mhassanch commited on
Commit
89a64ff
·
1 Parent(s): 5c7dcd7

add model card

Browse files
Files changed (1) hide show
  1. README.md +92 -0
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - geospatial
4
+ - geobase
5
+ - land-cover-classification
6
+ ---
7
+ | <img src="https://upload.wikimedia.org/wikipedia/commons/6/6a/JavaScript-logo.png" width="28" height="28"> | [@geobase-js/geoai](https://www.npmjs.com/package/@geobase-js/geoai) |
8
+ |---|---|
9
+
10
+
11
+
12
+ > `task = land-cover-classification`
13
+
14
+ ### 🛠 Model Purpose
15
+ This model is part of the **[@geobase-js/geoai](https://github.com/geobase-ai/geoai)** javascript library.
16
+
17
+ **GeoAi** enables geospatial AI inference **directly in the browser or Node.js** without requiring a heavy backend.
18
+
19
+ **GeoAi** pipeline accepts **geospatial polygons** as input (in GeoJSON format) and outputs results as a **GeoJSON FeatureCollection**, ready for use with libraries like **Leaflet** and **Mapbox GL**.
20
+
21
+ <video controls autoplay loop width="1024" height="720" src="https://geobase-docs.s3.amazonaws.com/geobase-ai-assets/land-cover-classification.mp4"></video>
22
+
23
+ ---
24
+ ### 🚀 Demo
25
+
26
+ Explore the model in action with the interactive [Demo](https://docs.geobase.app/geoai-live/tasks/land-cover-classification).
27
+
28
+ ### 📦 Model Information
29
+ - **Architecture**: MaskRCNN
30
+ - **Source Model**: https://opengeoai.org/examples/wetland_mapping/
31
+ - **Quantization**: Yes
32
+ ---
33
+
34
+ > **Note:** This model only works if the input imagery contains both RGB and NIR bands.
35
+
36
+
37
+ ### 💡 Example Usage
38
+
39
+ ```javascript
40
+ import { geoai } from "@geobase-js/geoai";
41
+
42
+ // Example polygon (GeoJSON)
43
+ const polygon = {
44
+ type: "Feature",
45
+ properties: {},
46
+ geometry: {
47
+ coordinates: [
48
+ [
49
+ [-99.98170823334205, 50.64318249608178],
50
+ [-99.9806018185983, 50.6431476272638],
51
+ [-99.9807117895183, 50.642403469489295],
52
+ [-99.98256386074388, 50.64247150726101],
53
+ [-99.98170823334205, 50.64318249608178]
54
+ ],
55
+ ],
56
+ type: "Polygon",
57
+ },
58
+ } as GeoJSON.Feature;
59
+
60
+ // Initialize pipeline
61
+ const pipeline = await geoai.pipeline(
62
+ [{ task: "land-cover-classification" }],
63
+ providerParams
64
+ );
65
+
66
+ // Run detection
67
+ const result = await pipeline.inference({
68
+ inputs: { polygon }
69
+ });
70
+
71
+ // Sample output format
72
+ // {
73
+ // "detections": [
74
+ // {type: 'FeatureCollection', features: Array(6)},
75
+ // {type: 'FeatureCollection', features: Array(43)},
76
+ // {type: 'FeatureCollection', features: Array(6)},
77
+ // {type: 'FeatureCollection', features: Array(3)},
78
+ // {type: 'FeatureCollection', features: Array(11)},
79
+ // {type: 'FeatureCollection', features: Array(36)},
80
+ // {type: 'FeatureCollection', features: Array(0)},
81
+ // {type: 'FeatureCollection', features: Array(9)}
82
+ // ]
83
+ // "outputimage": GeoRawImage {data: Uint8ClampedArray(1048576), width: 512, height: 512, channels: 4, bounds: {…}, …}
84
+ // }
85
+
86
+ ```
87
+ ### 📖 Documentation & Demo
88
+
89
+ - GeoBase Docs: https://docs.geobase.app/geoai
90
+ - NPM Package: https://www.npmjs.com/package/@geobase-js/geoai
91
+ - Demo Playground: https://docs.geobase.app/geoai-live/tasks/land-cover-classification
92
+ - GitHub Repo: https://github.com/decision-labs/geobase-ai.js