat peta jenis hutan global ForTy","summary":"Google DeepMind bekerja sama dengan World Resources Institute mengembangkan ForTy, peta hutan global pertama yang membagi lahan dunia 2020 menjadi 6 kategori dengan resolusi 10 meter, termasuk hutan primer, hutan alami yang beregenerasi, hutan buatan, perkebunan, tanaman kayu seperti kelapa sawit, karet, kakao, dan lahan lainnya. Peta ini terutama untuk menyediakan garis dasar 2020 untuk regulasi produk bebas deforestasi Uni Eropa (EUDR), dengan akurasi keseluruhan 90,2% dari 8190 titik sampel acak yang diverifikasi oleh ahli, namun hanya mengenali sekitar setengah dari tanaman kayu dan hutan buatan.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind menggunakan data satelit dan AI untuk membuat peta jenis hutan global ForTy - Berita AI Aioga","description":"Google DeepMind bekerja sama dengan World Resources Institute mengembangkan ForTy, peta hutan global pertama yang membagi lahan dunia 2020 menjadi 6 kategori dengan resolusi 10 met...","url":"https://www.aioga.com/id/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:43.334Z"},"th":{"title":"Google DeepMind ใช้ข้อมูลดาวเทียมและ AI จัดทำแผนที่ประเภทป่าทั่วโลก ForTy","summary":"Google DeepMind ร่วมกับสถาบันทรัพยากรโลกพัฒนา ForTy ซึ่งเป็นแผนที่ประเภทป่าทั่วโลกฉบับแรก โดยแบ่งพื้นที่ทั่วโลกในปี 2020 ออกเป็น 6 ประเภทด้วยความละเอียด 10 เมตร ได้แก่ ป่าดั้งเดิม ป่าที่ฟื้นฟูตามธรรมชาติ ป่าปลูก สวนเพาะปลูก พืชต้นไม้ เช่น ปาล์มน้ำมัน ยางพารา และโกโก้ รวมถึงพื้นที่อื่น ๆ แผนที่นี้จัดทำขึ้น בעיקרเพื่อใช้เป็นข้อมูลอ้างอิงปี 2020 สำหรับกฎระเบียบสินค้าปลอดการตัดไม้ทำลายป่าของสหภาพยุโรป (EUDR) โดยการตรวจสอบโดยผู้เชี่ยวชาญจากจุดสุ่มตัวอย่าง 8190 จุดพบว่ามีความแม่นยำโดยรวม 90.2% แต่สามารถจำแนกพืชต้นไม้และป่าปลูกได้เพียงประมาณครึ่งหนึ่งเท่านั้น","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind ใช้ข้อมูลดาวเทียมและ AI จัดทำแผนที่ประเภทป่าทั่วโลก ForTy - ข่าว AI Aioga","description":"Google DeepMind ร่วมกับสถาบันทรัพยากรโลกพัฒนา ForTy ซึ่งเป็นแผนที่ประเภทป่าทั่วโลกฉบับแรก โดยแบ่งพื้นที่ทั่วโลกในปี 2020 ออกเป็น 6 ประเภทด้วยความละเอียด 10 เมตร ได้แก่ ป่าดั้งเดิม...","url":"https://www.aioga.com/th/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:53.933Z"},"pl":{"title":"Google DeepMind wykorzystuje dane satelitarne i AI do stworzenia globalnej mapy typów lasów ForTy","summary":"Google DeepMind wraz z World Resources Institute opracował ForTy — pierwszą globalną mapę typów lasów, która z rozdzielczością 10 metrów dzieli światowe obszary lądowe z 2020 roku na 6 kategorii: lasy pierwotne, lasy odradzające się naturalnie, lasy sadzone, plantacje, uprawy drzewiaste, takie jak palma olejowa, kauczukowiec i kakaowiec, oraz pozostałe grunty. Mapa ma służyć przede wszystkim jako poziom odniesienia z 2020 roku na potrzeby unijnego rozporządzenia w sprawie produktów niepowodujących wylesiania (EUDR). Podczas eksperckiej weryfikacji 8190 losowo wybranych punktów jej ogólna dokładność wyniosła 90,2%, jednak prawidłowo rozpoznano jedynie około połowy upraw drzewiastych i lasów sadzonych.","category":"行业动态","source":"The Decoder：AI News","aggregationSource":"The Decoder：AI News","pageTitle":"Google DeepMind wykorzystuje dane satelitarne i AI do stworzenia globalnej mapy typów lasów ForTy - Aioga Wiadomości AI","description":"Google DeepMind wraz z World Resources Institute opracował ForTy — pierwszą globalną mapę typów lasów, która z rozdzielczością 10 metrów dzieli światowe obszary lądowe z 2020 roku...","url":"https://www.aioga.com/pl/news/cbmtamj96m73qxjf1nnhteiv9/","contentTranslated":true,"translationStatus":"translated","translationRetryAt":"","translationError":"","sourceHash":"79b879069739ee1c","translatedAt":"2026-10-31T12:03:57.578Z"}},"evidenceTier":"verified-news","reviewStatus":"automated-ingest","indexable":true,"editorialCover":""}}