Feature extraction from imagery is an important and long-standing problem in remote sensing. In this paper, we report on work using genetic programming to perform feature extraction simultaneously from multispectral and digital elevation model (DEM) data. We use the GENetic Imagery Exploitation (GENIE) software for this purpose, which produces image-processing software that inherently combines spatial and spectral processing. GENIE is particularly useful in exploratory studies of imagery, such as one often does in combining data from multiple sources. The user trains the software by painting the feature of interest with a simple graphical user interface. GENIE then uses genetic programming techniques to produce an image-processing pipeline. Here, we demonstrate evolution of image processing algorithms that extract a range of land cover features including towns, wildfire burnscars, and forest. We use imagery from the DOE/NNSA Multispectral Thermal Imager (MTI) spacecraft, fused with USGS 1:24000 scale DEM data.
Los Alamos National Laboratory has developed and demonstrated a highly capable system, GENIE, for the two-class problem of detecting a single feature against a background of non-feature. In addition to the two-class case, however, a commonly encountered remote sensing task is the segmentation of multispectral image data into a larger number of distinct feature classes or land cover types. To this end we have extended our existing system to allow the simultaneous classification of multiple features/classes from multispectral data. The technique builds on previous work and its core continues to utilize a hybrid evolutionary-algorithm-based system capable of searching for image processing pipelines optimized for specific image feature extraction tasks. We describe the improvements made to the GENIE software to allow multiple-feature classification and describe the application of this system to the automatic simultaneous classification of multiple features from MTI image data. We show the application of the multiple-feature classification technique to the problem of classifying lava flows on Mauna Loa volcano, Hawaii, using MTI image data and compare the classification results with standard supervised multiple-feature classification techniques.
The Cerro Grande/Los Alamos forest fire devastated over 43,000 acres (17,500 ha) of forested land, and destroyed over 200 structures in the town of Los Alamos and the adjoining Los Alamos National Laboratory. The need to measure the continuing impact of the fire on the local environment has led to the application of a number of remote sensing technologies. During and after the fire, remote-sensing data was acquired from a variety of aircraft- and satellite-based sensors, including Landsat 7 Enhanced Thematic Mapper (ETM+). We now report on the application of a machine learning technique to the automated classification of land cover using multi-spectral and multi-temporal imagery. We apply a hybrid genetic programming/supervised classification technique to evolve automatic feature extraction algorithms. We use a software package we have developed at Los Alamos National Laboratory, called GENIE, to carry out this evolution. We use multispectral imagery from the Landsat 7 ETM+ instrument from before, during, and after the wildfire. Using an existing land cover classification based on a 1992 Landsat 5 TM scene for our training data, we evolve algorithms that distinguish a range of land cover categories, and an algorithm to mask out clouds and cloud shadows. We report preliminary results of combining individual classification results using a K-means clustering approach. The details of our evolved classification are compared to the manually produced land-cover classification.
Between May 6 and May 18, 2000, the Cerro Grande/Los Alamos wildfire burned approximately 43,000 acres (17,500 ha) and 235 residences in the town of Los Alamos, NM. Initial estimates of forest damage included 17,000 acres (6,900 ha) of 70-100% tree mortality. Restoration efforts following the fire were complicated by the large scale of the fire, and by the presence of extensive natural and man-made hazards. These conditions forced a reliance on remote sensing techniques for mapping and classifying the burn region. During and after the fire, remote-sensing data was acquired from a variety of aircraft-based and satellite-based sensors, including Landsat 7. We now report on the application of a machine learning technique, implemented in a software package called GENIE, to the classification of forest fire burn severity using Landsat 7 ETM+ multispectral imagery. The details of this automatic classification are compared to the manually produced burn classification, which was derived from field observations and manual interpretation of high-resolution aerial color/infrared photography.
We describe the implementation and performance of a parallel, hybrid evolutionary-algorithm-based system, which optimizes image processing tools for feature-finding tasks in multi-spectral imagery (MSI) data sets. Our system uses an integrated spatio-spectral approach and is capable of combining suitably-registered data from different sensors. We investigate the speed-up obtained by parallelization of the evolutionary process via multiple processors (a workstation cluster) and develop a model for prediction of run-times for different numbers of processors. We demonstrate our system on Landsat Thematic Mapper MSI , covering the recent Cerro Grande fire at Los Alamos, NM, USA.
KEYWORDS: Digital signal processing, Reconfigurable computing, Sensors, Image segmentation, Image processing, Remote sensing, Field programmable gate arrays, Feature extraction, Signal processing, Algorithm development
Compute performance and algorithm design are key problems of image processing and scientific computing in general. For example, imaging spectrometers are capable of producing data in hundreds of spectral bands with millions of pixels. These data sets show great promise for remote sensing applications, but require new and computationally intensive processing. The goal of the Deployable Adaptive Processing Systems (DAPS) project at Los Alamos National Laboratory is to develop advanced processing hardware and algorithms for high-bandwidth sensor applications. The project has produced electronics for processing multi- and hyper-spectral sensor data, as well as LIDAR data, while employing processing elements using a variety of technologies. The project team is currently working on reconfigurable computing technology and advanced feature extraction techniques, with an emphasis on their application to image and RF signal processing. This paper presents reconfigurable computing technology and advanced feature extraction algorithm work and their application to multi- and hyperspectral image processing. Related projects on genetic algorithms as applied to image processing will be introduced, as will the collaboration between the DAPS project and the DARPA Adaptive Computing Systems program. Further details are presented in other talks during this conference and in other conferences taking place during this symposium.
We consider the problem of pixel-by-pixel classification of a multi- spectral image using supervised learning. Conventional spuervised classification techniques such as maximum likelihood classification and less conventional ones s uch as neural networks, typically base such classifications solely on the spectral components of each pixel. It is easy to see why: the color of a pixel provides a nice, bounded, fixed dimensional space in which these classifiers work well. It is often the case however, that spectral information alone is not sufficient to correctly classify a pixel. Maybe spatial neighborhood information is required as well. Or maybe the raw spectral components do not themselves make for easy classification, but some arithmetic combination of them would. In either of these cases we have the problem of selecting suitable spatial, spectral or spatio-spectral features that allow the classifier to do its job well. The number of all possible such features is extremely large. How can we select a suitable subset? We have developed GENIE, a hybrid learning system that combines a genetic algorithm that searches a space of image processing operations for a set that can produce suitable feature planes, and a more conventional classifier which uses those feature planes to output a final classification. In this paper we show that the use of a hybrid GA provides significant advantages over using either a GA alone or more conventional classification methods alone. We present results using high-resolution IKONOS data, looking for regions of burned forest and for roads.
We describe the implementation and performance of a genetic algorithm (GA) which evolves and combines image processing tools for multispectral imagery (MSI) datasets. Existing algorithms for particular features can also be “re-tuned” and combined with the newly evolved image processing tools to rapidly produce customized feature extraction tools. First results from our software system were presented previously. We now report on work extending our system to look for a range of broad-area features in MSI datasets. These features demand an integrated spatio- spectral approach, which our system is designed to use. We describe our chromosomal representation of candidate image processing algorithms, and discuss our set of image operators. Our application has been geospatial feature extraction using publicly available MSI and hyperspectral imagery (HSI). We demonstrate our system on NASA/Jet Propulsion Laboratory’s Airborne Visible and Infrared Imaging Spectrometer (AVIRIS) HSI which has been processed to simulate MSI data from the Department of Energy’s Multispectral Thermal Imager (MTI) instrument. We exhibit some of our evolved algorithms, and discuss their operation and performance.
We describe the implementation and performance of a genetic algorithm which generates image feature extraction algorithms for remote sensing applications. We describe our basis set of primitive image operators and present our chromosomal representation of a complete algorithm. Our initial application has been geospatial feature extraction using publicly available multi-spectral aerial-photography data sets. We present the preliminary results of our analysis of the efficiency of the classic genetic operations of crossover and mutation for our application, and discuss our choice of evolutionary control parameters. We exhibit some of our evolved algorithms, and discuss possible avenues for future progress.
The retrieval of scene properties (surface temperature, material type, vegetation health, etc.) from remotely sensed data is the ultimate goal of many earth observing satellites. The algorithms that have been developed for these retrievals are informed by physical models of how the raw data were generated. This includes models of radiation as emitted and/or reflected by the scene, propagated through the atmosphere, collected by the optics, detected by the sensor, and digitized by the electronics. To some extent, the retrieval is the inverse of this 'forward' modeling problem. But in contrast to this forward modeling, the practical task of making inferences about the original scene usually requires some ad hoc assumptions, good physical intuition, and a healthy dose of trial and error. The standard MTI data processing pipeline will employ algorithms developed with this traditional approach. But we will discuss some preliminary research on the use of a genetic programming scheme to 'evolve' retrieval algorithms. Such a scheme cannot compete with the physical intuition of a remote sensing scientist, but it may be able to automate some of the trial and error. In this scenario, a training set is used, which consists of multispectral image data and the associated 'ground truth;' that is, a registered map of the desired retrieval quantity. The genetic programming scheme attempts to combine a core set of image processing primitives to produce an IDL (Interactive Data Language) program which estimates this retrieval quantity from the raw data.
The ALEXIS mission, serving as the first dedicated all-sky monitor in the extreme UV (EUV), has been collecting data since its launch in 1993. ALEXIS operates in a 70 degree inclination orbit at an altitude of 800 km. The ALEXIS science mission is to observe the cosmic UV background and to study variability of EUV sources. The ALEXIS experiment is composed of six telescopes. Although ALEXIS was designed for a one-year technology verification mission, the telescopes are still functioning with much the same effectiveness as at the beginning of the mission. The telescopes comprise: 1) layered synthetic microstructure (LSM) spherical mirrors, 2) thin foil filters, and 3) microchannel plate detectors, all enshrouded within the telescope body. The LSM mirrors select the bandpass for each telescope, while providing enhanced rejection of the HeII 304 angstrom geocoronal radiation. The filters, constructed either form aluminum/carbon or Lexxan/titanium/boron, serve to strongly erect the geocoronal radiation, as well as longer wavelength emission from bright O or B stars. Each telescope detector consists of two plates, the outermost of which is curved to accurately match the spherical focal surface of the mirror. By reviewing the ground and flight histories, this paper analyzes the flight performance of the filters, including the effects of long term exposure and the formation of pinholes.
The array of low energy x-ray imaging sensors (ALEXIS) satellite was launched from the 4th flight of the Pegasus booster on 25 April 1993 into an 800 km, 70 degree inclination orbit. After an initial launch difficulty, the satellite was successfully recovered and is still producing 100 MB of mission data per day. ALEXIS, still going strong in its sixth year, was originally designed to be a high risk, single string, Smaller-Faster-Cheaper satellite, with a 1-year nominal and a 3-year design limit. This paper will discuss the on-orbit detector performance including microchannel plate operation, pre- and post-flight calibration efforts, observed backgrounds and impacts of flying in a high radiation environment.
The Array of Low Energy X-ray Imaging Sensors (ALEXIS) satellite is Los Alamos' first attempt at building and flying a small, low cost, rapid development, technology demonstration and scientific space mission. The ALEXIS satellite contains the two experiments: the ALEXIS telescope array, (which consists of six EUV/ultrasoft x-ray telescopes utilizing multilayer mirrors, each with a 33 degree field-of-view), and VHF ionospheric experiment called BLACKBEARD. The spacecraft is controlled exclusively from a ground station located at Los Alamos. The 113-kg ALEXIS satellite was launched by a Pegasus booster into a 750 X 850 km, 70 degree inclination orbit on April 25, 1993. Due to damage sustained at the time of launch, ground controllers did not make contact with the satellite until late June. By late July, full satellite operations had been restored through the implementation of new procedures for attitude control. Science operations with the two onboard experiments began at that time. This paper will discuss our experience gained in launching and managing this small scientific and technology demonstration satellite.
The Array of Low Energy X-ray Imaging Sensors (ALEXIS) satellite is Los Alamos' first attempt at building and flying a low cost, rapid development, technology demonstration and scientific space mission. The ALEXIS satellite contains the two experiments: the ALEXIS telescope array, (which consists of six EUV/ultrasoft x- ray telescopes utilizing multilayer mirrors, each with a 33 degree field-of-view), and a VHF ionospheric experiment called Blackbeard. A ground station located at Los Alamos exclusively controls the spacecraft. The 248 pound ALEXIS satellite was launched by a Pegasus booster into a 400 x 450 nautical mile, 70 degree inclination orbit on April 25, 1993. Images from a video system on the rocket indicated that ALEXIS had been severely damaged during launch with one of the 4 solar panels breaking away from its mounting. (It later turned out that the solar paddle was still attached to the spacecraft but only through cable bundles.) Attempts at communicating with the satellite were unsuccessful until a surprised ground crew received a short transmission on June 2. By mid July, ground station operators had regained full control of the satellite and began to initiate scientific operations with both the telescope array and the VHF experiment. In this paper we will discuss a preliminary analysis of the on-orbit performance of EUV telescopes on ALEXIS.
We report the launch and rescue of the ALEXIS small satellite. ALEXIS is a 113-kg satellite that carries an ultrasoft x-ray telescope array and a high-speed VHF receiver/digitizer (BLACKBEARD), supported by a miniature spacecraft bus. It was launched by a Pegasus booster on 1993 April 25, but a solar paddle was damaged during powered flight. Initial attempts to contact ALEXIS were unsuccessful. The satellite finally responded in June, and was soon brought under control. Because the magnetometer had failed, the rescue required the development of new attitude control techniques. The telemetry system has performed nominally. The BLACKBEARD experiment was turned on shortly after contact, and has returned its first data. We discuss preliminary lessons learned from ALEXIS.
The Array of Low Energy X-ray Imaging Sensors (ALEXIS) experiment consists of a mini-satellite containing six wide angle EUV/ultrasoft X-ray telescopes. Its purpose is to mp out the sky in three narrow (5%) baridpasses around 66, 71, arid 93 eV. The mission will be launched on the Pegasus Air Launched Vehicle in 1992 into a 400 nautical mile, high inclination orbit. The project is a collaborative effort between Los Alamos National Laboratory, Sandia National Laboratory, and the University of California-Berkeley Space Sciences Laboratory. The six telescopes are arranged in three pairs in such a manner that as the satellite spins twice a minute they scan the entire antisolar hemisphere. Each f/i telescope consists of a spherical multilayer coated mirror with a spherical microchannel plate detector located at the prime focus and a thin aluminum or lexan/boron filter in front of the detector. The multilayer coatings determine the bandpasses of the telescopes. Each telescope has a field of view of 33 degrees. Unlike grazing incidence x-ray telescopes, the point spread function is uniform over the entire field of view with a FWHM of O.5degrees determined by spherical aberration. In this paper we present the status of the project as of July i992 as well as summary results from the pre-flight telescope calibration procedures.
The Array of Low Energy X-ray Imaging Sensors (ALEXIS) experiment consists of six wide angle EUV/ultrasoft Xray
telescopes utilizing normal incidence multilayer mirrors, flown on a miniature satellite to map out the sky in three narrow
bandpasses around 66, 7 1, and 95eV.The 66 and 7 1 eV bandpasses are centered on intense Fe emission lines which are
characteristic of million degree plasmas such as the one thought to produce the soft X-ray background. The 95eVbandpass
has a higher throughput and is more sensitive to continuum sources. The mission will be launched into orbit on the Pegasus
Air Launched Vehicle in mid-1991.
We will present the details of the ALEXIS telescope optical design, initial characterizations of the first flight mirrors
and detectors, and the current schemes for characterizing and calibrating the completed telescope assemblies. We will also
discuss the details of a novel "wavetrap" feature incorporated into the multilayer mirror structure to greatly reduce the mirror's
reflectivity at 304A, a major background contamination flux of He II emission from the geocorona.
Several spherically curved microchannel plate (MCP) stack configurations were studied as part of an ongoing astrophysical detector development program, and as part of the development of the ALEXIS satellite payload. MCP pairs with surface radii of curvature as small as 7 cm, and diameters up to 46 mm have been evaluated. The experiments show that the gain (greater than 1.5 x 10 exp 7) and background characteristics (about 0.5 events/sq cm per sec) of highly curved MCP stacks are in general equivalent to the performance achieved with flat MCP stacks of similar configuration. However, gain variations across the curved MCP's due to variations in the channel length to diameter ratio are observed. The overall pulse height distribution of a highly curved surface MCP stack (greater than 50 percent FWHM) is thus broader than its flat counterpart (less than 30 percent). Preconditioning of curved MCP stacks gives comparable results to flat MCP stacks, but it also decreases the overall gain variations. Flat fields of curved MCP stacks have the same general characteristics as flat MCP stacks.
Metal multilayer mirrors have been designed for the ALEXIS satellite, which is to carry six wide-field telescopes to perform an all-sky survey in three narrow EUV/ultrasoft x-ray wavelength bands. Comprised
of alternating layers of molybdenum and silicon, the mirrors are optimized to provide maximum reflectivity at angles from 12.5 to 17.6° off normal incidence and at wavelengths of 133, 171, or 186 A. Simultaneously, the mirrors use a "wavetrap" to suppress reflectivity at 304 A, where the extremely strong geocoronal line of He II causes severe background problems. Low reflectivity at 304 A is achieved by superposing two layer pairs that provide destructive interference with an effective 2dspacing of 152 A. Calculations predict the 186 A design will have a peak reflectivity at 186 A of 35% and a 304 A reflectivity less than 10 compared to a peak reflectivity
at 186 A of 40% and 304 A reflectivity of 3 x 10 without the wavetrap. In the laboratory the 304 A reflectivity on a multilayer sample with a wavetrap has been measured to be as low as i0. We present details of the calculations and laboratory measurements of the reflectivity performance obtained with prototype mirrors.