Please wait a minute...
img

Wechat

Remote Sensing Technology and Application  2022, Vol. 37 Issue (2): 515-523    DOI: 10.11873/j.issn.1004-0323.2022.2.0515
    
Boosting Tree Model with Gabor and LPQ Feature Fusion of HSI Ground Object Recognition
Yanan Jiang1(),Chunlei Zhang2(),Xin Zhang3,Quanwei Xu1,Shutao Zhang1,Rui Zhou1
1.School of Science,China University of Geosciences(Beijing),Beijing 100083,China
2.Beijing Zhongdirunde Petroleum Technology Co. Ltd. ,Beijing 100083,China
3.School of Statistics,Beijing Normal University,Beijing 100875,China
Download:  HTML  PDF (4210KB) 
Export:  BibTeX | EndNote (RIS)      
Abstract  

To fully fuse the feature information in the spatial and frequency domains of hyperspectral image (HIS), a spatial-spectrum fusion HSI ground object recognition model that integrates multiscale features of Gabor and LPQ (Ms_GLPQ) is proposed. Firstly, the Gabor filter bank is used in the spatial domain to extract the multiscale and multidirectional spatial neighborhood information of various ground objects in HSI to describe the spatial structure of its edge and texture. Secondly, the Local Phase Quantization (LPQ) operator is utilized in the frequency domain to extract the multiscale frequency domain texture features, and the phase invariant feature description of HSI is obtained. Then the Principal Component Analysis (PCA) algorithm is used to reduce the dimensionality for the problem of feature redundancy, and the features in the spatial and frequency domains are fused to obtain the feature vector that fully describes the HSI information. Finally, the classifier based on Boosting tree (XGBoost, CatBoost, etc.) are utilized for recognition. Experiments on Indian Pines, Salinas, and tea farm datasets acquire accuracy rates of 85.88%, 94.42%, and 92.61%, respectively. The experimental results show that the Ms_GLPQ model can extract effective features in HSI and obtain more discriminative multi-featured region descriptors than traditional methods, and it performs better by using boosted tree model for ground object recognition and achieves higher accuracy than other classifiers.

Key words:  Hyperspectral image      Multiscale analysis      Gabor filter bank      Local Phase Quantization      Boosting tree model     
Received:  23 September 2020      Published:  17 June 2022
ZTFLH:  TP79  
Corresponding Authors:  Chunlei Zhang     E-mail:  2463613347@qq.com;676935005@qq.com
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Yanan Jiang
Chunlei Zhang
Xin Zhang
Quanwei Xu
Shutao Zhang
Rui Zhou

Cite this article: 

Yanan Jiang,Chunlei Zhang,Xin Zhang,Quanwei Xu,Shutao Zhang,Rui Zhou. Boosting Tree Model with Gabor and LPQ Feature Fusion of HSI Ground Object Recognition. Remote Sensing Technology and Application, 2022, 37(2): 515-523.

URL: 

http://www.rsta.ac.cn/EN/10.11873/j.issn.1004-0323.2022.2.0515     OR     http://www.rsta.ac.cn/EN/Y2022/V37/I2/515

Fig.1  Block diagram of the proposed Ms_GLPQ model
数据集获取地点成像仪器图像大小波长/μm波段数空间分辨率/m
Indian Pines美国印第安纳州印度松树林AVIRIS145×1450.40~2.5020020
Salinas美国加利福尼亚州 Salinas 山谷AVIRIS512×2170.40~2.502043.7
茶树江苏省常州市方麓村茶树种植基地PHI512×3480.417~0.855802.25
Table 1  Experimental datasets information
Fig.2  Feature maps of different Gabor filter parameters
Fig. 3  Statistical histogram features of LPQ at different scales
分类器Indian PinesSalinas茶树

原始

数据

PCA8原始数据PCA8原始数据PCA8
SVM55.6654.3678.7380.5390.9390.27
Bayes50.4051.8574.8278.0785.5589.2
BP73.7257.3488.9085.3191.9890.52
DT55.6256.5281.3480.8887.3686.88
RF72.3866.4687.1487.0491.190.98
XGBoost69.4364.5786.0186.0690.5890.42
LightGBM69.8864.4286.4785.8589.7389.07
CatBoost71.2667.3688.0487.2491.5491.39
Table 2  Comparison of image preprocessing effect
分类器Indian Pines
PCA8多尺度Gabor多尺度LPQMs_GLPQ
SVM54.3673.6962.2363.28
Bayes51.8561.3069.2969.36
BP57.3480.5271.4371.30
DT56.5269.8965.1666.90
RF66.4682.0382.4183.48
XGBoost64.5778.5580.3180.66
LightGBM64.4279.2881.2881.85
CatBoost67.3683.8085.5885.88
Table 3  Performance comparison of different algorithms in Indian Pines dataset
分类器Salinas
PCA8多尺度Gabor多尺度LPQMs_GLPQ
SVM80.5371.3884.8485.05
Bayes78.0769.6272.4272.77
BP85.3180.3189.5689.09
DT80.8881.4282.1881.47
RF87.0488.4893.7594.13
XGBoost86.0687.0189.4990.99
LightGBM85.8587.1290.8791.85
CatBoost87.2488.9294.2994.42
Table 4  Performance comparison of different algorithms in Salinas dataset
分类器茶树
PCA8多尺度Gabor多尺度LPQMs_GLPQ
SVM90.2777.4184.0485.20
Bayes89.2064.0585.1585.22
BP90.5283.6588.2288.48
DT86.8880.7781.5882.65
RF90.9887.4190.5390.35
XGBoost90.4286.9390.4990.72
LightGBM89.0786.2290.2990.45
CatBoost91.3988.5892.5992.61
Table 5  Performance comparison of different algorithms in Tea Farm dataset
Fig.4  Recognition results in HSI datasets
1 Jakob S, Zimmermann R, Gloaguen R. The need for accurate geometric and radiometric corrections of drone-borne hyperspectral data for mineral exploration: Mephysto—A toolbox for pre-processing drone-borne hyperspectral data[J]. Remote Sensing, 2017, 9(1): 88. DOI: .
doi: 10.3390/rs9010088
2 Carreiras J M B, Jones J, Lucas R M, et al. Mapping major land cover types and retrieving the age of secondary forests in the Brazilian Amazon by combining single-date optical and radar remote sensing data[J]. Remote Sensing of Environment, 2017, 194: 16-32. DOI: .
doi: 10.1016/j.rse.2017.03.016
3 Garg V, Kumar A S, Aggarwal S P, et al. Spectral similarity approach for mapping turbidity of an inland waterbody[J]. Journal of Hydrology, 2017, 550: 527-537. DOI: .
doi: 10.1016/j.jhydrol.2017.05.039
4 Zhong Bo, Ma Peng, Nie Aihua, et al. Land cover mapping using time series HJ-1 /CCD data[J]. Science China: Earth Sciences, 2014, 44(5): 967-977.
4 仲波, 马鹏, 聂爱华, 等. 基于时间序列 HJ-1 /CCD 数据的土地覆盖分类方法[J],中国科学:地球科学,2014,44(5):967-977.
5 Chen Ni, Ying Feng, Wang Jing, et al. Research on Land use information extraction based on U-Net[J]. Remote Sensing Technology and Application,2021,36(2):285-292.
5 陈妮,应丰,王静,等. 基于 U-Net 的高分辨率遥感图像土地利用信息提取[J]. 遥感技术与应用,2021,36(2):285-292.
6 Li J, Bioucas-Dias J M, Plaza A. Semisupervised hyperspectral image segmentation using multinomial logistic regression with active learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(11): 4085-4098. DOI: .
doi: 10.1109/TGRS.2010. 2060550
7 Garrity S R, Vierling L A, Smith A M S, et al. Automatic detection of shrub location, crown area, and cover using spatial wavelet analysis and aerial photography[J]. Canadian Journal of Remote Sensing, 2008, 34(2): S376-S384. DOI: .
doi: 10.5589/m08-056
8 Song Cuiyu, Li Peijun, Yang Fengjie. Remote sensing image classification based on texture features by multivariate local binary pattern[J]. Remote Sensing Technology and Application,2011,26(3):322-327,276.
8 宋翠玉,李培军,杨锋杰.基于多元局部二值模式的遥感图像纹理提取与分类[J].遥感技术与应用,2011,26(3):322-327,276.
9 Jiang Yanan, Zhang Xin, Zhang Chunlei, et al. Classification of remote sensing images based on multi-scale feature fusion using local binary patterns[J]. Remote Sensing for Natural Resources,2021,33(3):36-44.
9 姜亚楠,张欣,张春雷,等.基于多尺度LBP特征融合的遥感图像分类[J].自然资源遥感,2021,33(3):36-44.
10 Zhang Li. Remote sensing image feature and classification based on Gabor function[J]. Geomatics & Spatial Information Technology,2012,35(4):123-124,127.
10 张立. 基于Gabor函数的遥感影像特征与分类[J],测绘与空间地理信息,2012,35(4):123-124,127.
11 Liu Xun, Hua Wenshen, Yang Jia, et al. Spatial classification method for hyperspectral camouflage targets image based on local Gabor binary patterns[J]. High Power Laser and Particle Beams, 2014, 26(11): 49-56.
11 刘恂, 华文深, 杨佳, 等. 基于局部Gabor二进制模式的高光谱伪装目标图像分类(英文)[J],强激光与粒子束,2014,26(11):49-56.
12 Ojansivu V, Heikkilä J. Blur insensitive texture classification using local phase quantization[C]∥ International Conference on Image and Signal Processing. Berlin, Heidelberg: Springer, 2008: 236-243.
13 Jia Mingxing, Du Junqiang, Song Pengfei, et al. Face recogniton based on multi-feature optimization fusion of LBP and Gabor with multi-scale blocks[J]. Journal of Northeastern University(Natural Science Edition),2017,38(3):310-314.
13 贾明兴, 杜俊强, 宋鹏飞, 等. 基于不同分块多特征优化融合的人脸识别研究[J],东北大学学报(自然科学版),2017,38(3):310-314.
14 Chen Pengzhan, Hu Chao, Chen Xiaoyue. Facial expression recognition based on Gabor, LBP and LPQ feature[J]. Measurement & Control Technology,2018,37(8):16-20.
14 陈鹏展, 胡超, 陈晓玥. 融合Gabor、LBP与LPQ特征的面部表情识别[J],测控技术,2018,37(8):16-20.
15 Daugman J G. Two-dimensional spectral analysis of cortical receptive field profiles[J]. Vision Research,1980,20(10): 847-856. DOI: .
doi: 10.1016/0042-6989(80) 90065-6
16 Ojansivu V, Rahtu E, Heikkila J. Rotation invariant local phase quantization for blur insensitive texture analysis[C]∥2008 19th International Conference on Pattern Recognition. IEEE, 2008: 1-4.
17 Chen T, Guestrin C. Xgboost: A scalable tree boosting system[C]∥ Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining.2016, 785-794.
18 Prokhorenkova L, Gusev G, Vorobev A, et al. CatBoost: unbiased boosting with categorical features[C]∥ Advances in Neural Information Processing Systems,2018:6638-6648.
19 Vane G, Green R O, Chrien T G, et al. The Airborne Visible/Infrared Imaging Spectrometer (AVIRIS)[J]. Remote Sensing of Environment, 1987, 44(2–3): 127-143. DOI: .
doi: 10.1016/0034-4257(93)90012 -M
20 Zhang Xia, Zhang Bing, Zhang Lifu, et al. Hyperspectral remote sensing cataset for tea farm[DB/OL]. http:∥www. geodoi.ac.cn/webcn/doi. aspx?Id=720, 2017.08, 2019.10.张霞, 张兵, 张立福, 等. 茶树等十种地类高光谱遥感数据集[DB/OL]. http:∥www.geodoi.ac.cn/webcn/doi.aspx?Id=720,2017.08,2019.10. DOI: .
doi: 10.3974/geodb.2017. 03.04.V1
[1] Na Liu,Fengli Zhang,Zhikun Li,Yun Shao,Lei Pang,Lu Li. Updating of Built-up Areas based on SAR and Night Light Data and Its Optimization for WRF Model Wind Field Simulation[J]. Remote Sensing Technology and Application, 2022, 37(2): 488 -498 .
[2] . Remote Sensing Image Registration Based on SIFT and Mutual Information Screening Optimization[J]. Remote Sensing Technology and Application, 0, (): 0 .
[3] Jie Li,Kun Jia,Ning Zhang,Xiangqin Wei,Bing Wang. Evaluation of Ecological Protection Importance in Qingdao based on Remote Sensing and Ecological Service Model[J]. Remote Sensing Technology and Application, 2021, 36(6): 1329 -1338 .
[4] Jinming Zhu,Liwei Li,Gang Cheng,Lianru Gao,Bing Zhang. Detection and Analysis of High-rising Buildings within The sixth Ring Road of Beijing based on Sentinel-2 and Fully Convolutional Network[J]. Remote Sensing Technology and Application, 2021, 36(6): 1436 -1445 .
[5] Han Fu,Xiangtao Fan,Zhenzhen Yan,Xiaoping Du. Progress of Object Detection in Remote Sensing Images based on Deep Learning[J]. Remote Sensing Technology and Application, 2022, 37(2): 290 -305 .
[6] . Cross-validation Analysis of Sea Surface Wind and Significant Wave Height over the Arctic from the Autonomous HY-2B Satellite[J]. Remote Sensing Technology and Application, 0, (): 465 -0 .
[7] Huiyun Ma,Yanan Li,Xiaojing Wu,Yinze Ran,Junjie Yan. Automatic Detection of Night Land Fog based on H8/AHI Satellite Data[J]. Remote Sensing Technology and Application, 2022, 37(2): 408 -415 .
[8] Bingxue Fu,Jichao Zhang,Wenjie Du,Penglong Wang,Zhongchang Sun. Effective and Novel Impervious Surface Fine Mapping Method and Its Application on Monitoring Urban Sustainable Development Goals[J]. Remote Sensing Technology and Application, 2021, 36(6): 1339 -1349 .
[9] Qian Liu,Xinyu Hu,Xiaotong Li,Xianlin Qin. Building Recognition Method in Forest Districts Combining the Pixel-level and Object-level[J]. Remote Sensing Technology and Application, 2021, 36(6): 1350 -1357 .
[10] Yongkang Li,Xinjun Wang,Yanfei Ma,Bei Chen,Linan Yan,Guanhong Zhang. Downscaling Land Surface Temperature through AMSR-2 Observations by Using Machine Learning Algorithms[J]. Remote Sensing Technology and Application, 2022, 37(2): 474 -487 .