Yang Liu , Fuqin Yang , Jibo Yue , Wanxue Zhu , Yiguang Fan , Jiejie Fan , Yanpeng Ma , Mingbo Bian , Riqiang Chen , Guijun Yang , Haikuan Feng
{"title":"利用基于无人机的 RGB 图像中的颜色参数对作物冠层体积进行加权，以估算马铃薯的地上生物量","authors":"Yang Liu , Fuqin Yang , Jibo Yue , Wanxue Zhu , Yiguang Fan , Jiejie Fan , Yanpeng Ma , Mingbo Bian , Riqiang Chen , Guijun Yang , Haikuan Feng","doi":"10.1016/j.compag.2024.109678","DOIUrl":null,"url":null,"abstract":"<div><div>Current techniques to estimate crop aboveground biomass (AGB) across the multiple growth stages mainly used optical remote-sensing techniques. However, this technology was limited by saturation of the canopy spectrum. To meet this problem, this study used digital images obtained by an unmanned aerial vehicle to extract the spectral and structural indicators of the crop canopy in three key potato growth stages. We took the color parameters (CP) of assorted color space transformations as the canopy spectral information, and crop height (CH), crop coverage (CC), and crop canopy volume (CCV) as the canopy structural indicators. Based on the complementary advantages of CP and CCV, we proposed a new metric: the color parameter-weighted crop-canopy volume (CCV<sub>CP</sub>). Results showed that the CH, CCV, and CCV<sub>CP</sub> correlated more strongly with potato AGB during the multi-growth stages than do CP and CC. The hue-weighted crop-canopy volume (CCV<sub>H</sub>) correlated most strongly with the potato AGB among all structural indicators. Using CH was more accurate in estimating potato AGB compared to CP and CC. Combining indicators (CP + CC/CH, CP + CC + CH) improved the accuracy of potato AGB estimation over the multi-growth stages. Except for the CP + CC + CH model, other AGB estimation models produced inaccurate AGB estimation than the models based on CCV and CCV<sub>H</sub>. The AGB estimation accuracy produced by the univariate-based CCV<sub>H</sub> model (R<sup>2</sup> = 0.65, RMSE = 281 kg/hm<sup>2</sup>, and NRMSE = 23.61 %) was comparable to that of the complex model [CP + CC + CH using random forest (RF) or multiple stepwise regression (MSR)]. Compared with CP + CC + CH using RF and MSR, the RMSE decreased and increased by 0.35 % and 4.24 %, respectively. Compared with CP, CP + CC, CP + CH, and CCV, the use of CCV<sub>H</sub> to estimate AGB decreased the RMSE by 10.24 %, 7.42 %, 6.36 %, and 6.33 %, respectively. Meanwhile, the performance of CCV<sub>H</sub> was verified at different stages and among varieties. Thus, this indicator can be used for monitoring potato growth to help guide field production management.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109678"},"PeriodicalIF":7.7000,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Crop canopy volume weighted by color parameters from UAV-based RGB imagery to estimate above-ground biomass of potatoes\",\"authors\":\"Yang Liu , Fuqin Yang , Jibo Yue , Wanxue Zhu , Yiguang Fan , Jiejie Fan , Yanpeng Ma , Mingbo Bian , Riqiang Chen , Guijun Yang , Haikuan Feng\",\"doi\":\"10.1016/j.compag.2024.109678\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Current techniques to estimate crop aboveground biomass (AGB) across the multiple growth stages mainly used optical remote-sensing techniques. However, this technology was limited by saturation of the canopy spectrum. To meet this problem, this study used digital images obtained by an unmanned aerial vehicle to extract the spectral and structural indicators of the crop canopy in three key potato growth stages. We took the color parameters (CP) of assorted color space transformations as the canopy spectral information, and crop height (CH), crop coverage (CC), and crop canopy volume (CCV) as the canopy structural indicators. Based on the complementary advantages of CP and CCV, we proposed a new metric: the color parameter-weighted crop-canopy volume (CCV<sub>CP</sub>). Results showed that the CH, CCV, and CCV<sub>CP</sub> correlated more strongly with potato AGB during the multi-growth stages than do CP and CC. The hue-weighted crop-canopy volume (CCV<sub>H</sub>) correlated most strongly with the potato AGB among all structural indicators. Using CH was more accurate in estimating potato AGB compared to CP and CC. Combining indicators (CP + CC/CH, CP + CC + CH) improved the accuracy of potato AGB estimation over the multi-growth stages. Except for the CP + CC + CH model, other AGB estimation models produced inaccurate AGB estimation than the models based on CCV and CCV<sub>H</sub>. The AGB estimation accuracy produced by the univariate-based CCV<sub>H</sub> model (R<sup>2</sup> = 0.65, RMSE = 281 kg/hm<sup>2</sup>, and NRMSE = 23.61 %) was comparable to that of the complex model [CP + CC + CH using random forest (RF) or multiple stepwise regression (MSR)]. Compared with CP + CC + CH using RF and MSR, the RMSE decreased and increased by 0.35 % and 4.24 %, respectively. Compared with CP, CP + CC, CP + CH, and CCV, the use of CCV<sub>H</sub> to estimate AGB decreased the RMSE by 10.24 %, 7.42 %, 6.36 %, and 6.33 %, respectively. Meanwhile, the performance of CCV<sub>H</sub> was verified at different stages and among varieties. Thus, this indicator can be used for monitoring potato growth to help guide field production management.</div></div>\",\"PeriodicalId\":50627,\"journal\":{\"name\":\"Computers and Electronics in Agriculture\",\"volume\":\"227 \",\"pages\":\"Article 109678\"},\"PeriodicalIF\":7.7000,\"publicationDate\":\"2024-11-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computers and Electronics in Agriculture\",\"FirstCategoryId\":\"97\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S016816992401069X\",\"RegionNum\":1,\"RegionCategory\":\"农林科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"AGRICULTURE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers and Electronics in Agriculture","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S016816992401069X","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURE, MULTIDISCIPLINARY","Score":null,"Total":0}
Crop canopy volume weighted by color parameters from UAV-based RGB imagery to estimate above-ground biomass of potatoes
Current techniques to estimate crop aboveground biomass (AGB) across the multiple growth stages mainly used optical remote-sensing techniques. However, this technology was limited by saturation of the canopy spectrum. To meet this problem, this study used digital images obtained by an unmanned aerial vehicle to extract the spectral and structural indicators of the crop canopy in three key potato growth stages. We took the color parameters (CP) of assorted color space transformations as the canopy spectral information, and crop height (CH), crop coverage (CC), and crop canopy volume (CCV) as the canopy structural indicators. Based on the complementary advantages of CP and CCV, we proposed a new metric: the color parameter-weighted crop-canopy volume (CCVCP). Results showed that the CH, CCV, and CCVCP correlated more strongly with potato AGB during the multi-growth stages than do CP and CC. The hue-weighted crop-canopy volume (CCVH) correlated most strongly with the potato AGB among all structural indicators. Using CH was more accurate in estimating potato AGB compared to CP and CC. Combining indicators (CP + CC/CH, CP + CC + CH) improved the accuracy of potato AGB estimation over the multi-growth stages. Except for the CP + CC + CH model, other AGB estimation models produced inaccurate AGB estimation than the models based on CCV and CCVH. The AGB estimation accuracy produced by the univariate-based CCVH model (R2 = 0.65, RMSE = 281 kg/hm2, and NRMSE = 23.61 %) was comparable to that of the complex model [CP + CC + CH using random forest (RF) or multiple stepwise regression (MSR)]. Compared with CP + CC + CH using RF and MSR, the RMSE decreased and increased by 0.35 % and 4.24 %, respectively. Compared with CP, CP + CC, CP + CH, and CCV, the use of CCVH to estimate AGB decreased the RMSE by 10.24 %, 7.42 %, 6.36 %, and 6.33 %, respectively. Meanwhile, the performance of CCVH was verified at different stages and among varieties. Thus, this indicator can be used for monitoring potato growth to help guide field production management.
期刊介绍：
Computers and Electronics in Agriculture provides international coverage of advancements in computer hardware, software, electronic instrumentation, and control systems applied to agricultural challenges. Encompassing agronomy, horticulture, forestry, aquaculture, and animal farming, the journal publishes original papers, reviews, and applications notes. It explores the use of computers and electronics in plant or animal agricultural production, covering topics like agricultural soils, water, pests, controlled environments, and waste. The scope extends to on-farm post-harvest operations and relevant technologies, including artificial intelligence, sensors, machine vision, robotics, networking, and simulation modeling. Its companion journal, Smart Agricultural Technology, continues the focus on smart applications in production agriculture.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
