Side-looking Airborne Radar Images Research Articles

Semantic Segmentation of SLAR Imagery with Convolutional LSTM Selectional AutoEncoders

We present a method to detect maritime oil spills from Side-Looking Airborne Radar (SLAR) sensors mounted on aircraft in order to enable a quick response of emergency services when an oil spill occurs. The proposed approach introduces a new type of neural architecture named Convolutional Long Short Term Memory Selectional AutoEncoders (CMSAE) which allows the simultaneous segmentation of multiple classes such as coast, oil spill and ships. Unlike previous works using full SLAR images, in this work only a few scanlines from the beam-scanning of radar are needed to perform the detection. The main objective is to develop a method that performs accurate segmentation using only the current and previous sensor information, in order to return a real-time response during the flight. The proposed architecture uses a series of CMSAE networks to process in parallel each of the objectives defined as different classes. The output of these networks are given to a machine learning classifier to perform the final detection. Results show that the proposed approach can reliably detect oil spills and other maritime objects in SLAR sequences, outperforming the accuracy of previous state-of-the-art methods and with a response time of only 0.76 s.

Two-Stage Convolutional Neural Network for Ship and Spill Detection Using SLAR Images

This paper presents a system for the detection of ships and oil spills using side-looking airborne radar (SLAR) images. The proposed method employs a two-stage architecture composed of three pairs of convolutional neural networks (CNNs). Each pair of networks is trained to recognize a single class (ship, oil spill, and coast) by following two steps: a first network performs a coarse detection, and then, a second specialized CNN obtains the precise localization of the pixels belonging to each class. After classification, a postprocessing stage is performed by applying a morphological opening filter in order to eliminate small look-alikes, and removing those oil spills and ships that are surrounded by a minimum amount of coast. Data augmentation is performed to increase the number of samples, owing to the difficulty involved in obtaining a sufficient number of correctly labeled SLAR images. The proposed method is evaluated and compared to a single multiclass CNN architecture and to previous state-of-the-art methods using accuracy, precision, recall, F-measure, and intersection over union. The results show that the proposed method is efficient and competitive, and outperforms the approaches previously used for this task.

Oil Spill Detection in Terma-Side-Looking Airborne Radar Images Using Image Features and Region Segmentation.

This work presents a method for oil-spill detection on Spanish coasts using aerial Side-Looking Airborne Radar (SLAR) images, which are captured using a Terma sensor. The proposed method uses grayscale image processing techniques to identify the dark spots that represent oil slicks on the sea. The approach is based on two steps. First, the noise regions caused by aircraft movements are detected and labeled in order to avoid the detection of false-positives. Second, a segmentation process guided by a map saliency technique is used to detect image regions that represent oil slicks. The results show that the proposed method is an improvement on the previous approaches for this task when employing SLAR images.

Sensoriamento remoto da vegetação: evolução e estado-da-arte

Systematic surveys of the vegetative cover in Brazil include the semi-detailed mappings conducted within the scope of the Radam and Radambrasil projects, between 1971 and 1977, based on side-looking airborne radar images, and more recently, the assessments, based on Landsat 7 ETM+ imagery, of the remnant vegetation cover of the Brazilian biomes (MMA/PROBIO), finished in 2006. In a little more than 30 years, many changes have occurred, both in vegetation cover as well as in orbital remote sensing. In this review article, we discuss, from a conceptual and technological perspective, the evolution and the state of the art of imager sensors, as well as some of the enhancement techniques utilized for decoding (e.g., in biophysical parameters, etc.) and classifying the spectral response of vegetation. Likewise, we present the main current (pioneering) initiatives for land cover monitoring of the Brazilian biomes.

Analysis of linear features mapped in Landsat Thematic Mapper and side-looking airborne radar images of the Reno 1° by 2° quadrangle, Nevada and California: implications for mineral resource studies : L. C. Rowan & T. L. Bowers, Photogrammetric Engineering & Remote Sensing, 61(6), 1995, pp 749–759

Analysis of linear features mapped in Landsat Thematic Mapper and side-looking airborne radar images of the Reno 1° by 2° quadrangle, Nevada and California: implications for mineral resource studies : L. C. Rowan & T. L. Bowers, Photogrammetric Engineering & Remote Sensing, 61(6), 1995, pp 749–759

The USGS Side-Looking Airborne Radar (SLAR) Program: CD-ROMs Expand Potential for Petroleum Exploration

The United States Geological Survey (USGS) began the systematic collection of Side-Looking Airborne Radar (SLAR) data in 1980. The SLAR image data, useful for many geologic applications including petroleum exploration, are compiled into mosaics using the USGS 1:250,000-scale topographic map series for format and control. Mosaics have been prepared for over 35% of the United States. Image data collected since 1985 are also available as computer compatible tapes (CCTs) for digital analysis. However, the use of tapes is often cumbersome. To make digital data more readily available for use on a microcomputer, the USGS has started to prepare compact discs-read only memory (CD-ROM). Several experimental discs have been compiled to demonstrate the utility of the medium to make available very large data sets. These discs include necessary nonproprietary software text, radar, and other image data. The SLAR images selected for these discs show significantly different geologic features and include the Long Valley caldera, a section of the San Andreas fault in the Monterey area, the Grand Canyon, and glaciers in southeastern Alaska. At present, several CD-ROMs are available as standard products distributed by the USGS EROS Data Center in Sioux Falls, South Dakota 57198. This is also the source formore » all USGS SLAR photographic and digital material.« less

Large quaternary landslides in the central appalachian valley and ridge province near Petersburg, West Virginia

Geological mapping and photointerpretation of side-looking airborne radar images and color-infrared aerial photographs reveal two large Quaternary landslides in the Valley and Ridge province of the central Appalachians near Petersburg, W. Va. The Elkhorn Mountain rock avalanche occurs on the thrust-faulted northwestern flank of the Elkhorn Mountain anticlinorium. A minimum of 7 × 10 6 m 3 of quartzite colluvium was transported more than 3 km from a 91 m high escarpment of Silurian Tuscarora Quartzite. The extensively vegetated deposit may owe, in part, its transport and weathering to periglacial conditions during the Pleistocene. In contrast, the Gap Mountain rock block slide is a single allochthonous block that is 1.2 km long, 0.6 km wide, and at least 60 m thick. The 43 × 10 6 m 3 block is composed of limestone of the Helderberg Group and the Oriskany Sanstone of Early Devonian age. Planar detachment probably occurred along a dissolution bedding plane near the Shriver Chert and the Oriskany Sandstone contact. Failure probably was initiated by downcutting of the South Branch Potomac River during the Pleistocene. Landslides of this magnitude suggest accelerated erosion during periglacial climates in the Pleistocene. The recognition of these large slope failures may provide evidence of paleoclimatic conditions and, thereby, increase our understanding of the geomorphologic development of the Valley and Ridge province.

Central Appalachian Cross-Strike Structural Discontinuities and Lineaments Compiled on Side-Looking Airborne Radar Image Mosaics: ABSTRACT

Seven cross-strike structural discontinuities (CSDs), 17 lineaments, and 4 crustal blocks, previously recognized in the central Appalachians, were compiled on side-looking airborne radar (SLAR) image mosaics to develop a structural model. The high-resolution, synoptic view and detailed expression of surficial morphology on x-band SLAR images provide a preliminary means of mapping CSDs and lineaments on the basis of alignment or disruption of structural and geomorphic patterns. Wheeler defined CSDs as structural lineaments or alignments at high angles to regional strike that are recognizable because they disrupt strike-parallel structural, geophysical, geomorphic, sedimentologic, or other patterns in allochthonous fold and thrust belts. CSDs are broad zones that may contai many lineaments of varying size and orientation. Geologic and geophysical data suggest that some Appalachian CSDs and lineaments are the surficial expression of crustal block boundaries. The previously defined CSDs, lineaments, and crustal blocks were compiled on a 1:1,000,000-scale SLAR image mosaic of eight 1:250,000-scale quadrangles from central Pennsylvania to southern Virginia. Data compiled on the likely origin of these 7 CSDs suggest that three involve basement, three involve only lateral decollement ramps (zones where a decollement transfers to a higher stratigraphic level along regional strike), and one may involve both basement and a lateral decollement ramp. CSD lateral ramps can dip either north or south. Geologic literature suggests that lateral decollement ramps alternate in dip direction along strike of the CSD. Reported evidence for basement involvement in the origin of CSDs includes spatially related centers of earthquake intensity, intrusive igneous rocks, stratigraphic evidence of syndepositional uplift, and basement faults beneath splay faults as seen in seismic reflection data. High-resolution strike-line seismic reflection, gravity, and magnetic profiles are necessary to further understand the origin and formation mechanisms of CSDs. CSDs are an exploration target for natural gas because they are areas where fracture permeability is enhanced. The spatial relationship of CSDs to gas fields in the Valley and Ridge province suggests structural closure of anticlinal traps due to differential movement along the CSD zone. End_of_Article - Last_Page 1448------------

Imaging radar observations of directional properties of ocean waves

SEASAT‐A synthetic aperture radar (SAR) and side‐looking airborne radar (SLAR) images of ocean waves are examined in the form of normalized directional distributions of backscatter variance at series of frequencies. This method provides a more detailed description of radar results than have contoured two‐dimensional wave number spectra and reduces some of the uncertainties in relating radar measurements to the waves. The range of aspects of the radar distribution that parallel those of ocean waves is defined. Within this restriction, not only can dominant wave frequencies and directions be determined accurately, but also the shape of a directional peak at a frequency, its directional width, and the background level can be determined approximately. Some of these aspects are examined with SLAR images obtained near reference wave measurements. Through its superior directional resolution, the radar appears to have distinguished two wave trains at a single frequency only 20° different in direction. The SEASAT‐A satellite SAR provided an unusual opportunity to examine directional properties of waves in the hostile environment about Hurricane Fico. A swell highly dispersed in frequency and direction at a distance from the center of 450 km had a minimum observed directional width of 11°. Wave directions, their changes with frequency, and directional widths were in accord with those expected from the hurricane winds. Thematic maps of the direction and width of the swell energy as it spread across the ocean surface show smooth changes in these properties over distance, with relatively small scatter of individual values. These patterns also are in accord with those from a simple hurricane wave emission concept, but details of the distributions show distinct departures that must represent unrecognized smaller‐scale fluctuations of the process.

Seasat Radar Image of San Andreas Fault, California

On a Seasat radar image (23.5-cm wavelength) of the Durmid Hills in southern California, the San Andreas fault is expressed as a prominent southeast-trending tonal lineament that is bright on the southwest side and dark on the northeast side. Field investigation established that the bright signature corresponds to outcrops of the Borrego Formation, which weathers to a rough surface. The dark signature corresponds to the Lake Cahuilla sand and silt deposits which are smooth at the wavelength of the Seasat radar. These signatures and field characteristics agree with calculations of the smooth and rough radar criteria. On Landsat and Skylab images of the Durmid Hills, the Borrego and Lake Cahuilla surfaces have similar bright tones and the San Andreas fault is not detectable On a side-looking airborne radar image (0.86-cm wavelength), both the Borrego and Lake Cahuilla surfaces appear rough, which results in bright signatures on both sides of the San Andreas fault. Because of this lack of roughness contrast, the fault cannot be distinguished on the aircraft image acquired at a short radar wavelength. The wavelength of the Seasat radar system is well suited for mapping Durmid Hills geologic features that are obscure on the other remote sensing images evaluated in this report.

Characteristics of Remotely Sensed Data Acquired from Aircraft: ABSTRACT

Remote sensing technology has advanced rapidly during the past 2 decades, to the point where sensor systems are capable of acquiring data in wavelength bands from the ultraviolet through the visible and infrared to the microwave. Imaging sensor systems used in aircraft can be grouped into three general catagories: camera, optical-mechanical scanner, and sidelooking radar. The aerial camera records radiation reflected from features on the earth's surface on black-and-white, normal color, and false-color infrared films. Black-and-white films are used frequently for mapping purposes; however, usage of color films is increasing. Normal color films have advantages in ease of interpretation because surface features are displayed in colors similar to those observed by the human eye. However, atmospheric scattering of incident solar radiation can have a significant deleterious effect upon the processed color photograph. False-color infrared films record reflected radiation in the near-infrared, red and green portions of the electromagnetic spectrum. The false-color photograph displays healthy vegetation as red, water as dark blue to black and red rocks or soil as a hue of yellow. Advantages of false-color infrared photography include less sensitivity to atmospheric scatter; greater contrast between vegetated and nonvegetated areas, moist and dry soils, clear and turbid water, and fresh and old snow; and, in some situations, an enhanced depiction of variations in vegetation owing to stress or species differences. Optical-mechanical scanners employ photo-electric sensors to record radiation reflected from the earth's surface. The instantaneous-field-of-view (IFOV), normally expressed in milli-stereradians, and the altitude of the aircraft determine the ground resolution (size of the area on the ground that is sampled at any instant of time). Multispectral scanners record reflected and emitted energy in a number of wavelength intervals within the ultraviolet, visible (0.4 to 0.7µm) and infrared (0.7 to 14µm) portions of the electromagnetic spectrum. The infrared (IR) region generally is subdivided into two parts: the reflected IR (0.7 to 3.0µm) and the thermal IR (3.0 to 14.0µm). Thermal IR scanners record emitted radiation in two atmospheric windows (generally 3 to 5µm nd 8 to 14µm). The apparent temperature of earth materials is a function of their emissivity and their absolute temperature. Conventional thermal infrared images display variations in apparent surface radiant temperatures in shades of gray with light tones representing warm temperatures and dark tones representing cool temperatures. Side-looking airborne radar (SLAR) is an active imaging system because if transmits electromagnetic energy that illuminates the terrain. Radar systems operate in the 0.8 cm to 1 m wavelength range of the spectrum. They can be used during day or night and under most weather conditions. Direction and angle of illumination can be controlled to enhance topographic expression of geologic features. The interaction between terrain materials and the transmitted pulse of electromagnetic energy is complex and depends on orientation, surface roughness, and electrical properties of terrain features. Conventional SLAR images display the intensity of the energy returned to the radar received in tones of gray: light tones represent high intensity returns (such as buildings and slopes facing the ante na) and dark tones indicate low intensity returns (such as smooth water, dry lake beds, and roads). Radar has several advantages for regional analysis of geologic structure and terrain analysis: control of illumination direction, day and night and all-weather capability, suppression of surface detail, and continuous image strips. End_of_Article - Last_Page 506------------

Comparison of SLAR Images and Small-Scale, Low-Sun Aerial Photographs

A comparison of side-looking airborne radar (SLAR) images and black and white aerial photos of similar scale and illumination of an area in the Mojave Desert of California shows that aerial photos yield far more information about geology than do SLAR images because of greater resolution, tonal range, and geometric fidelity, and easier use in stereo. Nevertheless, radar can differentiate some materials or surfaces that aerial photos cannot; thus, they should be considered as complementary, rather than competing tools in geologic investigations. The most significant advantage of SLAR, however, is its freedom from the stringent conditions of weather, date, and time that are required by small-scale aerial photos taken with a specified direction and angle of illumination. Indeed, in low latitudes, SLAR is the only way to obtain small-scale images with low illumination from certain directions; moreover, in areas of nearly continuous cloudiness, radar may be the only practical source of small-scale images.