Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator
Tree species classification using hyperspectral imagery is a challenging task due to the high spectral similarity between species and large intra-species variability. This paper proposes a solution using the Multiple Instance Adaptive Cosine Estimator (MI-ACE) algorithm. MI-ACE estimates a discrimin...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2019-02-01
|
| Series: | PeerJ |
| Subjects: | |
| Online Access: | https://peerj.com/articles/6405.pdf |
| _version_ | 1827605917728768000 |
|---|---|
| author | Sheng Zou Paul Gader Alina Zare |
| author_facet | Sheng Zou Paul Gader Alina Zare |
| author_sort | Sheng Zou |
| collection | DOAJ |
| description | Tree species classification using hyperspectral imagery is a challenging task due to the high spectral similarity between species and large intra-species variability. This paper proposes a solution using the Multiple Instance Adaptive Cosine Estimator (MI-ACE) algorithm. MI-ACE estimates a discriminative target signature to differentiate between a pair of tree species while accounting for label uncertainty. Multi-class species classification is achieved by training a set of one-vs-one MI-ACE classifiers corresponding to the classification between each pair of tree species and a majority voting on the classification results from all classifiers. Additionally, the performance of MI-ACE does not rely on parameter settings that require tuning resulting in a method that is easy to use in application. Results presented are using training and testing data provided by a data analysis competition aimed at encouraging the development of methods for extracting ecological information through remote sensing obtained through participation in the competition. The experimental results using one-vs-one MI-ACE technique composed of a hierarchical classification, where a tree crown is first classified to one of the genus classes and one of the species classes. The species-level rank-1 classification accuracy is 86.4% and cross entropy is 0.9395 on the testing data, provided by the competition organizer, without the release of ground truth for testing data. Similarly, the same evaluation metrics are computed on the training data, where the rank-1 classification accuracy is 95.62% and the cross entropy is 0.2649. The results show that the presented approach can not only classify the majority species classes, but also classify the rare species classes. |
| first_indexed | 2024-03-09T06:30:00Z |
| format | Article |
| id | doaj.art-783588ab9ecd40d39338993fd0ed7bad |
| institution | Directory Open Access Journal |
| issn | 2167-8359 |
| language | English |
| last_indexed | 2024-03-09T06:30:00Z |
| publishDate | 2019-02-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ |
| spelling | doaj.art-783588ab9ecd40d39338993fd0ed7bad2023-12-03T11:07:00ZengPeerJ Inc.PeerJ2167-83592019-02-017e640510.7717/peerj.6405Hyperspectral tree crown classification using the multiple instance adaptive cosine estimatorSheng Zou0Paul Gader1Alina Zare2Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL, United States of AmericaDepartment of Computer & Information Science & Engineering, University of Florida, Gainesville, FL, United States of AmericaDepartment of Electrical and Computer Engineering, University of Florida, Gainesville, FL, United States of AmericaTree species classification using hyperspectral imagery is a challenging task due to the high spectral similarity between species and large intra-species variability. This paper proposes a solution using the Multiple Instance Adaptive Cosine Estimator (MI-ACE) algorithm. MI-ACE estimates a discriminative target signature to differentiate between a pair of tree species while accounting for label uncertainty. Multi-class species classification is achieved by training a set of one-vs-one MI-ACE classifiers corresponding to the classification between each pair of tree species and a majority voting on the classification results from all classifiers. Additionally, the performance of MI-ACE does not rely on parameter settings that require tuning resulting in a method that is easy to use in application. Results presented are using training and testing data provided by a data analysis competition aimed at encouraging the development of methods for extracting ecological information through remote sensing obtained through participation in the competition. The experimental results using one-vs-one MI-ACE technique composed of a hierarchical classification, where a tree crown is first classified to one of the genus classes and one of the species classes. The species-level rank-1 classification accuracy is 86.4% and cross entropy is 0.9395 on the testing data, provided by the competition organizer, without the release of ground truth for testing data. Similarly, the same evaluation metrics are computed on the training data, where the rank-1 classification accuracy is 95.62% and the cross entropy is 0.2649. The results show that the presented approach can not only classify the majority species classes, but also classify the rare species classes.https://peerj.com/articles/6405.pdfTree crownClassificationHyperspectralMultiple instanceNEONSpecies classification |
| spellingShingle | Sheng Zou Paul Gader Alina Zare Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator PeerJ Tree crown Classification Hyperspectral Multiple instance NEON Species classification |
| title | Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| title_full | Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| title_fullStr | Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| title_full_unstemmed | Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| title_short | Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| title_sort | hyperspectral tree crown classification using the multiple instance adaptive cosine estimator |
| topic | Tree crown Classification Hyperspectral Multiple instance NEON Species classification |
| url | https://peerj.com/articles/6405.pdf |
| work_keys_str_mv | AT shengzou hyperspectraltreecrownclassificationusingthemultipleinstanceadaptivecosineestimator AT paulgader hyperspectraltreecrownclassificationusingthemultipleinstanceadaptivecosineestimator AT alinazare hyperspectraltreecrownclassificationusingthemultipleinstanceadaptivecosineestimator |