高光谱图像数据集（Hyperspectral Image Datasets）（1）

下载链接：【免费】高光谱图像数据集（HyperspectralImageDatasets）（1）资源-CSDN文库

1.Indian Pines

（1）数据集介绍

Indian Pines Dataset: The Indian Pines dataset was acquired in 1992 by the Airborne Visible Infrared Imaging Spectrometer(AVIRIS) in the Indian Pines region of Indiana, USA. The data size is 145 × 145 pixels, and a total of 10,249 pixel samples are labeled for training and testing purposes in HSI classification. The wavelength range of the AVIRIS spectrometer is from 0.4 to 2.5 μm, capable of imaging the ground over 220 consecutive bands. 20 bands that could not be reflected by water were removed, leaving a total of 200 bands for analysis.

（2）类别介绍

Indian PinesNo.Class NameNumber1Alfalfa462Corn-notill14283Corn-mintill8304Corn2375Grass-pasture4836Grass-trees7307Grass-pasture-mowed288Hay-windrowed4789Oats2010Soybean-notill97211Soybean-mintill245512Soybean-clean59313Wheat20514Woods126515Buildings38616Stone-Steel-Towers93Total10249

（3）图像

Indian Pines_false color

Indian Pines_gt

Indian Pines_pred_1

Indian Pines_pred_2

（4）调用代码

data = sio.loadmat('./data/Indian_Pines/Indian_pines_corrected.mat')['indian_pines_corrected']

labels = sio.loadmat('./data/Indian_Pines/Indian_pines_gt.mat')['indian_pines_gt']

2.Pavia University

（1）数据集介绍

Pavia University Dataset:The Pavia University dataset was acquired in 2003 by the Reflective Optics Spectrographic Imaging System (ROSIS-03) in parts of the city of Pavia, Italy.The data size is 610 x 340 pixels, with 42,776 pixel samples labeled for training and testing purposes in HSI classification.The wavelength range of the ROSIS-03 spectrometer is from 0.43 to 0.86 μm, capable of imaging the ground over 115 consecutive bands. 12 bands severely affected by noise were removed, leaving a total of 103 bands for analysis.

（2）类别介绍

Pavia UniversityNo.Class NameNumber1Asphalt66312Meadows186493Gravel20994Trees30645Painted metal sheets13456Bare soil50297Bitumen13308Self-blocking bricks36829Shadows947Total42776

（3）图像

Pavia University_false color

Pavia University_gt

Pavia University_pred_1

Pavia University_pred_2

（4）调用代码

data = sio.loadmat('./data/Pavia University/PaviaU.mat')['paviaU']

labels = sio.loadmat('./data/Pavia University/PaviaU_gt.mat')['paviaU_gt']

3.Salinas

（1）数据集介绍

Salinas Dataset:The Salinas dataset was acquired by the Airborne Visible Infrared Imaging Spectrometer (AVIRIS) in the Salinas Valley, California, USA.The data size is 512 x 217 pixels, with 54,129 pixel samples labeled for training and testing purposes in HSI classification.The image was continuously imaged over 224 bands, from which 20 bands that could not be reflected by water were removed, leaving a total of 204 bands for analysis.

（2）类别介绍

SalinasNo.Class NameNumber1Broccoli 120092Broccoli 237263Fallow19764Rough Plow13945Smooth26786Stubble39597Celery35798Untrained Grapes112719Vineyard Soil620310Green Weeds327811Romaine 4wk106812Romaine 5wk192713Romaine 6wk91614Romaine 7wk107015Untrained Vineyard726816Vertical Trellis1807Total54129

（3）图像

Salinas_false color

Salinas_gt

Salinas_pred_1

Salinas_pred_2

（4）调用代码

data = sio.loadmat('./data/Salinas/Salinas_corrected.mat')['salinas_corrected']

labels = sio.loadmat('./data/Salinas/Salinas_gt.mat')['salinas_gt']

4.Houston 2013

（1）数据集介绍

Houston 2013 Dataset:This scene was acquired by the ITRES CASI-1500 sensor over the campus of the University of Houston and its neighboring rural areas, and it was used in the 2013 GRSS Data Fusion Contest.The data comprises 144 spectral bands in the 380–1050 nm region and has dimensions of 349 × 1905 pixels with a spatial resolution of 2.5 m.

（2）类别介绍

Houston 2013No.Class NameNumber1Healthy grass13632Stressed grass13663Synthetic grass7604Trees13555Soil13536Water3547Residential13828Commercial13559Road136410Highway133711Railway134512Parking Lot 1134313Parking Lot 251114Tennis Court46615Running Track719Total16373

（3）图像

Houston 2013_false color

Houston 2013_gt

Houston 2013_pred_1

Houston 2013_pred_2

（4）调用代码

data = sio.loadmat('./data/Houston 2013/HustonU_IM.mat')['hustonu']

labels = sio.loadmat('./data/Houston 2013/HustonU_gt.mat')['hustonu_gt']

5.Houston 2018

（1）数据集介绍

Houston 2018 Dataset:The Houston 2018 dataset was acquired by the Hyperspectral Image Analysis Group and the NSF-funded Airborne Laser Mapping Center (NCALM) at the University of Houston, covering the university campus and surrounding areas.The data size is 601 x 2384 pixels, with 504,856 pixel samples labeled for training and testing purposes in hyperspectral image classification.The wavelength range of the image data is from 0.38 to 1.05 μm, with a total of 50 bands.

（2）类别介绍

Houston 2018No.Class NameNumber1Healthy grass97992Stressed grass325023Synthetic grass6844Evergreen trees135955Deciduous trees50216Soil45167Water2668Residential buildings397729Non-residential buildings22375210Roads4586611Sidewalks3402912Crosswalks151813Major thoroughfares4634814Highways986515Railways693716Paved parking lots1150017Unpaved parking lots14618Cars654719Trains536920Stadium seats6824Total504856

（3）图像

Houston 2018_false color

Houston 2018_gt

Houston 2018_pred_1

Houston 2018_pred_2

（4）调用代码

data = sio.loadmat('./data/Houston 2018/houstonU2018.mat')['houstonU']

labels = sio.loadmat('./data/Houston 2018/houstonU2018.mat')['houstonU_gt']