Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Remote Sensing on 26/10/2016, available online: http://www.tandfonline.com/10.1080/01431161.2016.1246771
Accepted author manuscript, 1.68 MB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - A novel unsupervised bee colony optimization (UBCO) method for remote sensing image classification
T2 - a case study in a heterogeneous marsh area
AU - Li, Huapeng
AU - Zhang, Shuqing
AU - Ding, Xiaohui
AU - Zhang, Ce
AU - Cropp, Roger
N1 - This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Remote Sensing on 26/10/2016, available online: http://www.tandfonline.com/10.1080/01431161.2016.1246771
PY - 2016/10/27
Y1 - 2016/10/27
N2 - Unsupervised image classification is an important means to obtain land use/cover informationin the field of remote sensing, since it does not require initial knowledge (training samples)for classification. Traditional methods such as k-means and ISODATA have limitations insolving this NP-hard unsupervised classification problem, mainly due to their strictassumptions about the data distribution. The bee colony optimization (BCO) is a new type ofswarm intelligence, based upon which a simple and novel unsupervised bee colonyoptimization (UBCO) method is proposed for remote sensing image classification. UBCOpossesses powerful exploitation and exploration capacities that are carried out by employedbees, onlookers and scouts. This enables the promising regions to be globally searchedquickly and thoroughly, without becoming trapped on local optima. In addition, it has norestrictions on data distribution, and thus is especially suitable for handling complex remotesensing data. We tested the method on the Zhalong National Nature Reserve (ZNNR)—atypical inland wetland ecosystem in China, whose landscape is heterogeneous. Thepreliminary results showed that UBCO (overall accuracy = 80.81%) achieved statisticallysignificant better classification result (McNemar test) in comparison with traditional k-means(63.11%) and other intelligent clustering methods built on genetic algorithm (UGA, 71.49%),differential evolution (UDE, 77.57%) and particle swarm optimization (UPSO, 69.86%). Therobustness and superiority of UBCO were also demonstrated from the two other study sitesnext to the ZNNR with distinct landscapes (urban and natural landscapes). Enabling toconsistently find the optimal or nearly optimal global solution in image clustering, the UBCOis thus suggested as a robust method for unsupervised remote sensing image classification,especially in the case of heterogeneous areas.
AB - Unsupervised image classification is an important means to obtain land use/cover informationin the field of remote sensing, since it does not require initial knowledge (training samples)for classification. Traditional methods such as k-means and ISODATA have limitations insolving this NP-hard unsupervised classification problem, mainly due to their strictassumptions about the data distribution. The bee colony optimization (BCO) is a new type ofswarm intelligence, based upon which a simple and novel unsupervised bee colonyoptimization (UBCO) method is proposed for remote sensing image classification. UBCOpossesses powerful exploitation and exploration capacities that are carried out by employedbees, onlookers and scouts. This enables the promising regions to be globally searchedquickly and thoroughly, without becoming trapped on local optima. In addition, it has norestrictions on data distribution, and thus is especially suitable for handling complex remotesensing data. We tested the method on the Zhalong National Nature Reserve (ZNNR)—atypical inland wetland ecosystem in China, whose landscape is heterogeneous. Thepreliminary results showed that UBCO (overall accuracy = 80.81%) achieved statisticallysignificant better classification result (McNemar test) in comparison with traditional k-means(63.11%) and other intelligent clustering methods built on genetic algorithm (UGA, 71.49%),differential evolution (UDE, 77.57%) and particle swarm optimization (UPSO, 69.86%). Therobustness and superiority of UBCO were also demonstrated from the two other study sitesnext to the ZNNR with distinct landscapes (urban and natural landscapes). Enabling toconsistently find the optimal or nearly optimal global solution in image clustering, the UBCOis thus suggested as a robust method for unsupervised remote sensing image classification,especially in the case of heterogeneous areas.
U2 - 10.1080/01431161.2016.1246771
DO - 10.1080/01431161.2016.1246771
M3 - Journal article
VL - 37
SP - 5627
EP - 5748
JO - International Journal of Remote Sensing
JF - International Journal of Remote Sensing
SN - 0143-1161
IS - 24
ER -