-
[show abstract]
[hide abstract]
ABSTRACT: Recent advances in spectral imaging technology have enabled the development of models that estimate various crop parameters
from spectral imagery data. We developed partial least square (PLS) models to predict fruit yield of Satsuma mandarin using
airborne hyperspectral imagery obtained several months before harvesting. Hyperspectral images in the 72 visible and near-infrared
(NIR) wavelengths (from 407 to 898nm) were acquired over a citrus orchard during the early growing seasons of 2003, 2004
and 2005. The canopy features of individual trees were identified using pixel-based average spectral reflectance values for
all 72 wavelengths from the acquired images. The acquired canopy features were then used as prediction variables to develop
yield prediction models. These were developed using three techniques: (1) normalized difference vegetation index (NDVI), simple
ratio (SR) and photochemical reflectance index (PRI), (2) conventional multiple linear regression (MLR) models, and (3) PLS
regression models. As we intended to predict yield several months before the harvesting season (generally late December),
the conventional techniques (vegetation indices and MLR) did not predict well. In contrast, PLS models gave successful predictions
for the three years. These results confirmed the hypothesized correlation between canopy features and citrus yield. The successful
forecasting of yields several months or even one year ahead of the harvest season is expected to contribute to planning harvest
schedules, generating prescription maps for dealing with fluctuations of yield in specific trees, control measures, and management
practices.
Precision Agriculture 04/2012; 8(3):111-125. · 1.55 Impact Factor
-
Ecological Informatics 01/2008; · 1.43 Impact Factor
-
Ecological Informatics. 01/2008; 3:237-244.
-
[show abstract]
[hide abstract]
ABSTRACT: A cropping system is usually characterized by continuous spatio-temporal vegetation variability. Vegetation variability can be detected by changes in several vegetation parameters defined according to purpose. Estimation of these vegetation parameters has been made possible by calculating various vegetation indices (VIs), usually by ratioing, differencing, ratioing differences and sums, or by forming linear combinations of spectral band data. Spectrometers or sensors have been used to acquire visible and infrared spectral properties of vegetation. This paper presents a ground-based hyperspectral imaging system for characterizing vegetation spectral features. The hyperspectral sensor used was a ground-based line sensor, ImSpector (V10-12-102), with a nominal spectral resolution of 1.5–2 nm and a wavelength range of 360–1010 nm. A graphical user interface (GUI) was developed in a MATLAB environment to aid in processing and analysis of acquired multidimensional spectral image data. Issues that arise when applying the imaging system to a particular field include acquiring hyperspectral images, selecting appropriate vegetation features or VIs, and quantifying the selected vegetation features or indices with the GUI developed. Vegetation features extracted by the proposed imaging system contribute not only to monitoring vegetation variability in crop systems, but also provide a potential source of relevant variables that can be used to estimate various vegetation parameters. A study that was set up to investigate the alternate bearing phenomenon of citrus trees illustrates the basic elements of the proposed approach.
Computers and Electronics in Agriculture. 01/2008;
-
[show abstract]
[hide abstract]
ABSTRACT: Reliable information on weed abundance and distribution within fields is essential for weed management in agricultural systems. Such information is necessary to adopt localized and variable rates of herbicide spraying, thus reducing chemical waste, crop damage, and environmental pollution. This paper examined the potential of airborne multispectral imagery to discriminate and map weed infestations in an experimental citrus orchard in Japan. Using an airborne digital sensor, multispectral imagery was acquired over the study site on 10 April 2003. The obtained reflectance imagery was analyzed using an image object-based approach in eCognition. After creating image objects on the image, the spectral information for weeds and citrus, represented by corresponding selected sample image objects, was extracted. Significant differences in the spectral characteristics between weeds and citrus were observed in each of the red, green, and blue wavebands. The simple average values of these wavebands were used to classify image objects with the nearest neighbor algorithm. Maps were generated with different classes or levels of class groups. A subsequent accuracy assessment demonstrated that the weeds were successfully discriminated from other image objects with a classification accuracy of 99.07%. Therefore, maps generated based on the classification result could provide valuable information for developing a site-specific weed management program for the study orchard.
Weed Biology and Management 02/2007; 7(1):23 - 30. · 0.71 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We developed a multiple linear regression methodology for estimating acorn yield of Quercus serrata from airborne multi-spectral images. Using the models developed, we estimated the spatial distribution of acorn yields on the aerial images. We also calculated spatial autocorrelation from the estimated spatial distribution of yields, and evaluated the spatial pattern by comparison with the simulation output from Satake and Iwasa's [Satake, A., Iwasa, Y., 2002a. Spatially limited pollen exchange and a long-range synchronization of trees. Ecology 83, 993–1005] theoretical models, which assume internal allocation and pollen exchange between trees within a finite range.A significant correlation was found between logarithmic acorn yield and the multi-spectral data observed on June 6, 2003 (R2 = 0.37, p < 0.05) and on May 26, 2004 (R2 = 0.44, p < 0.01); these relationships were also confirmed by leave-one-out cross-validation (p < 0.05). Through an image segmentation procedure, approximately 5700 canopies were identified. The acorn yields of these canopies were estimated by applying the model developed. The correlation coefficients calculated from the estimated spatial distributions of yield decreased as distance increased. Our experimentally estimated spatial distribution agrees with the patterns of the correlograms of spatial pattern derived from Sakate and Iwasa's models. Combinations of our findings and their simulation results suggest that an endogenous mechanism may drive the masting of Q. serrata.
Ecological Modelling.
