Publications

2D convolution is a staple of digital image processing. The advent of large format imagers makes it possible to literally ''pave'' with silicon the focal plane of an optical sensor, which results in very large images that can require a significant amount computation to process. Filtering of large images via 2D convolutions is often complicated by a variety of effects (e.g., non-uniformities found in wide field of view instruments) which must be compensated for in the filtering process by changing the filter across the image. This paper describes a fast (FFT based) method for convolving images with slowly varying filters. A parallel version of the method is implemented using a multi-threaded approach, which allows more efficient load balancing and a simpler software architecture. The method has been implemented within a high level interpreted language (IDL), while also exploiting open standards vector libraries (VSIPL) and open standards parallel directives (OpenMP). The parallel approach and software architecture are generally applicable to a variety of algorithms and has the advantage of enabling users to obtain the convenience of an easy operating environment while also delivering high performance using a fully portable code.

The CSKETCH image processing library is a collection of C++ classes and global functions which comprise a development environment for meteorological algorithms. The library is best thought of as a 'tool-kit' which contains many standard mathematical and signal processing functions often employed in the analysis of weather radar data. A tutorial-style introduction to the library is given, complete with many examples of class and global function usage. Included is an in-depth look at the main class of the library, the SKArray class, which is a templatized and encapsulated class for storing numerical data arrays of one, two, or three dimensions. Following the tutorial is a complete reference for the library which describes all publicly-available class data members and class member functions, as well as all global functions included in the library.

Thunderstorm activity and associated low-altitude wind shear constitute a significant safety hazard to aviation, particularly during operations near airport terminals where aircraft altitude is low and flight routes are constrained. The Federal Aviation Administration (FAA) has procured several dedicated meteorological sensors (Terminal Doppler Weather Radar (TDWR), Network Expansion Low Level Wind Shear Alert System (LLWAS) at major airports to enhance the safety and efficiency of operations during convective weather. A hardware and software modification to existing Airport Surveillance Radars (ASR-9)-the Weather Systems Processor (WSP)-will provide similar capabilities at much lower cost, thus allowing the FAA to extend its protection envelope to medium density airports and airports where thunderstorm activity is less frequent. Following successful operation demonstrations of a prototype ASR-WSP, the FAA has procured approximately 35 WSP's for nationwide deployment. Lincoln Laboratory was responsible for development of all data processing algorithms, which were provided as Government Furnished Equipment (GFE), to be implemented by the full-scale development (FSD) contractor without modification. This report defines the operations that are used to produce images of atmospheric reflectivity, Doppler velocity and data quality that are used by WSP's meteorological product algorithms to generate automated information on hazardous wind shear and other phenomena. Principle requirements are suppression of interference (e.g. ground clutter, moving points targets, meteorological and ground echoes originating from beyond the radar's unambiguous range), generation of meteorologically relevant images and estimates of data quality. Hereafter, these operations will be referred to as "signal processing" and the resulting images as "base data."

Providing an accurate picture of the weather conditions in the pilot's area of interest is a highly useful application for ground-to-air datalinks. The problem with using data links to transmit weather graphics is the large number of bits required to exactly specify the weather image. To make transmission of weather images practical, a means must be found to compress the data to a size compatible with a limited datalink capacity. The Weather-Huffman (WH) Algorithm developed in this report incorporates several subalgorithms in order to encode as faithfully as possible an input weather image within a specified datalink bit limitation. The main algorithm component is the encoding of a version of the input image via the Weather Huffman runlength code, a variant of the standard Huffman code tailored to the peculiarities of weather images. If possible, the input map itself is encoded. Generally, however, a resolution-reduced version of the map must be created prior to the encoding to meet the bit limitation. In that case, the output map will contain blocky regions, and higher weather level areas will tend to bloom in size. Two routines are included in WH to overcome these problems. The first is a Smoother Process, which corrects the blocky edges of weather regions. The second, more powerful routine, is the Extra Bit Algorithm (EBA). EBA utilizes all bits remaining in the message after the Huffman encoding to correct pixels set at too high a weather level. Both size and shape of weather regions are adjusted by this algorithim. Pictorial examples of the operation of this algorithm on several severe weather images derived from NEXRAD are presented.

An important component of any automated image analysis system is the detection and classification of objects. In this report, we consider the first of these problems where the specific goal is to detect anomalous areas (e.g., man-made objects) in textured backgrounds such as trees, grass, and fields of aerial photographs. Our detection algorithm relies on a significance test which adapts itself to the changing background in such a way that a constant false alarm rate is maintained. Furthermore, this test has a potentially practical implementation since it can be expressed in terms of the residuals of an adaptive two-dimensional linear predictor. The algorithm is demonstrated with both synthetic and realworld images.