An Unsupervised Crop Classification Method Based on Principal Components Isometric Binning

The accurate and timely access to the spatial distribution information of crops is of great importance for agricultural production management. Although widely used, supervised classification mapping requires a large number of field samples, and is consequently costly in terms of time and money. In order to reduce the need for sample size, this paper proposes an unsupervised classification method based on principal components isometric binning (PCIB). In particular, principal component analysis (PCA) dimensionality reduction is applied to the classification features, followed by the division of the top k principal components into equidistant bins. Bins of the same category are subsequently merged as a class label. Multitemporal Gaofen 1 satellite (GF-1) remote sensing images were collected over the southwest of Hulin City and Luobei County of Hegang City, Heilongjiang Province, China in order to map crop types in 2016 and 2017. Our proposed method was compared with commonly used classifiers (random forest, K-means and Iterative Self-Organizing Data Analysis Techniques Algorithm (ISODATA)). Results demonstrate PCIB and random forest to have the highest classification accuracies, reaching 82% in 2016 in the southwest of Hulin City. In Luobei County in 2016, the accuracies of PCIB and random forest were determined as 81% and 82%, respectively. It can be concluded that the overall accuracy of our proposed method meets the basic requirements of classification accuracy. Despite exhibiting a lower accuracy than that of random forest, PCIB does not require a large field sample size, thus making it more suitable for large-scale crop mapping.

[1]  Shaowen Wang,et al.  A high-performance and in-season classification system of field-level crop types using time-series Landsat data and a machine learning approach , 2018, Remote Sensing of Environment.

[2]  Liang Tong,et al.  Identification of Seed Maize Fields With High Spatial Resolution and Multiple Spectral Remote Sensing Using Random Forest Classifier , 2020, Remote. Sens..

[3]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[4]  H. Abdi,et al.  Principal component analysis , 2010 .

[5]  Hui Lin,et al.  A small-patched convolutional neural network for mangrove mapping at species level using high-resolution remote-sensing image , 2019, Ann. GIS.

[6]  Y. Hirosawa,et al.  Application of standardized principal component analysis to land-cover characterization using multitemporal AVHRR data , 1996 .

[7]  B. Wardlow,et al.  Analysis of time-series MODIS 250 m vegetation index data for crop classification in the U.S. Central Great Plains , 2007 .

[8]  R. Congalton,et al.  Automated cropland mapping of continental Africa using Google Earth Engine cloud computing , 2017 .

[9]  Le Yu,et al.  A 30 meter land cover mapping of China with an efficient clustering algorithm CBEST , 2014, Science China Earth Sciences.

[10]  Xin Huang,et al.  A novel co-training approach for urban land cover mapping with unclear Landsat time series imagery , 2018, Remote Sensing of Environment.

[11]  AbdiHervé,et al.  Principal Component Analysis , 2010, Essentials of Pattern Recognition.

[12]  Lorenzo Bruzzone,et al.  A Novel Transductive SVM for Semisupervised Classification of Remote-Sensing Images , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[13]  Huajun Tang,et al.  How do temporal and spectral features matter in crop classification in Heilongjiang Province, China? , 2017 .

[14]  Danny Lo Seen,et al.  A Remote Sensing Approach for Regional-Scale Mapping of Agricultural Land-Use Systems Based on NDVI Time Series , 2017, Remote. Sens..

[15]  Ying Liu,et al.  A self-trained semisupervised SVM approach to the remote sensing land cover classification , 2013, Comput. Geosci..

[16]  Li Wang,et al.  Using Moderate-Resolution Temporal NDVI Profiles for High-Resolution Crop Mapping in Years of Absent Ground Reference Data: A Case Study of Bole and Manas Counties in Xinjiang, China , 2016, ISPRS Int. J. Geo Inf..

[17]  Adriaan Van Niekerk,et al.  Pre-harvest classification of crop types using a Sentinel-2 time-series and machine learning , 2020, Comput. Electron. Agric..

[18]  Jawad Iounousse,et al.  Using an unsupervised approach of Probabilistic Neural Network (PNN) for land use classification from multitemporal satellite images , 2015, Appl. Soft Comput..

[19]  M. Dharani,et al.  Land use and land cover change detection by using principal component analysis and morphological operations in remote sensing applications , 2019, International Journal of Computers and Applications.

[20]  Changqing Song,et al.  RDCRMG: A Raster Dataset Clean & Reconstitution Multi-Grid Architecture for Remote Sensing Monitoring of Vegetation Dryness , 2018, Remote. Sens..

[21]  Lorenzo Bruzzone,et al.  Mean Map Kernel Methods for Semisupervised Cloud Classification , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[22]  Mingquan Wu,et al.  Crop classification using crop knowledge of the previous-year: Case study in Southwest Kansas, USA , 2016 .

[23]  Matthew C. Hansen,et al.  National-scale soybean mapping and area estimation in the United States using medium resolution satellite imagery and field survey , 2017 .

[24]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[25]  Farid Melgani,et al.  Genetic SVM Approach to Semisupervised Multitemporal Classification , 2008, IEEE Geoscience and Remote Sensing Letters.

[26]  Leila Maria Garcia Fonseca,et al.  Urban population estimation based on residential buildings volume using IKONOS-2 images and lidar data , 2016 .

[27]  D. Nagesh Kumar,et al.  Evaluation of Feature Selection and Feature Extraction Techniques on Multi-Temporal Landsat-8 Images for Crop Classification , 2019, Remote Sensing in Earth Systems Sciences.

[28]  Jason Weston,et al.  Semisupervised Neural Networks for Efficient Hyperspectral Image Classification , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[29]  V. Mani,et al.  Crop Stage Classification of Hyperspectral Data Using Unsupervised Techniques , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[30]  Hang Zhou,et al.  Deep learning based multi-temporal crop classification , 2019, Remote Sensing of Environment.

[31]  Limin Yang,et al.  Development of a global land cover characteristics database and IGBP DISCover from 1 km AVHRR data , 2000 .

[32]  Yu Wang,et al.  A study of a Gaussian mixture model for urban land-cover mapping based on VHR remote sensing imagery , 2016 .

[33]  Namita M Kulkarni.  Crop Identification Using Unsuperviesd ISODATA and K-Means from Multispectral Remote Sensing Imagery , 2017 .

[34]  Yi Li,et al.  Winter wheat mapping using a random forest classifier combined with multi-temporal and multi-sensor data , 2018, Int. J. Digit. Earth.

[35]  Saroj K. Meher,et al.  Semisupervised classification of remote sensing images using efficient neighborhood learning method , 2020, Eng. Appl. Artif. Intell..

[36]  Shuqing Zhang,et al.  Crop classification from full-year fully-polarimetric L-band UAVSAR time-series using the Random Forest algorithm , 2020, Int. J. Appl. Earth Obs. Geoinformation.

[37]  Shaoming Li,et al.  Crop Mapping Based on Historical Samples and New Training Samples Generation in Heilongjiang Province, China , 2019, Sustainability.

[38]  Lin Yan,et al.  Improved time series land cover classification by missing-observation-adaptive nonlinear dimensionality reduction , 2015 .

[39]  David B. Lobell,et al.  Crop type mapping without field-level labels: Random forest transfer and unsupervised clustering techniques , 2019, Remote Sensing of Environment.

[40]  Dehai Zhu,et al.  Large-Scale Crop Mapping Based on Machine Learning and Parallel Computation with Grids , 2019, Remote. Sens..

[41]  Wei Liu,et al.  A Cloud Detection Approach Based on Hybrid Multispectral Features with Dynamic Thresholds for GF-1 Remote Sensing Images , 2020, Remote. Sens..

[42]  Adriano L. I. Oliveira,et al.  Letters: Novelty detection with constructive probabilistic neural networks , 2008 .

[43]  Prasad S. Thenkabail,et al.  Mapping rice-fallow cropland areas for short-season grain legumes intensification in South Asia using MODIS 250 m time-series data , 2016, Int. J. Digit. Earth.