• Korean Journal of Remote Sensing
• /
• v.24 no.6
• /
• pp.551-558
• /
• 2008

It is well known that the protein content of rice grain is an indicator of taste of cooked rice in the countries where people as the staple food. Ground-based optical sensing over the crop canopy would provide information not only on the mass of plant body which reflects the light, but also on the crop nitrogen content which is closely related to the greenness of plant leaves. The vegetation index has been related to crop variables such as biomass, leaf nitrogen, plant cover, and chlorophyll in cereals. The objective of this study was to investigate the correlation between GNDVI and NDVI values, and grain protein content at different dates and to estimate the grain protein content using G(NDVI) values. We measured Green normalized difference vegetation index [$GNDVI=({\rho}0.80{\mu}m-{\rho}0.55{\mu}m)/({\rho}0.80{\mu}m+{\rho}0.55{\mu}m)$] and [$GNDVI=({\rho}0.80{\mu}m-{\rho}0.68{\mu}m)/({\rho}0.80{\mu}m+{\rho}0.68{\mu}m)$] by using two different active sensors. The study was conducted during the rice growing season for three years from 2005 through 2007 at the experimental plots of National Institute of Agricultural Science and Technology. The experiments were carried out by randomized complete block design with the application of four levels of nitrogen fertilizers(0, 70, 100, 130kg N/ha) and the same amount of phosphorous and potassium content of the fertilizers. After heading stage, relationships between GNDVI of rice canopy and grain protein content showed the highly positive correlation at different dates for three years. GNDVI values showed higher correlation coefficients than that of NDVI during growing season in 2005-07. The correlation between GNDVI values at different dates and grain protein contents was highly correlated at early July. We attempted to estimate the grain protein content at harvesting stage using GNDVI values from early July for three years. The determination coefficients of the linear model by GNDVI values were 0.9l and the measured and estimated grain protein content at harvesting stage using GNDVI values highly correlated($R^2=0.96^{***}$). Results from this study show that GNDVI appeared very effective to estimate leaf nitrogen and grain protein content of rice canopy.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.151-154
• /
• 2008

Information on the area and spatial distribution of crop fields is needed for biomass production, arrangement of water resources, trace gas emission estimates, and food security. The present study aims to monitor crops status during the growing season by estimating its aboveground biomass and leaf area index (LAI) from field reflectance taken with a hand-held radiometer. Field reflectance values were collected over specific spectral bandwidths using a handheld radiometer(LI-1800). A methodology is described to use spectral reflectance as indicators of the vegetative status in crop cultures. Two vegetation indices were derived from these spectral measurements. In this paper, first we analyze each spectral reflectance characteristics of vegetation in the order of growth stage. Vegetation indices (NDVI, GNDVI) were calculated from crop reflectance. And assess the nature of relationships between LAI and VI, as measured by the in situ NDVI and GNDVI. Among the two VI, NDVI showed predictive ability across a wider range of LAI than did GNDVI. Specific objectives were to determine the relative accuracy of these two vegetation indices for predicting LAI. The results of this study indicated that the NDVI and GNDVI could potentially be applied to monitor crop agriculture on a timely and frequent basis.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.40 no.4
• /
• pp.196-202
• /
• 2020

Rye, whole-crop barley and Italian Ryegrass are major winter forage species in Korea, and yield monitoring of winter forage species is important to improve forage productivity by precision management of forage. Forage monitoring using Unmanned Aerial Vehicle (UAV) has offered cost effective and real-time applications for site-specific data collection. To monitor forage crop by multispectral camera with UAV, we tested four types of vegetation index (Normalized Difference Vegetation Index; NDVI, Green Normalized Difference Vegetation Index; GNDVI, Normalized Green Red Difference Index; NGRDI and Normalized Difference Red Edge Index; NDREI). Field measurements were conducted on paddy field at Naju City, Jeollanam-do, Korea between February to April 2019. Aerial photos were obtained by an UAV system and NDVI, GNDVI, NGRDI and NDREI were calculated from aerial photos. About rye, whole-crop barley and Italian Ryegrass, regression analysis showed that the correlation coefficients between dry matter and NDVI were 0.91~0.92, GNDVI were 0.92~0.94, NGRDI were 0.71~0.85 and NDREI were 0.84~0.91. Therefore, GNDVI were the best effective vegetation index to predict dry matter of rye, wholecrop barley and Italian Ryegrass by UAV system.

• Proceedings of the KSRS Conference
• /
• 2007.03a
• /
• pp.43-48
• /
• 2007

본 연구는 벼 군락의 분광반사율 지표를 측정할 수 있는 인공광원을 사용하는 능동형 광학 센서인 GNDVI 를 활용하여 생육시기별 식생지수와 엽 질소함량과의 관계를 구명하여 벼 군락의 엽 질소함량을 추정하고자 하였다. 농업과학기술원 답작 포장에서 공시품 종인 추청벼를 이용하여 난괴법 3반복으로 시험구 배치를 하고 질소 4수준 3반복으로 실험을 수행하였다. 벼 생육시기별 GNDVI 와 엽 질소함량과의 관계를 2년 간 (2005, 2006)의 자료를 통하여 분석해 보았다. 2005년의 경우 벼 생육시기동안 9시기의 GNDVI값과 그 당시 시료를 샘플링하여 분석한 엽 질소함량과의 관계를 벼 출수전과 출수후로 구분하여 분석해 본 결과 GNDVI 값은 출수전 ($r=0.78^{***}$ n=60) 보다 출수후 ($r=0.89^{***}$, n=59) 가 엽 질소함량과의 상관계수가 높았다. 2006년은 20시기동 안 생육시기별 식생지수와 엽 질소함량과의 상관 분석한 결과 착근기 (6월 5일) $r=0.84^{***}$, 유수분화기 (7월 11일) $r=0.95^{***}$, 출수기 (8 월 16일) $r=0.87^{***}$, 수확기 (10월 13일) $r=0.90^{***}$ 으로 출수전의 경우 7월 11일이 상관계수가 가장 높았고， 이 결과는 2005년 동일시기 (7월 11일) 식생지수와 벼 엽 질소함량과의 상관계수가 가장 높았던 ($r=0.91^{***}$) 것과 일치하였다. 벼 생육시기 변화에 따른 식생지수와 엽 질소 흡수량과의 관계를 살펴보았는데 벼 출수전의 경우 GNDVI는 7월 11일 에 엽 질소 흡수량과의 상관계수 ($r=93^{***}$)가 가장 높은 결과를 보였고 출수후의 경우에는 시기에 따라 상관계수가 고르게 높게 나타났다. 엽 질소함량과의 상관관계가 높았던 2005년，2006년 7월 11일 식생지수 데이터를 함께 이용하여 엽 질소함량과의 관계를 추정식으로 작성하였다. GNDVI를 이용하여 2005년과 2006년 실측한 엽 질소함량 값과 추정 값을 비교해 본 결과 2005년과 2006년의 결정계수가 각각 0.88, 0.94로 2006년이 더 예측률이 높게 나타났다. GNDVI 값을 이용하여 엽 질소함량 추정값과 실측값을 비교해 본 결과 결정계수가 0.86으로 추정값과 실측값이 근접하게 분포하였다.

• Journal of Cadastre & Land InformatiX
• /
• v.50 no.1
• /
• pp.201-214
• /
• 2020

This study aimed to early detect damaged trees by pine wilt disease using the vegetation indices of UAV images. The location data of 193 pine wilt disease trees were constructed through field surveys and vegetation index analyses of NDVI, GNDVI, NDRE and SAVI were performed using multi-spectral UAV images at the same time. K-Means algorithm was adopted to classify damaged trees and confusion matrix was used to compare and analyze the classification accuracy. The results of the study are summarized as follows. First, the overall accuracy of the classification was analyzed in order of NDVI (88.04%, Kappa coefficient 0.76) > GNDVI (86.01%, Kappa coefficient 0.72) > NDRE (77.35%, Kappa coefficient 0.55) > SAVI (76.84%, Kappa coefficient 0.54) and showed the highest accuracy of NDVI. Second, K-Means unsupervised classification method using NDVI or GNDVI is possible to some extent to find out the damaged trees. In particular, this technique is to help early detection of damaged trees due to its intensive operation, low user intervention and relatively simple analysis process. In the future, it is expected that the utilization of time series images or the application of deep learning techniques will increase the accuracy of classification.

• Journal of Biosystems Engineering
• /
• v.34 no.4
• /
• pp.269-277
• /
• 2009

In this research, the vegetation growth and rice taste of the liquid fertilizer applied fields (LF) were compared with those of chemical fertilizer applied fields(CF) in order to confirm the possibility of methane fermentation digested sludge as liquid fertilizer using precision agriculture and remote sensing technology. In panicle initiation stage, the vegetation growth at LF was 60%~80% of it at CF and there were significant difference of nitrogen contents between CF and LF. The estimation model of nitrogen contents was established by GNDVI (R=0.607, RMSE=$1.04\;g/m^2$, n=36, p<0.01). In heading stage, vegetation growth at LF went close to it at CF as ratio of 80%~95%. The nitrogen content estimation model was also established (R=0.650, RMSE=$1.73\;g/m^2$, n=35, p<0.01) and there were significant difference of spatial variability between LF and CF. There were not significant difference of rice taste and it's elements, when three samples, which were more than twice of standard deviation, were excepted. The protein contents estimation model using GNDVI of before harvesting (R=0.700, RMSE=0.470%, n=29, p<0.01) were more suitable to predict the protein contents at harvesting comparing with it of heading stage(R=0.610, RMSE=0.521%, n=29, p<0.01).

• Korean Journal of Soil Science and Fertilizer
• /
• v.47 no.3
• /
• pp.217-224
• /
• 2014

Ground-based remote sensing can be used as one of the non-destructive, fast, and real-time diagnostic tools for predicting yield, biomass, and nitrogen stress during growing season. The objectives of this study were: 1) to assess biomass and nitrogen (N) status of tobacco (Nicotiana tabacum L.) plants under N stress using ground-based remote sensors; and 2) to evaluate the feasibility of spectral reflectance indices for estimating an application rate of N and predicting yield of tobacco. Dry weight (DW), N content, and N uptake at the 40th and 50th day after transplanting (DAT) were positively correlated with chlorophyll content and normalized difference vegetation indexes (NDVIs) from all sensors (P<0.01). Especially, Green NDVI (GNDVI) by spectroradiometer and Crop Circle-passive sensors were highly correlated with DW, N content and N uptake. The yield of tobacco was positively correlated with canopy reflectance indices measured at each growth stage (P<0.01). The regression of GNDVI by spectroradiometer on yield showed positively quadratic curve and explained about 90% for the variability of measured yield. The sufficiency index (SI) calculated from data/maximum value of GNDVI at the $40^{th}$ DAT ranged from 0.72 to 1.0 and showed the same positively quadratic regression with N application rate explaining 84% for the variability of N rate. These results suggest that use of reflectance indices measured with ground-based remote sensors may assist in determining application rate of fertilizer N at the critical season and estimating yield in mid-season.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.218-222
• /
• 2006

본 연구에서는 광학센서를 이용한 벼 군락의 질소수준별 생육단계별 식생지수와 쌀 단백질함량과의 관계를 구명하여 쌀 단백질함량을 추정하는 것을 목적으로 하였다. 질소의 경우 0, 7, 10, 13kg/10a등 4수준으로 범위를 두고 처리하여 인공광원을 사용하는 2종의 능동형 광학(G)NDVI 센서를 이용하여 벼 군락의 반사특성을 측정하였고, 동시에 식물체의 생육량, 엽면적지수, 엽 질소함량 등을 분석하였다. 생육단계에 따른 식생지수 변화를 분석해 본 결과 (G)NDVI값은 이앙기 이후 급속히 증가하다가 수잉기 전후로 수확기에 이르기까지 감소하는 경향을 보였다. 질소 수준에 따른 식생지수 변화의 경우 무처리구를 제외하고는 처리수준별 G(NDVI)값이 큰 변이가 나타나지는 않았지만, 처리 수준에 따라 일정하게 식생지수 차이를 보였다. (G)NDVI값 과 엽질소 함량과의 시기별 상관분석 결과 유효분얼기, 유수형성기 보다는 출수기, 결실기에 엽 질소 함량과의 상관이 더 높게 나타났고, GNDVI값이 NDVI값보다 상관이 더 높게 나타났다. 출수 후 쌀 단백질 함량과 엽 질소 함량과의 관계를 조사해보았는데 높은 정의 상관관계($r=0.96^{**}$)를 보였다. 출수기에서 수확기까지 자료를 이용한 각 시기별 G(NDVI)값과 쌀 단백질 함량과의 상관분석 결과 수확기에 가까울수록 상관계수가 높게 나타났다. GNDVI값을 이용한 수확기 쌀 단백질 함량 추정식($R^2=0.92$)을 작성하였고, 쌀 단백질 함량 추정값과 실측값을 비교해보았더니 1:1선에 근접하게 분포하였다($R^2=0.90$).

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.21 no.4
• /
• pp.327-336
• /
• 2019

Soybean is one of the most important crops of which the grains contain high protein content and has been consumed in various forms of food. Soybean plants are generally cultivated on the field and their yield and quality are strongly affected by climate change. Recently, the abnormal climate conditions, including heat wave and heavy rainfall, frequently occurs which would increase the risk of the farm management. The real-time assessment techniques for quality and growth of soybean would reduce the losses of the crop in terms of quantity and quality. The objective of this work was to develop a simple model to estimate the growth of soybean plant using a multispectral sensor mounted on a rotor-wing unmanned aerial vehicle(UAV). The soybean growth model was developed by using simple linear regression analysis with three phenotypic data (fresh weight, dry weight, leaf area index) and two types of vegetation indices (VIs). It was found that the accuracy and precision of LAI model using GNDVI (R²= 0.789, RMSE=0.73 ㎡/㎡, RE=34.91%) was greater than those of the model using NDVI (R²= 0.587, RMSE=1.01 ㎡/㎡, RE=48.98%). The accuracy and precision based on the simple ratio indices were better than those based on the normalized vegetation indices, such as RRVI (R²= 0.760, RMSE=0.78 ㎡/㎡, RE=37.26%) and GRVI (R²= 0.828, RMSE=0.66 ㎡/㎡, RE=31.59%). The outcome of this study could aid the production of soybeans with high and uniform quality when a variable rate fertilization system is introduced to cope with the adverse climate conditions.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.23 no.2
• /
• pp.21-35
• /
• 2020

Vegetation information is a very important factor used in various fields such as urban planning, landscaping, water resources, and the environment. Vegetation varies according to canopy density or chlorophyll content, but vegetation vitality is not considered when classifying vegetation areas in previous studies. In this study, in order to satisfy various applied studies, a study was conducted to set a threshold value of vegetation index considering vegetation vitality. First, an eBee fixed-wing drone was equipped with a multi-spectral camera to construct optical and near-infrared orthomosaic images. Then, GIS calculation was performed for each orthomosaic image to calculate the NDVI, GNDVI, SAVI, and MSAVI vegetation index. In addition, the vegetation position of the target site was investigated through VRS survey, and the accuracy of each vegetation index was evaluated using vegetation vitality. As a result, the scenario in which the vegetation vitality point was selected as the vegetation area was higher in the classification accuracy of the vegetation index than the scenario in which the vegetation vitality point was slightly insufficient. In addition, the Kappa coefficient for each vegetation index calculated by overlapping with each site survey point was used to select the best threshold value of vegetation index for classifying vegetation by scenario. Therefore, the evaluation of vegetation index accuracy considering the vegetation vitality suggested in this study is expected to provide useful information for decision-making support in various business fields such as city planning in the future.