LujianYao commited on
Commit
c724c1b
·
verified ·
1 Parent(s): b1ceeeb

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. ._README.md +0 -0
  2. README.md +1 -42
._README.md ADDED
Binary file (4.1 kB). View file
 
README.md CHANGED
@@ -1,6 +1,5 @@
1
  # Hyperspectral and Multispectral Smoke Segmentation Datasets
2
 
3
- This repository contains two smoke segmentation datasets introduced in our paper **"Hyperspectral Smoke Segmentation via Mixture of Prototypes"**.
4
  ---
5
 
6
  ## Overview
@@ -11,7 +10,7 @@ Smoke segmentation is critical for wildfire management and industrial safety app
11
  *Figure 1: Motivation for hyperspectral smoke segmentation. The upper part shows challenging smoke scenarios with cloud interference and semi-transparent regions in the visible light band. The lower part plots the spectral distribution of marked points, where yellow shaded regions highlight key discriminatory band ranges.*
12
 
13
  To address these challenges, we introduce:
14
- 1. **HSSDataset**: The first hyperspectral smoke segmentation dataset with 25 spectral bands (600-974nm)
15
  2. **MSSDataset**: A multispectral dataset with RGB-infrared paired images
16
 
17
  ---
@@ -139,46 +138,6 @@ FLAME2_VOC/
139
 
140
  ---
141
 
142
- ## Dataset Statistics
143
-
144
- ### HSSDataset Summary
145
-
146
- | Metric | Value |
147
- |--------|-------|
148
- | Total Annotated Samples | 1,007 |
149
- | Spectral Bands | 25 |
150
- | Wavelength Range | 600-974 nm |
151
- | Real-world Scenarios | 20 |
152
- | Raw Video Frames | 18,000+ |
153
- | Annotators per Sample | 3 |
154
- | Image Resolution | Variable (industrial setting) |
155
-
156
- ### MSSDataset Summary
157
-
158
- | Metric | Value |
159
- |--------|-------|
160
- | Total Annotated Samples | 200 |
161
- | Spectral Channels | 4 (RGB + IR) |
162
- | Scenarios | Wildland fires |
163
- | Source Dataset | [FLAME2](https://ieee-dataport.org/open-access/flame-2-fire-detection-and-modeling-aerial-multi-spectral-image-dataset) |
164
- | Image Resolution | Variable |
165
-
166
- ### Challenging Scenario Distribution (HSSDataset)
167
-
168
- | Scenario Category | Sample Count | Description |
169
- |-------------------|--------------|-------------|
170
- | **Scene-based Classification** |
171
- | High Exposure | 214 | Bright lighting with overexposed regions |
172
- | Low Visibility | 118 | Poor lighting and atmospheric conditions |
173
- | Complex Background | 411 | Industrial environments with clutter |
174
- | Cloud Interference | 264 | Cloud-smoke confusion scenarios |
175
- | **Smoke-based Classification** |
176
- | Early-stage Smoke | 268 | Small smoke plumes in initial phases |
177
- | High Transparency Smoke | 258 | Varying opacity and transparency |
178
- | Complex-shaped Smoke | 481 | Irregular patterns with unclear boundaries |
179
-
180
- ---
181
-
182
  ## Data Format
183
 
184
  ### Hyperspectral Data (HSSDataset)
 
1
  # Hyperspectral and Multispectral Smoke Segmentation Datasets
2
 
 
3
  ---
4
 
5
  ## Overview
 
10
  *Figure 1: Motivation for hyperspectral smoke segmentation. The upper part shows challenging smoke scenarios with cloud interference and semi-transparent regions in the visible light band. The lower part plots the spectral distribution of marked points, where yellow shaded regions highlight key discriminatory band ranges.*
11
 
12
  To address these challenges, we introduce:
13
+ 1. **HSSDataset**: The first hyperspectral smoke segmentation dataset with 25 spectral bands
14
  2. **MSSDataset**: A multispectral dataset with RGB-infrared paired images
15
 
16
  ---
 
138
 
139
  ---
140
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
141
  ## Data Format
142
 
143
  ### Hyperspectral Data (HSSDataset)