A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring

Near-infrared (NIR) modification of low-cost cameras is considered an important method to acquire high-resolution NIR images on an unmanned aerial vehicle (UAV) platform. However, few studies have examined filter selection methods to modify consumer-grade cameras for UAV-based agricultural crop moni...

Descripción completa

Detalles Bibliográficos
Autores principales: Wang, Chufeng, Zhang, Jian, Wu, Hao, Liu, Bin, Wang, Botao, You, Yunhao, Tan, Zuojun, Xie, Jing, You, Liangzhi, Zhang, Junqiang, Wen, Ping
Formato: Journal Article
Lenguaje:Inglés
Publicado: Elsevier 2025
Materias:
Acceso en línea:https://hdl.handle.net/10568/173594
_version_ 1855532690376228864
author Wang, Chufeng
Zhang, Jian
Wu, Hao
Liu, Bin
Wang, Botao
You, Yunhao
Tan, Zuojun
Xie, Jing
You, Liangzhi
Zhang, Junqiang
Wen, Ping
author_browse Liu, Bin
Tan, Zuojun
Wang, Botao
Wang, Chufeng
Wen, Ping
Wu, Hao
Xie, Jing
You, Liangzhi
You, Yunhao
Zhang, Jian
Zhang, Junqiang
author_facet Wang, Chufeng
Zhang, Jian
Wu, Hao
Liu, Bin
Wang, Botao
You, Yunhao
Tan, Zuojun
Xie, Jing
You, Liangzhi
Zhang, Junqiang
Wen, Ping
author_sort Wang, Chufeng
collection Repository of Agricultural Research Outputs (CGSpace)
description Near-infrared (NIR) modification of low-cost cameras is considered an important method to acquire high-resolution NIR images on an unmanned aerial vehicle (UAV) platform. However, few studies have examined filter selection methods to modify consumer-grade cameras for UAV-based agricultural crop monitoring. This study addresses a key challenge: how to balance imaging quality with spectral sensitivity when selecting filters for the modification of consumer-grade cameras. To this end, the normalized difference spectral index (NDSI) and the ratio spectral index (RSI) formulations were used to calculate the spectral indices (SIs) from all possible combinations of any two center wavelengths in UAV hyperspectral data. The contour maps of the coefficient of determination (R2) between the SIs and ground-measured rapeseed LAI were then computed to automatically generate the broadband combinations with optimized center wavelengths and effective bandwidths for selecting filters on camera modification. Results showed that a consumer-grade camera (Nikon D7000) modified by the selected filters had performance comparable with a multispectral camera (RedEdge Micasense 3), but slightly worse than a research-grade hyperspectral camera (Nano-Hyperspec®) in terms of SIs for LAI estimation. In addition, the high-resolution images from the modified camera were processed to obtain accurate crop plant height information. The SIs coupled with plant height from the modified camera (rRMSE = 18.1 % for field 1 and 14.3 % for field 2) was found to perform similar to, and in some cases even better than, those from the research-grade multispectral (rRMSE = 17.9 % and 16.7 % for the respective fields) and hyperspectral (rRMSE = 18.8 % for field 1) cameras for UAV-based LAI estimation. The findings from this study indicate that the proposed camera modification method is feasible and adaptable to agricultural crop monitoring. Thus, appropriately modified consumer-grade cameras can be a cost-effective replacement for research-grade sensors to rapidly and accurately assess crop growth status.
format Journal Article
id CGSpace173594
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling CGSpace1735942025-10-26T12:56:30Z A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring Wang, Chufeng Zhang, Jian Wu, Hao Liu, Bin Wang, Botao You, Yunhao Tan, Zuojun Xie, Jing You, Liangzhi Zhang, Junqiang Wen, Ping cameras rapeseed sensors crop monitoring aerial photography Near-infrared (NIR) modification of low-cost cameras is considered an important method to acquire high-resolution NIR images on an unmanned aerial vehicle (UAV) platform. However, few studies have examined filter selection methods to modify consumer-grade cameras for UAV-based agricultural crop monitoring. This study addresses a key challenge: how to balance imaging quality with spectral sensitivity when selecting filters for the modification of consumer-grade cameras. To this end, the normalized difference spectral index (NDSI) and the ratio spectral index (RSI) formulations were used to calculate the spectral indices (SIs) from all possible combinations of any two center wavelengths in UAV hyperspectral data. The contour maps of the coefficient of determination (R2) between the SIs and ground-measured rapeseed LAI were then computed to automatically generate the broadband combinations with optimized center wavelengths and effective bandwidths for selecting filters on camera modification. Results showed that a consumer-grade camera (Nikon D7000) modified by the selected filters had performance comparable with a multispectral camera (RedEdge Micasense 3), but slightly worse than a research-grade hyperspectral camera (Nano-Hyperspec®) in terms of SIs for LAI estimation. In addition, the high-resolution images from the modified camera were processed to obtain accurate crop plant height information. The SIs coupled with plant height from the modified camera (rRMSE = 18.1 % for field 1 and 14.3 % for field 2) was found to perform similar to, and in some cases even better than, those from the research-grade multispectral (rRMSE = 17.9 % and 16.7 % for the respective fields) and hyperspectral (rRMSE = 18.8 % for field 1) cameras for UAV-based LAI estimation. The findings from this study indicate that the proposed camera modification method is feasible and adaptable to agricultural crop monitoring. Thus, appropriately modified consumer-grade cameras can be a cost-effective replacement for research-grade sensors to rapidly and accurately assess crop growth status. 2025-03 2025-03-12T21:14:25Z 2025-03-12T21:14:25Z Journal Article https://hdl.handle.net/10568/173594 en Open Access Elsevier Wang, Chufeng; Zhang, Jian; Wu, Hao; Liu, Bin; Wang, Botao; You, Yunhao ; et al. 2025. A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring. Smart Agricultural Technology 10(March 2025): 100830. https://doi.org/10.1016/j.atech.2025.100830
spellingShingle cameras
rapeseed
sensors
crop monitoring
aerial photography
Wang, Chufeng
Zhang, Jian
Wu, Hao
Liu, Bin
Wang, Botao
You, Yunhao
Tan, Zuojun
Xie, Jing
You, Liangzhi
Zhang, Junqiang
Wen, Ping
A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title_full A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title_fullStr A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title_full_unstemmed A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title_short A band selection method for consumer-grade camera modification for UAV-based rapeseed growth monitoring
title_sort band selection method for consumer grade camera modification for uav based rapeseed growth monitoring
topic cameras
rapeseed
sensors
crop monitoring
aerial photography
url https://hdl.handle.net/10568/173594
work_keys_str_mv AT wangchufeng abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT zhangjian abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wuhao abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT liubin abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wangbotao abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT youyunhao abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT tanzuojun abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT xiejing abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT youliangzhi abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT zhangjunqiang abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wenping abandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wangchufeng bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT zhangjian bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wuhao bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT liubin bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wangbotao bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT youyunhao bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT tanzuojun bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT xiejing bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT youliangzhi bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT zhangjunqiang bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring
AT wenping bandselectionmethodforconsumergradecameramodificationforuavbasedrapeseedgrowthmonitoring