John A. Thomasson

Combining fuzzy set theory and nonlinear stretching enhancement for unsupervised classification of cotton root rot

Abstract. Cotton root rot is a destructive disease affecting cotton production. Accurate identification of infected areas within fields is useful for cost-effective control of the disease. The uncertainties caused by various infection stages and newly infected plants make it difficult to achieve accurate classification results from airborne imagery. The objectives of this study were to apply fuzzy set theory and nonlinear stretching enhancement to airborne multispectral imagery for unsupervised classification of cotton root rot infections. Four cotton fields near Edroy and San Angelo, Texas, were selected for this study. Airborne multispectral imagery was taken and the color-infrared (CIR) composite images were used for classification. The intensity component was enhanced by using a fuzzy-set based method, and the saturation component was enhanced by a nonlinear stretching image enhancement algorithm. The enhanced CIR composite images were then classified into infected and noninfected areas. Iterative self organization data analysis and adaptive Otsu’s method were used to compare the performance of the proposed image enhancement method. The results showed that image enhancement has improved the classification accuracy of these two unsupervised classification methods for all four fields. The results from this study will be useful for detection of cotton root rot and for site-specific treatment of the disease.

A Wireless GPS System for in-field Cotton Fiber Quality Mapping

Cotton fiber quality maps show spatial variability of fiber quality and thus allow site-specific management of fiber quality at the production level. However, the complex nature of fiber quality and the lack of real-time in-situ fiber quality sensing technologies prevent a straightforward way for fiber quality mapping. In this article, a system which can delineate the harvesting areas of each seed cotton module was described. The system uses a single board computer and a GPS receiver to log the GPS signals while harvesters are picking seed cotton in the field. The wireless communication technology is used to transmit and receive module (or boll buggy) ID numbers when the harvester basket is full and a dump occurs. Ultimately a unique seed cotton module ID can be attached to each continuous harvesting area with a unit of harvesting basket. When seed cotton modules are ginned and classed, the fiber quality information together with the geographic information of each module can be integrated into farmer’s GIS system for fiber quality mapping at the module level.

Plant health sensing system for determining nitrogen status in plants

A plant health sensing system was developed for determining nitrogen status in plants. The system consists of a multi-spectral optical sensor and a data-acquisition and processing unit. The optical sensor’s light source provides modulated panchromatic illumination of a plant canopy with light-emitting diodes, and the sensor measures spectral reflectance through optical filters that partition the energy into blue, green, red, and near-infrared wavebands. Spectral reflectance of plants is detected in situ, at the four wavebands, in real time. The data-acquisition and processing unit is based on a single board computer that collects data from the multi-spectral sensor and spatial information from a global positioning system receiver. Spectral reflectance at the selected wavebands is analyzed, with algorithms developed during preliminary work, to determine nitrogen status in plants. The plant health sensing system has been tested primarily in the laboratory and field so far, and promising results have been obtained. This article describes the development, theory of operation, and test results of the plant health sensing system.

Remote Sensing in Dryland Cotton: Relation to Yield Potential and Soil Properties

The use of soil and topography information to explain crop yield variation across fields is often applied for crop management purposes. Remote sensed data is a potential source of information for site-specific crop management, providing both spatial and temporal information about soil and crop condition. Studies were conducted in a 104-acre (42-hectare) dryland cotton field in 2001 and 2002 in order to (1) qualitatively assess the spatial variability of soil physical properties from kriged estimates, (2) compare actual yields with normalized difference vegetation reflectance indices (NDVI) obtained from multispectral imagery and from in situ radiometer data, and (3) predict site-specific cotton yields using a crop simulation model, GOSSYM. An NDVI map of soybean in 2000 obtained from a multispectral image was used to establish four sites in each low, medium and high NDVI class. These 12 sites were studied in 2001 and 12 more sites selected at random were studied in 2002 (n = 24). Site-specific measurements included leaf area index (LAI), canopy hyperspectral reflectance, and three-band multispectral image data for green, red, and near-infrared reflectance wavebands at spatial resolutions of 2 m in 2001 and 0.5 m in 2002. Imagery was imported into the image analysis software Imagine (ERDAS, v. 8.5) for georegistration and image analysis. A 6x6 pixels (144 m2) area of interest was established on top of each field plot site and digital numbers (DN) from reflectance imagery were extracted from each band for derivation of NDVI maps for each of four sampling dates. Lint yield from each plot site was collected by hand and also by a cotton picker equipped with AgLeader yield monitor and OmniStar differential global positioning system. We found plant height, leaf area index, and lint yield were closely associated with NDVI maps and with NIR band values acquired from either an aircraft or handheld (GER-1500) sensor during peak bloom in mid July. Results indicate NDVI and NIR bands could be used to produce estimated field maps of plant height, leaf area index and yield, which offer a potentially attractive mid-season management tool for site specific farming in dryland cotton.

Remote sensing and weather information in cotton yield prediction

If farmers could predict yield on a spatially variable basis, they could better understand risks and returns in applying costly inputs such as fertilizers, etc. To this end, several remotely sensed images of a cotton field were collected during the 2002 growing season, along with daily high and low temperatures. Image data were converted to normalized-difference vegetation index (NDVI), and temperature data were used to normalize NDVI changes over periods between image collections. Remote-sensing and weather data were overlaid in a geographic information system (GIS) with data from the field: topography, soil texture, and historical cotton yield. All these data were used to develop relationships with yield data collected at the end of the 2002 season. Stepwise regression was conducted at grid-cell sizes from 10 m square (100 m2) to 100 m square (10,000 m2) in 10-m increments. Relationships at each cell size were calculated with data available at the beginning of the season, at the first image date, at the second image date, and so on. Stepwise linear regression was used to select variables at each date that would constitute an appropriate model to predict yield. Results indicated that, at most dates, model accuracy was highest at the 100-m cell size. Remotely sensed data combined with weather data contributed much information to the models, particularly with data collected within 2.5 months of planting. The most appropriate model had an R2 value of 0.63, and its average prediction error was about 0.5 bale/ha (0.2 bale/ac, or roughly 100 lb/ac).

Spatial validation of cotton simulation model in relation to soils and multispectral imagery

Field studies were conducted in 1998 and 1999 in Livingston Field at Perthshire Farm, Bolivar County which is located in west-central Mississippi along the Mississippi River. It is a 162 ha field and has a 2-m elevation range. The dominant soil series of the field are Commerce silt loam, Robinsonville fine sandy loam and Souva silty clay loam. The objectives of the study were to (1) compare GOSSYM simulated yield with actual yield, (2) study spatial and temporal pattern of cotton crop across two growing seasons using multispectral imagery, 3) predict field based lint yield from remote sensed data, and determine age of the crop most suitable for aerial image acquisition in predicting yield and/or discriminating differences in cotton growth. Two transects were selected for GOSSYM study, each containing twelve sites. A 1-m length of single row plot was established at each profile. Plant mapping was done five times in 1998 and seven times in 1999 growing seasons. GOSSYM simulation runs were made for each profile and compared with actual crop parameters using root mean square error (RMSE). Results based on averaging common soil mapping units indicate that GOSSYM accuracy in predicting yield varied from 0.45 bales acre-1 to 0.61 bales acre-1. To monitor and estimate the biophysical condition of the cotton crop, airborne multispectral images were acquired on 10 dates in 1998 and 17 dates in 1999 from April to September. In both years site-specific yield and normalized difference vegetation index (NDVI) were significantly (p < 0.0001) correlated in July. Changes in NDVI in 1999 across sampling dates for the different sites showed the least distinctiveness due to somewhat wetter weather conditions, as compared to drier weather in 1998. Crop growing in or near the drainage areas were low in NDVI and lint yield. Multispectral images acquired between ~ 300 - 600 growing degree days above 60°C (GDD60) may be useful decision tools for replanting certain areas of the field, especially in dry weather conditions when variability in crop growth pattern is enhanced due to spatial variability in soil texture, which influences the capacity of a soil to hold moisture and to release it to plants for growth. Results suggest that 2-3 multispectral images acquired between 800 and 1500 GDD60 can be used to monitor crop health and predict yield in a normal weather condition.