Greg S. Biging

OBJECT-BASED DETAILED VEGETATION CLASSIFICATION WITH AIRBORNE HIGH SPATIAL RESOLUTION REMOTE SENSING IMAGERY

In this paper, we evaluate the capability of the high spatial resolution airborne Digital Airborne Imaging System (DAIS) imagery for detailed vegetation classification at the alliance level with the aid of ancillary topographic data. Image objects as minimum classification units were generated through the Fractal Net Evolution Approach (FNEA) segmentation using eCognition software. For each object, 52 features were calculated including spectral features, textures, topographic features, and geometric features. After statistically ranking the importance of these features with the classification and regression tree algorithm (CART), the most effective features for classification were used to classify the vegetation. Due to the uneven sample size for each class, we chose a non-parametric (nearest neighbor) classifier. We built a hierarchical classification scheme and selected features for each of the broadest categories to carry out the detailed classification, which significantly improved the accuracy. Pixel-based maximum likelihood classification (MLC) with comparable features was used as a benchmark in evaluating our approach. The objectbased classification approach overcame the problem of saltand-pepper effects found in classification results from traditional pixel-based approaches. The method takes advantage of the rich amount of local spatial information present in the irregularly shaped objects in an image. This classification approach was successfully tested at Point Reyes National Seashore in Northern California to create a comprehensive vegetation inventory. Computer-assisted classification of high spatial resolution remotely sensed imagery has good potential to substitute or augment the present ground-based inventory of National Park lands.

Extraction of red edge optical parameters from Hyperion data for estimation of forest leaf area index

A correlation analysis was conducted between forest leaf area index (LAI) and two red edge parameters: red edge position (REP) and red well position (RWP), extracted from reflectance image retrieved from Hyperion data. Field spectrometer data and LAI measurements were collected within two days after the Earth Observing One satellite passed over the study site in the Patagonia region of Argentina. The two red edge parameters were extracted with four approaches: four-point interpolation, polynomial fitting, Lagrangian technique, and inverted-Gaussian (IG) modeling. Experimental results indicate that the four-point approach is the most practical and suitable method for extracting the two red edge parameters from Hyperion data because only four bands and a simple interpolation computation are needed. The polynomial fitting approach is a direct method and has its practical value if hyperspectral data are available. However, it requires more computation time. The Lagrangian method is applicable only if the first derivative spectra are available; thus, it is not suitable to multispectral remote sensing. The IG approach needs further testing and refinement for Hyperion data.

Fuzzy classification systems

In this paper it is pointed out that a classification is always made taking into account all the available classes, i.e., by means of a classification system. The approach presented in this paper generalizes the classical definition of fuzzy partition as defined by Ruspini, which is now conceived as a quite often desirable objective that can be usually obtained only after a long learning process. In addition, our model allows the evaluation of the resulting classification, according to several indexes related to covering, relevance and overlapping.

A phenology-based approach to map crop types in the San Joaquin Valley, California

An accurate and timely crop-type map is essential in water planning in California. So far, no effort has been made to effectively and efficiently identify specific crop types on an annual basis in this area. We have explored the potential of Moderate Resolution Imaging Spectroradiometer (MODIS) reflectance images to annually map major crop types in the San Joaquin Valley, California. A phenology-based classification approach has been employed, which has extracted phenological metrics from normalized difference vegetation index (NDVI) profiles and identified crop types based on these metrics using decision trees. According to a comparison with traditional maximum-likelihood classification, this phenology-based approach has shown great advantages when the size of the training set was limited by ground-truth availability and when the central tendency was absent in agricultural systems heavily influenced by human activities.

Snail Density Prediction for Schistosomiasis Control Using Ikonos and ASTER Images

Schistosomiasis is a water-borne parasitic disease endemic in tropical and subtropical areas. Its transmission depends upon the presence of snails, which serve as intermediate hosts for the parasite. Some efforts have been made to classify snail habitats with remotely sensed data, but not to estimate snail abundance that is an important parameter in schistosomiasis transmission modeling. In this research, snail density was predicted by integrating the field survey and satellite images of different spatial resolution. A mountainous environment near Xichang city, in southwest Sichuan province, China, was chosen as the test site. Land-cover and land-use information extracted from 4 m resolution Ikonos data and elevation data derived from ASTER (Advanced Space-borne Thermal Emission and Reflection Radiometer) data were used as reference for scaling up to greater spatial extents. Therefore, we estimated land-cover and land-use fraction data at the 30 m resolution level based on classification results from the Ikonos data. Snail abundance for each 30 m resolution grid was then predicted by regressing field survey data with land-cover and land-use fractions. Subsequently, a snail density map was generated using the territory of each of the over 200 residential groups as a mapping unit. An R 2 of 0.87 was obtained between the average snail density predicted and that surveyed for 19 groups. With such a model, we were able to extrapolate scattered snail abundance surveyed at a limited number of sites to the entire area. Spatial autocorrelation of snail distribution was considered as one of the possible factors in predicting snail density and tested for further model calibration.

Photo ecometrics for forest inventory

Abstract In this paper, we report the results obtained from the application of digital photogrammetry and hyperspectral data analysis for forest inventory purposes. Our long term goal is to provide low-cost yet accurate estimates of as many important forest biophysical parameters as can be measured and inferred with airborne digital cameras. Accuracies of traditional multispectral image analysis algorithms of remotely sensed data are low. Traditional photo interpretation is error prone and expensive. We propose new image analysis strategies that make use of the 3D spatial morphological information from stereo images and the multispectral, texture and contextual information inherent in the imagery. Research on the use of 3D crown shape information in automated tree species recognition has not been reported before. The minimum requirements of image spatial resolution for deriving estimates of tree heights and crown size with high accuracies are not known. With digital photogrammetry, it has been proven that...

Forest Cover Classification by Optimal Segmentation of High Resolution Satellite Imagery

This study investigated whether high-resolution satellite imagery is suitable for preparing a detailed digital forest cover map that discriminates forest cover at the tree species level. First, we tried to find an optimal process for segmenting the high-resolution images using a region-growing method with the scale, color and shape factors in Definiens® Professional 5.0. The image was classified by a traditional, pixel-based, maximum likelihood classification approach using the spectral information of the pixels. The pixels in each segment were reclassified using a segment-based classification (SBC) with a majority rule. Segmentation with strongly weighted color was less sensitive to the scale parameter and led to optimal forest cover segmentation and classification. The pixel-based classification (PBC) suffered from the “salt-and-pepper effect” and performed poorly in the classification of forest cover types, whereas the SBC helped to attenuate the effect and notably improved the classification accuracy. As a whole, SBC proved to be more suitable for classifying and delineating forest cover using high-resolution satellite images.

CLASSIFYING PIXELS BY MEANS OF FUZZY RELATIONS

Most of the classification models assume, in a direct or indirect way, the possibility of a representation of the object to be classified within a good properties space, in which one is able to define some distances. Quite often this representation, which usually is the base of intuitive arguments made by the decision maker, is elaborated from a systematic comparison between different available options. The target of this article is to model a classification problem in the case that a comparative analysis is an essential part of the available information. This will be accomplished by using a description of the option set quality.

Geostatistical analysis of regional differences in stem taper form of Pinus densiflora in central Korea

This study verified regional differences in the stem form of Pinus densiflora Sieb. et Zucc. (red pine) and identified the relationship between stem form and climatic factors in the central region of the Korean peninsula. Regional differences in stem form index at tree base (butt) and top stem section were found. Compared to the stem form in the eastern uplands, the stem form in the western lowlands could be characterized by a more conical butt section and more cylindrical middle and upper section. Through geostatistical analysis of kriging and spatial regression, several climatic factors proved to exert a meaningful influence on stem taper form. On the stem form at the butt section, the precipitation during the late growing season exerts statistically significant effects. High precipitation during the growing season in the western lowland and coastal region causes the stem form at the butt section to be more tapered. On the stem form at the middle and upper section, temperature and precipitation during the growing season, and wind during the late growing season have statistically meaningful influences. High temperature, precipitation, and wind during the growing season in the western lowland and coastal region jointly influence the stem form at the middle and upper sections which result in more cylindrical profiles. This study can be considered an initial investigation into the factors controlling stem form variability in the central region of the Korean peninsula. The results can be used to develop more accurate regional stem taper models needed for reasonable management of red pine stands in different regions.

Improving Fuzzy Classification by Means of a Segmentation Algorithm

In this chapter we consider remotely sensed images, where land surface should be classified depending on their uses. On one hand, we discuss the advantages of the fuzzy classification model proposed by Amo et al. (European Journal of Operational Research, 2004) versus standard approaches. On the other hand, we introduce a coloring algorithm by to Gomez et al. (Omega, to appear) in order to produce a supervised algorithm that takes into account a previous segmentation of the image that pursues the identification of possible homogeneous regions. This algorithm is applied to a real image, showing its high improvement in accuracy, which is then measured.