Genetic Programming for High-Level Feature Learning in Crop Classification

Information on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are...

Full description

Bibliographic Details
Main Authors: Miao Lu, Ying Bi, Bing Xue, Qiong Hu, Mengjie Zhang, Yanbing Wei, Peng Yang, Wenbin Wu
Format: Article
Language:English
Published: MDPI AG 2022-08-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/14/16/3982
_version_ 1797442327680122880
author Miao Lu
Ying Bi
Bing Xue
Qiong Hu
Mengjie Zhang
Yanbing Wei
Peng Yang
Wenbin Wu
author_facet Miao Lu
Ying Bi
Bing Xue
Qiong Hu
Mengjie Zhang
Yanbing Wei
Peng Yang
Wenbin Wu
author_sort Miao Lu
collection DOAJ
description Information on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are critical to the accuracy of crop maps. Although several approaches are available for extracting spectral, temporal, and phenological features for crop identification, these methods depend heavily on domain knowledge and human experiences, adding uncertainty to the final crop classification. This study proposed a novel Genetic Programming (GP) approach to learning high-level features from time series images for crop classification to address this issue. We developed a new representation of GP to extend the GP tree’s width and depth to dynamically generate either fixed or flexible informative features without requiring domain knowledge. This new GP approach was wrapped with four classifiers, i.e., K-Nearest Neighbor (KNN), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM), and was then used for crop classification based on MODIS time series data in Heilongjiang Province, China. The performance of the GP features was compared with the traditional features of vegetation indices (VIs) and the advanced feature learning method Multilayer Perceptron (MLP) to show GP effectiveness. The experiments indicated that high-level features learned by GP improved the classification accuracies, and the accuracies were higher than those using VIs and MLP. GP was more robust and stable for diverse classifiers, different feature numbers, and various training sample sets compared with classification using VI features and the classifier MLP. The proposed GP approach automatically selects valuable features from the original data and uses them to construct high-level features simultaneously. The learned features are explainable, unlike those of a black-box deep learning model. This study demonstrated the outstanding performance of GP for feature learning in crop classification. GP has the potential of becoming a mainstream method to solve complex remote sensing tasks, such as feature transfer learning, image classification, and change detection.
first_indexed 2024-03-09T12:40:14Z
format Article
id doaj.art-d8b6f63fe5d24d3ea5a504fd10384489
institution Directory Open Access Journal
issn 2072-4292
language English
last_indexed 2024-03-09T12:40:14Z
publishDate 2022-08-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj.art-d8b6f63fe5d24d3ea5a504fd103844892023-11-30T22:19:42ZengMDPI AGRemote Sensing2072-42922022-08-011416398210.3390/rs14163982Genetic Programming for High-Level Feature Learning in Crop ClassificationMiao Lu0Ying Bi1Bing Xue2Qiong Hu3Mengjie Zhang4Yanbing Wei5Peng Yang6Wenbin Wu7Key Laboratory of Agricultural Remote Sensing, Ministry of Agriculture and Rural Affairs/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, ChinaSchool of Engineering and Computer Science, Victoria University of Wellington, Wellington 6140, New ZealandSchool of Engineering and Computer Science, Victoria University of Wellington, Wellington 6140, New ZealandSchool of Urban and Environmental Sciences, Central China Normal University, Wuhan 430079, ChinaSchool of Engineering and Computer Science, Victoria University of Wellington, Wellington 6140, New ZealandKey Laboratory of Agricultural Remote Sensing, Ministry of Agriculture and Rural Affairs/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, ChinaKey Laboratory of Agricultural Remote Sensing, Ministry of Agriculture and Rural Affairs/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, ChinaKey Laboratory of Agricultural Remote Sensing, Ministry of Agriculture and Rural Affairs/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, ChinaInformation on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are critical to the accuracy of crop maps. Although several approaches are available for extracting spectral, temporal, and phenological features for crop identification, these methods depend heavily on domain knowledge and human experiences, adding uncertainty to the final crop classification. This study proposed a novel Genetic Programming (GP) approach to learning high-level features from time series images for crop classification to address this issue. We developed a new representation of GP to extend the GP tree’s width and depth to dynamically generate either fixed or flexible informative features without requiring domain knowledge. This new GP approach was wrapped with four classifiers, i.e., K-Nearest Neighbor (KNN), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM), and was then used for crop classification based on MODIS time series data in Heilongjiang Province, China. The performance of the GP features was compared with the traditional features of vegetation indices (VIs) and the advanced feature learning method Multilayer Perceptron (MLP) to show GP effectiveness. The experiments indicated that high-level features learned by GP improved the classification accuracies, and the accuracies were higher than those using VIs and MLP. GP was more robust and stable for diverse classifiers, different feature numbers, and various training sample sets compared with classification using VI features and the classifier MLP. The proposed GP approach automatically selects valuable features from the original data and uses them to construct high-level features simultaneously. The learned features are explainable, unlike those of a black-box deep learning model. This study demonstrated the outstanding performance of GP for feature learning in crop classification. GP has the potential of becoming a mainstream method to solve complex remote sensing tasks, such as feature transfer learning, image classification, and change detection.https://www.mdpi.com/2072-4292/14/16/3982crop classificationgenetic programmingfeature learninghigh-level featuresgenetic programming representation
spellingShingle Miao Lu
Ying Bi
Bing Xue
Qiong Hu
Mengjie Zhang
Yanbing Wei
Peng Yang
Wenbin Wu
Genetic Programming for High-Level Feature Learning in Crop Classification
Remote Sensing
crop classification
genetic programming
feature learning
high-level features
genetic programming representation
title Genetic Programming for High-Level Feature Learning in Crop Classification
title_full Genetic Programming for High-Level Feature Learning in Crop Classification
title_fullStr Genetic Programming for High-Level Feature Learning in Crop Classification
title_full_unstemmed Genetic Programming for High-Level Feature Learning in Crop Classification
title_short Genetic Programming for High-Level Feature Learning in Crop Classification
title_sort genetic programming for high level feature learning in crop classification
topic crop classification
genetic programming
feature learning
high-level features
genetic programming representation
url https://www.mdpi.com/2072-4292/14/16/3982
work_keys_str_mv AT miaolu geneticprogrammingforhighlevelfeaturelearningincropclassification
AT yingbi geneticprogrammingforhighlevelfeaturelearningincropclassification
AT bingxue geneticprogrammingforhighlevelfeaturelearningincropclassification
AT qionghu geneticprogrammingforhighlevelfeaturelearningincropclassification
AT mengjiezhang geneticprogrammingforhighlevelfeaturelearningincropclassification
AT yanbingwei geneticprogrammingforhighlevelfeaturelearningincropclassification
AT pengyang geneticprogrammingforhighlevelfeaturelearningincropclassification
AT wenbinwu geneticprogrammingforhighlevelfeaturelearningincropclassification