공지사항:
2025년 4월 15일 전에 Earth Engine 사용을 위해 등록된 모든 비상업용 프로젝트는 Earth Engine 액세스를 유지하기 위해
비상업용 자격 요건을 인증해야 합니다.
ee.Algorithms.Image.Segmentation.SNIC
컬렉션을 사용해 정리하기
내 환경설정을 기준으로 콘텐츠를 저장하고 분류하세요.
SNIC (Simple Non-Iterative Clustering) 기반 슈퍼픽셀 클러스터링 클러스터 ID와 각 입력 밴드의 클러스터별 평균을 출력합니다. '시드' 이미지가 입력으로 제공되지 않으면 생성된 시드 위치가 포함된 '시드' 밴드가 출력에 포함됩니다. Achanta, Radhakrishna, Susstrunk, Sabine, 'Superpixels and Polygons using Simple Non-Iterative Clustering', CVPR, 2017 참고
사용 | 반환 값 |
---|
ee.Algorithms.Image.Segmentation.SNIC(image, size, compactness, connectivity, neighborhoodSize, seeds) | 이미지 |
인수 | 유형 | 세부정보 |
---|
image | 이미지 | 클러스터링할 입력 이미지입니다. |
size | 정수, 기본값: 5 | 슈퍼픽셀 시드 위치 간격(픽셀)입니다. '시드' 이미지가 제공되면 그리드가 생성되지 않습니다. |
compactness | 부동 소수점 수, 기본값: 1 | 간결성 요소입니다. 값이 클수록 클러스터가 더 콤팩트해집니다 (사각형). 이 값을 0으로 설정하면 공간 거리 가중치가 사용 중지됩니다. |
connectivity | 정수, 기본값: 8 | 연결을 탭합니다. 4 또는 8입니다. |
neighborhoodSize | 정수, 기본값: null | 타일 인접 크기 (타일 경계 아티팩트 방지) 기본값은 2 * 크기입니다. |
seeds | 이미지, 기본값: null | 제공된 경우 0이 아닌 값의 픽셀이 시드 위치로 사용됩니다. '연결성'에 따라 서로 닿는 픽셀은 동일한 클러스터에 속하는 것으로 간주됩니다. |
예
코드 편집기 (JavaScript)
// Note that the compactness and size parameters can have a significant impact
// on the result. They must be adjusted to meet image-specific characteristics
// and patterns, typically through trial. Pixel scale (map zoom level) is also
// important to consider. When exploring interactively through map tile
// visualization, the segmentation result it dependent on zoom level. If you
// need to evaluate the result at a specific scale, call .reproject() on the
// result, but do so with caution because it overrides the default scaling
// behavior that makes tile computation fast and efficient.
// Load a NAIP image for a neighborhood in Las Vegas.
var naip = ee.Image('USDA/NAIP/DOQQ/m_3611554_sw_11_1_20170613');
// Apply the SNIC algorithm to the image.
var snic = ee.Algorithms.Image.Segmentation.SNIC({
image: naip,
size: 30,
compactness: 0.1,
connectivity: 8,
});
// Display the original NAIP image as RGB.
// Lock map zoom to maintain the desired scale of the segmentation computation.
Map.setLocked(false, 18, 18);
Map.setCenter(-115.32053, 36.182016, 18);
Map.addLayer(naip, null, 'NAIP RGB');
// Display the clusters.
Map.addLayer(snic.randomVisualizer(), null, 'Clusters');
// Display the RGB cluster means.
var visParams = {
bands: ['R_mean', 'G_mean', 'B_mean'],
min: 0,
max: 255
};
Map.addLayer(snic, visParams, 'RGB cluster means');
Python 설정
Python API 및 geemap
를 사용한 대화형 개발에 관한 자세한 내용은
Python 환경 페이지를 참고하세요.
import ee
import geemap.core as geemap
Colab (Python)
# Note that the compactness and size parameters can have a significant impact
# on the result. They must be adjusted to meet image-specific characteristics
# and patterns, typically through trial. Pixel scale (map zoom level) is also
# important to consider. When exploring interactively through map tile
# visualization, the segmentation result it dependent on zoom level. If you
# need to evaluate the result at a specific scale, call .reproject() on the
# result, but do so with caution because it overrides the default scaling
# behavior that makes tile computation fast and efficient.
# Load a NAIP image for a neighborhood in Las Vegas.
naip = ee.Image('USDA/NAIP/DOQQ/m_3611554_sw_11_1_20170613')
# Apply the SNIC algorithm to the image.
snic = ee.Algorithms.Image.Segmentation.SNIC(
image=naip, size=30, compactness=0.1, connectivity=8
)
# Display the original NAIP image as RGB.
m = geemap.Map()
m.set_center(-115.32053, 36.182016, 18)
m.add_layer(naip, None, 'NAIP RGB')
# Display the clusters.
m.add_layer(snic.randomVisualizer(), None, 'Clusters')
# Display the RGB cluster means.
vis_params = {'bands': ['R_mean', 'G_mean', 'B_mean'], 'min': 0, 'max': 255}
m.add_layer(snic, vis_params, 'RGB cluster means')
m
달리 명시되지 않는 한 이 페이지의 콘텐츠에는 Creative Commons Attribution 4.0 라이선스에 따라 라이선스가 부여되며, 코드 샘플에는 Apache 2.0 라이선스에 따라 라이선스가 부여됩니다. 자세한 내용은 Google Developers 사이트 정책을 참조하세요. 자바는 Oracle 및/또는 Oracle 계열사의 등록 상표입니다.
최종 업데이트: 2025-07-26(UTC)
[null,null,["최종 업데이트: 2025-07-26(UTC)"],[[["\u003cp\u003eApplies the Simple Non-Iterative Clustering (SNIC) algorithm to generate superpixels from an image.\u003c/p\u003e\n"],["\u003cp\u003eOutputs an image containing cluster IDs and per-cluster band averages.\u003c/p\u003e\n"],["\u003cp\u003eAlgorithm parameters like \u003ccode\u003esize\u003c/code\u003e and \u003ccode\u003ecompactness\u003c/code\u003e impact the resulting superpixel shapes and sizes and may require adjustments.\u003c/p\u003e\n"],["\u003cp\u003eIf no seed locations are provided, the algorithm generates them based on a grid defined by the \u003ccode\u003esize\u003c/code\u003e parameter.\u003c/p\u003e\n"],["\u003cp\u003eSuperpixel clustering is sensitive to pixel scale and zoom level during interactive visualization.\u003c/p\u003e\n"]]],["SNIC clustering segments an image into superpixels, outputting cluster IDs and per-cluster averages for each input band. Key parameters include `size` (seed spacing), `compactness` (cluster shape), and `connectivity`. A user can provide `seeds` to define seed locations; otherwise, they are generated. The output `Image` includes cluster IDs, band averages, and optionally generated seed locations. Adjusting `size` and `compactness` is crucial for optimal results, which are also affected by pixel scale.\n"],null,["# ee.Algorithms.Image.Segmentation.SNIC\n\nSuperpixel clustering based on SNIC (Simple Non-Iterative Clustering). Outputs a band of cluster IDs and the per-cluster averages for each of the input bands. If the 'seeds' image isn't provided as input, the output will include a 'seeds' band containing the generated seed locations. See: Achanta, Radhakrishna and Susstrunk, Sabine, 'Superpixels and Polygons using Simple Non-Iterative Clustering', CVPR, 2017.\n\n\u003cbr /\u003e\n\n| Usage | Returns |\n|----------------------------------------------------------------------------------------------------------------------------------|---------|\n| `ee.Algorithms.Image.Segmentation.SNIC(image, `*size* `, `*compactness* `, `*connectivity* `, `*neighborhoodSize* `, `*seeds*`)` | Image |\n\n| Argument | Type | Details |\n|--------------------|------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| `image` | Image | The input image for clustering. |\n| `size` | Integer, default: 5 | The superpixel seed location spacing, in pixels. If 'seeds' image is provided, no grid is produced. |\n| `compactness` | Float, default: 1 | Compactness factor. Larger values cause clusters to be more compact (square). Setting this to 0 disables spatial distance weighting. |\n| `connectivity` | Integer, default: 8 | Connectivity. Either 4 or 8. |\n| `neighborhoodSize` | Integer, default: null | Tile neighborhood size (to avoid tile boundary artifacts). Defaults to 2 \\* size. |\n| `seeds` | Image, default: null | If provided, any non-zero valued pixels are used as seed locations. Pixels that touch (as specified by 'connectivity') are considered to belong to the same cluster. |\n\nExamples\n--------\n\n### Code Editor (JavaScript)\n\n```javascript\n// Note that the compactness and size parameters can have a significant impact\n// on the result. They must be adjusted to meet image-specific characteristics\n// and patterns, typically through trial. Pixel scale (map zoom level) is also\n// important to consider. When exploring interactively through map tile\n// visualization, the segmentation result it dependent on zoom level. If you\n// need to evaluate the result at a specific scale, call .reproject() on the\n// result, but do so with caution because it overrides the default scaling\n// behavior that makes tile computation fast and efficient.\n\n\n// Load a NAIP image for a neighborhood in Las Vegas.\nvar naip = ee.Image('USDA/NAIP/DOQQ/m_3611554_sw_11_1_20170613');\n\n// Apply the SNIC algorithm to the image.\nvar snic = ee.Algorithms.Image.Segmentation.SNIC({\n image: naip,\n size: 30,\n compactness: 0.1,\n connectivity: 8,\n});\n\n// Display the original NAIP image as RGB.\n// Lock map zoom to maintain the desired scale of the segmentation computation.\nMap.setLocked(false, 18, 18);\nMap.setCenter(-115.32053, 36.182016, 18);\nMap.addLayer(naip, null, 'NAIP RGB');\n\n// Display the clusters.\nMap.addLayer(snic.randomVisualizer(), null, 'Clusters');\n\n// Display the RGB cluster means.\nvar visParams = {\n bands: ['R_mean', 'G_mean', 'B_mean'],\n min: 0,\n max: 255\n};\nMap.addLayer(snic, visParams, 'RGB cluster means');\n```\nPython setup\n\nSee the [Python Environment](/earth-engine/guides/python_install) page for information on the Python API and using\n`geemap` for interactive development. \n\n```python\nimport ee\nimport geemap.core as geemap\n```\n\n### Colab (Python)\n\n```python\n# Note that the compactness and size parameters can have a significant impact\n# on the result. They must be adjusted to meet image-specific characteristics\n# and patterns, typically through trial. Pixel scale (map zoom level) is also\n# important to consider. When exploring interactively through map tile\n# visualization, the segmentation result it dependent on zoom level. If you\n# need to evaluate the result at a specific scale, call .reproject() on the\n# result, but do so with caution because it overrides the default scaling\n# behavior that makes tile computation fast and efficient.\n\n\n# Load a NAIP image for a neighborhood in Las Vegas.\nnaip = ee.Image('USDA/NAIP/DOQQ/m_3611554_sw_11_1_20170613')\n\n# Apply the SNIC algorithm to the image.\nsnic = ee.Algorithms.Image.Segmentation.SNIC(\n image=naip, size=30, compactness=0.1, connectivity=8\n)\n\n# Display the original NAIP image as RGB.\nm = geemap.Map()\nm.set_center(-115.32053, 36.182016, 18)\nm.add_layer(naip, None, 'NAIP RGB')\n\n# Display the clusters.\nm.add_layer(snic.randomVisualizer(), None, 'Clusters')\n\n# Display the RGB cluster means.\nvis_params = {'bands': ['R_mean', 'G_mean', 'B_mean'], 'min': 0, 'max': 255}\nm.add_layer(snic, vis_params, 'RGB cluster means')\nm\n```"]]