Digital Commons Network^™

Statistical Analysis Of Texture In Landsat Images Of The United Kingdom, R. M. Lee, P. Gray, M. E. Barnett

LARS Symposia

The agricultural regions of England appear in LANDSAT IMAGES as extended mosaic patterns of small fields. Individual fields, especially in the hillier north and west of the country may be as small as a few pixels and thus it is the field patterns themselves which make up the texture of the image. We have been concerned with the problem of mapping the average orientation of fields (a secondary indicator for geomorphology) and the distribution of field size.

Two approaches are being used:

(i) Holographic derivation of local autocorrelation information.

(ii) Digital extraction of statistical parameters.

For our test area in …

Digital Analysis Of Human Impact On Tropical Vegetation, Suzanne Textor, Jerry C. Coiner

LARS Symposia

The purpose of these studies was to develop digital-processing techniques for LANDSAT Multispectral Scanner (MSS) data to use in monitoring the effect of human activity upon Amazonian vegetation. The procedure involves: (1) development of signatures for major vegetation communities, and (2) measurement of the extent of vegetation modification by human groups.

Study sites were located in lowland Peru and in the Amazon basin of Brazil, south of Manaus, where tropical ombrophilous forest is the dominant natural vegetation. Two groups are currently modifying the natural vegetation: native Indiana populations and representatives of the "modern" national economies. The mode of occupance of …

Landsat Estimation With Cloud Cover, George A. Hanuschak

LARS Symposia

The problem addressed in this paper is crop acreage estimation techniques which utilize LANDSAT I imagery that is not cloud free. Several statistical techniques are proposed that would allow inferences about the population even if cloud cover is an extensive problem. These techniques are entirely dependent upon a random sample of ground data; from the total population of interest.

Pacific Northwest Resources Inventory Demonstration, J. D. Nichols

LARS Symposia

The Pacific Northwest Land Resource Inventory Demonstration project is being carried out jointly by NASA, the U.S. Department of the Interior (USDI) and Pacific Northwest Regional Commission (PNRC) through the technical capability provided by NASA, USDI and contractor support. The project is designed to demonstrate to users from state and local agencies in Washington, Oregon, and Idaho the cost effective role that LANDSAT derived information can play in natural resource planning and management when properly supported by ground data and aircraft data.

The Tasselled Cap -- A Graphic Description Of The Spectral-Temporal Development Of Agricultural Crops As Seen By Landsat, R. J. Kauth, G. S. Thomas

LARS Symposia

The time trajectories of agricultural data points as seen in LANDSAT signal space form a pattern suggestive of a tasselled woolly cap. Using this easily visualized three dimensional construct most of the important phenomena of crop development and observation variables are pointed out, named, discussed and measured. The important crop phenomena described are the distribution of signals from bare soil, the processes of green development, of yellow development, of shadowing and of harvesting. The important external variables include view angle, sun angle, atmospheric haze, and atmospheric water vapor.

The development of a quantitative picture of the tasselled cap depends upon …

The Earth Resources Interactive Processing System (Erips) Image Data Access Method (Idam), A. E. Pape, D. L. Truitt

LARS Symposia

The Image Data Access Method (IDAM) was developed for NASA at the Johnson Space Center as the access method for the Earth Resources Interactive Processing System (ERIPS). It was designed to satisfy the need for an image data storage technique which is very efficient with storage space and data access time, but which also provides direct retrieval of all of the image data and easy programmer utilization. A set of eight user macros have been provided which satisfy the utilization requirement, and the result has proven to be a very successful image data storage and retrieval technique.

The Application Of A Parallel Processing Computer In The Large Area Crop Inventory Experiment, Sherwin Ruben, John C. Lyon, Matthew J. Quinn

LARS Symposia

The Large Area Crop Inventory Experiment (LACIE) is a joint investigation by NASA, USDA, and NOAA to determine the practicality and utility of computer analyzed remotely sensed data in crop forecasting. LANDSAT imagery combined with NOAA-supplied meteorological data and ground truth history are the principal sources of LACIE information. The LACIE responsibility of NASA-Johnson Space Center (JSC) includes a Classification and Mensuration Subsystem (CAMS) which is tailored to the production problem presented by LACIE requirements to classify large numbers of fundamentally similar regions in the same manner. In short, LACIE and CAMS represent an essential change from research and development …

General Polygons Used To Determine Training And Test Areas In Digital Remote Sensing Imagery, Ross H. Hieber

LARS Symposia

Polygonal fields specified by the (x, y) coordinates of the polygon vertices offer several advantages over the more conventional rectangular are as specified by the beginning and ending scan line numbers and inter-line point numbers.

Processing Remotely Sensed Data With Array Processors, A. S. Margulies

LARS Symposia

Array processors have been used extensively in military applications involving sonar and radar signal processing, but they have not been as widely employed in image processing applications. The constraints of limited word-size and limited programmability which previously made array processors unattractive have been mitigated with the architecture of the modern machines and the low cost of these modern processors relative to the cost of large scale computers. The array processors are not general purpose computers. Instead, they are extremely fast specialized processors in a parallel architecture with a memory structure matched to the processor capabilities. This arrangement is less flexible …

Analysis Of Geophysical Remote Sensing Data Using Multivariate Pattern Recognition Techniques, Paul E. Anuta, Hans Hauska, Donald W. Levandowski

LARS Symposia

Multivariate statistical pattern recognition techniques have been widely used in the analysis of multispectral scanner remote sensing data for crop surveys, forest mapping, land use surveys and in many other applications. These applications are restricted basically to surface cover reflectance and emissivity phenomena. In the study described in this paper multivariate analysis techniques were applied to geophysical remote sensing data which measures phenomena occurring beneath the surface of the earth. Three types of geophysical data: magnetic anomaly, induced pulse transient, and gamma ray data were digitized, registered and analyzed to observe relationships to known geology. In addition several types of …

Procams: A Second Generation Multispectral-Multitemporal Data Processing System For Agricultural Mensuration, Jon D. Erickson, Richard F. Nalepka

LARS Symposia

A prototype operational data-processing system has been defined, implemented, and tested by Multispectral Analysis Section personnel of the Environmental Research Institute of Michigan (ERIM). This system has been designed for the classification and mensuration of agricultural crops through the use of data provided by the LANDSAT satellite scanner. The specific crops for which the system was designed are the small grains including wheat, rye, oats, and barley, although the system as designed is not limited to these particular crops.

The processing system, known as PROCAMS (Prototype Classification and Mensuration System), has been built based on the experience gained and to …

A Remote Sensing-Aided System For Evaporation And Watershed-Wide Evapotranspiration Estimation, Siamak Khorram, Randall W. Thomas

LARS Symposia

Timely, spatial, and relatively inexpensive remotely sensed data when combined with other conventional ground meteorological data can potentially provide accurate, location specific estimates of evaporation and evapotranspiration. Reliable evaporation data are required for planning, designing and operating reservoirs, ponds, shipping canals, and irrigation and drainage systems. Data on evapotranspiration are useful for estimating irrigation requirements, rainfall disposition, yield of ground water basins, water yields from mountain watershed, streamflow depletion in river basins, and in crop production management.

The approach uses a multistage, multiphase sampling technique employing three increasingly involved levels of information. The input variables for level I models are …

Determination Of Planimetric Features By Interactive Image Processing, John Y. C. Wang

LARS Symposia

In aerial photography since roads and other man-made structures are small compared to natural features, their edges are usually indistinct from the background and are often undetected by automatic systems. A major problem in the development of such systems is the delineation of boundaries around areas containing features of interest. The computer is often unable to accomplish the detection process alone. Therefore, it is reasonable to combine the superior pattern recognition abilities of a human being with the computational power of a digital computer to form an interactive system.

In this paper, an interactive feature extraction system is discussed. The …

Linear Dimensionality Of Landsat Agricultural Data With Implications For Classification, S. G. Wheeler, P. N. Misra, Q. A. Holmes

LARS Symposia

A model for the LANDSAT multispectral scanner data, representing a generalization of the commonly used Gaussian model, has been formulated and analyzed. The model hypothesizes that the data for different crop types essentially lie on distinct hyperplanes in the feature space. Tests of this model reveal that: (1) the agricultural data from any single acquisition (i.e., four-channel) of LANDSAT are essentially two dimensional, regardless of the crop type; and (2) the data from different sites and different stages of crop development all lie on planes which are parallel. These findings have significant implications for data display, classification, feature extraction, and …

Classification By Clustering, Alex Pentland

LARS Symposia

Conventional classification procedures have several difficulties which sometimes limit the usefulness of computer aided analysis techniques on multispectral scanner data. In order to minimize some of these problems, the clustering algorithm used at ERIM (called CLUSTR) was adapted for use as a classifier. Briefly, the technique devised is to cluster the scene, assigning each pixel to a cluster, and then to identify the crop type of the clusters by examining training areas to determine the crop type of pixels assigned to each cluster. In this manner, the classification of each pixel to a particular crop class is accomplished.

Number Of Signatures Necessary For Accurate Classification, W. Richardson, A. Pentland, R. Crane, H. Horwitz

LARS Symposia

This paper presents a procedure for determining the number of signatures to use in classifying multispectral scanner data. A large initial set of signatures is obtained by clustering the training points within each category (such as "wheat" or "other") to be recognized. These clusters are then combined into broader signatures by a program that considers each pair of signatures within a category, combines the best pair in the light of certain criteria, saves the combined signature and repeats the procedure until there is one signature for each category. The result is a collection of sets of signatures, one set for …

Computer Location Of Citrus Trees Using Color Aerial Infrared Transparencies, D. H. Williams, J. K. Aggarwal

LARS Symposia

An algorithm is described designed to determine automatically citrus tree locations from color aerial infrared transparencies. It permits further processing to be performed to check trees for three major types of citrus infestations: Brown Soft Scale, Gummosis, and Citrus Mealybug, which cause millions of dollars of damage annually in the Rio Grande Valley of Texas. It would be advantageous to detect these infestations in the early stages,

Forest Type Mapping Using Computer Classification Of Landsat Data, Emily Bryant, Arthur G. Dodge

LARS Symposia

Computer classification of LANDSAT data from July 24, 1973 has resulted in measurements and maps of forest types for two New Hampshire counties. Signatures were developed from training sites and applied to test areas and then to the counties. Acreages of hardwood and softwood type and total forested area derived through this process compare favorably with Forest Service statistics for the same areas. Computer generated maps located hardwood, softwood and mixed wood types accurately when compared with low level aerial photography. Additional measurements made using data from August 29, 1973 and September 21, 1972 for one of the counties are …

Signature Extension Through The Application Of Cluster Matching Algorithms To Determine Appropriate Signature Transformations, Peter F. Lambeck, Daniel P. Rice

LARS Symposia

Signature extension is a process intended to increase the spatial-temporal range over which a set of training statistics can be used to classify data without significant loss of recognition accuracy. The goal of signature extension is to minimize the requirements for collecting ground truth and extracting training statistics, thus reducing the costs and time delays associated with those procedures. Signature extension would then help to provide timely and cost-effective classification over extensive land areas, including remote areas for which ground truth information may not be readily available.

Many current signature extension techniques are based on a transformation of training statistics …

Applicability Of Landsat Data To Water Management And Control Needs, J. W. Jarman

LARS Symposia

When the Landsat satellite was launched in June 1972, the Corps of Engineers undertook several experiments related to the use in water resources development programs of automatic classification of the digital data gathered by the satellite. In one application, the U.S. Army Engineer Waterways Experiment Station in Vicksburg, Mississippi, developed techniques for analyzing the multispectral data to determine surface concentrations of pollutants in the Chesapeake Bay and several tributary rivers. Our San Francisco District, with contract support from North American Rockwell, used automatic classification of multi-spectral data to delineate flow patterns and sediment movement in the San Francisco Bay and …

A Computerized Mapping System For Forest Resource Management Planning, D. W. Smith, S. A. Nottingham, C. W. Wade

LARS Symposia

Large volumes of inventory data are collected and analyzed with the idea of developing resource management schemes for the future. Unless this inventory information is easily accessible, is of the type needed to make management decisions in accordance with current policy, and has a readily available updating system, the entire management plan often becomes a seldom, if ever, used document.

This study deals with the application of several inventory collection and display techniques to assist in making rapid and accurate resource management decisions on a continuing basis. The objective of the study is to develop a comprehensive forest resource management …

Illinois Crop-Acreage Estimation Experiment, Robert M. Ray, Harold F. Huddleston

LARS Symposia

This paper describes remote-sensing data analysis research conducted collaboratively during the last year by personnel of the Center for Advanced Computation (CAC) of the University of Illinois at Urbana-Champaign and the Statistical Reporting Service (SRS) of the U.S. Department of Agriculture. The research reported has been undertaken by CAC and SRS to assess the practicalities of existing high-volume earth observations data acquisition, processing, and communication technologies such as LANDSAT, the ILLIAC IV parallel computer, and the ARPA Network as mechanisms for improving the accuracy of USDA annual estimates of agricultural crop acreages for geographic regions corresponding to U.S. states.

Remote Sensing Technology--A Look To The Future, David Landgrebe

LARS Technical Reports

If one is bothered by the possibility of being wrong perhaps one of the most risky things to do is to attempt to predict future directions of technology. However, in these days of extremely limited resources, good planning for development is essential to get the greatest bang for the buck spent on technology development. Such a plan must be based upon some anticipation of the direction that development will go. Thus, one must attempt to project both the potential for and direction of the future development of a technology. It is important to know at any given time what one's …

Kandidats Image Processing System, Robert M. Haralick, Gary J. Minden, Dale R. Johnson, Amrendra Singh, William F. Bryant, Craig A. Paul

LARS Symposia

KANDIDATS is a comprehensive digital image processing system that interacts with the user at a command string level. It includes prompting for parameter input and checks user input for errors. Image analyses available in KANDIDATS consist of utility functions, image transform operations, spatial clustering, and Bayesian classification. The versatility and capabilities of KANDIDATS arise from a modular programming structure and file structure. These attributes allow processing of images that are a few thousand rows by a few thousand columns in a minicomputer system with only 32 K words of main memory.

Tree System Approach For Landsat Data Interpretation, R. Y. Li, K. S. Fu

LARS Symposia

This paper describes a tree system approach which interpretates highways and rivers from LANDSAT pictures. The basic definitions of tree grammars and tree automaton and a grammatical inference procedure are first introduced. The interpretation process is conceived as a process of continuous verification of the hypothesized descriptions of objects in the picture. The LANDSAT imagery map of Lafayette, Indiana is used as a training data set and tree grammar is inferred from the interpretation process. The versatility of this set of syntactic rules is tested on a different data set and the initial results are reported.

A Non-Parametric Approach To Classifying Remotely Sensed Data, Jack Tubbs, John Engvall

LARS Symposia

A non-parametric classification procedure which categorizes a given pixel as wheat or non-wheat according to the trend it follows throughout the growing season is discussed. The cycle wheat makes during the growing season can be described by observing that wheat goes from "brown" (bare soil) to "green" (greening stage) to "brown" (Harvested or golden stage). Since wheat is the only significant U.S. crop which has this cycle during the growing season for winter wheat, we felt that if we could develop a procedure which would enable one to visually follow this type of trend, then perhaps one would be able …

Processing Multispectral Scanner Data Using Correlation Clustering And Nonparametric Classification Techniques, Richard E. Haskell, David E. Boddy

LARS Symposia

A two-step classification algorithm for processing multispectral scanner data has been developed and tested. The algorithm is carried out by two separate programs called CLUSTX and GROUPX. The program CLUSTX is a single pass clustering algorithm that assigns each pixel, based on its spectral signature, to one of NCLUST clusters. The output of the program CLUSTX is a cluster tape in which a single integer is associated with each pixel, This integer is the cluster number to which the pixel has been assigned by the program.

Selecting Class Weights To Minimize Classification Bias In Acreage Estimation, W. M. Belcher, T. C. Minter

LARS Symposia

Classification of multispectral data by the use of a maximum likelihood classifier is dependent upon knowing in advance a set of prior probabilities. Therefore, the selection of an optimal set of prior probabilities is critical to the estimation of proportions for each class. In the proposed procedure, a function is minimized to yield a set of optimal prior probabilities for a specific data set. Classification results using optimal, actual, and default (equal prior probabilities for each class) values are compared.

Estimation Of The Probability Of Error Without Ground Truth And Known A Priori Probabilities, K. A. Havens, T. C. Minter, S. G. Thadani

LARS Symposia

The probability of error or, alternatively, the probability of correct classification (PCC) is an important criterion in analyzing the performance of a classifier. Labeled samples (those with ground truth) are usually employed to evaluate the performance of a classifier. Occasionally, the numbers of labeled samples are inadequate, or no labeled samples are available to evaluate a classifier's performance; for example, when crop signatures from one area from which ground truth is available are used to classify another area from which no ground truth is available. This paper reports the results of an experiment to estimate the probability of error using …

A New Computer Approach To Mixed Feature Classification For Forestry Application, E. P. Kan, J. K. Lo

LARS Symposia

A new computer approach for mapping mixed forest features (i.e., classes, types) from computes classification maps is presented in both theory and application.

This approach is particularly useful and applicable to forestry stand mapping, where small areas are required to be absorbed into the surrounding to form "homogeneous" stands, and where mixed stands contain mixed proportions of different species of trees. Previous studies involving LANDSAT data show that mixed pine-hardwood stands are often erroneously classified as either pine or hardwood.

The present work utilizes a modification and an iterative application of a previously developed computer program called "CLEAN". The program …