Yossi Bushlin

Semi-empirical model-based approach for IR scene simulation

This paper describes a novel approach of generating radiometric scenes of natural backgrounds that will serve as an input for simulating E-O sensor scene outputs in the thermal band. The methodology is based on segmentation of a measured scene (in any spectral band) into elements that have similar thermal behavior. The thermal radiance value for each thermal element is calculated using a set of four semi empirically determined coefficients that relate the surface temperature to the local meteorological parameters such as solar radiation, longwave sky radiation, air temperature and wind speed. The thermal coefficients are determined using a theoretical model and an experimental data base. The diurnal variations of the scene are thus easily predicted by knowing the meteorological parameters and the individual set of thermal coefficients for the various thermal elements of the scene. Since the approach is based on a real scene image and an experimental database the generated images have a realistic appearance including realistic clutter properties. The generated thermal scene will serve as the input to a sensor model that will calculate the expected image of a thermal camera viewing the scene. The paper describes the methodology of the scene generation, the sensor model and demonstrates the approach by giving some examples.

Multispectral radiometers (ColoRad) for spectro-temporal flares intensity emission measurements

As more and more spectral ranges are used by different threat detecting sensors, the effectiveness of a countermeasure is becoming more and more dependent on how similar its emitted spectrum is to the object that it is supposed to simulate. As a result, the need to model and test the countermeasure radiometric output (in radiance units) and contrast (in radiant intensity units) or effective temperature at different wavelengths simultaneously increases in importance during both R&D and production for both the producer of countermeasures (to confuse the seekers) and the producer of missile seekers (to prevent seeker confusion). We have developed a family of multi-spectral radiometers (ColoRad) specifically designed to quantitatively measure countermeasure spectral signatures dynamically for precise characterization. In this paper we describe the design of such instrumentation, including the various modes of operation and highlighting the important instrument features for the present application. In addition an example of measurement is given here to demonstrate its usefulness. The ColoRad performance parameter values are also given in this paper.

Dew, dust, and wind influencing thermal signatures of objects

Dew and dust layers on the surface of an object may significantly affect its thermal state and IR signature. Dew formation begins when the object surface temperature falls below atmospheric dew point temperature. Due to the latent heat released by the water accumulated on the surface the temperature drop stagnates and the object appears warmer then it would be without dew formation. An attempt was made to modify RadThermIR software to account for dew effects. A simple plate model and the more elaborate CUBI thermal modeling benchmarking object were used to study the extent to which dew may change thermal object signatures. A dust layer on an object surface may affect its optical properties and may act as additional thermal insulation when it is thick enough. Both effects influence the temperature and IR signature of the object. Parametric calculations by RadThermIR were performed for various dust thicknesses and optical properties. This data was used in an object/background contrast analysis. The obtained dust/dew layer results will be used in the planning of the next CUBI experiment in natural desert environments. In addition, CUBI data from another geographic location was used for studying different wind models resulting in some interesting conclusions concerning the applicability of the wind model used in RadThermIR.

Variance properties in IR background images

Background images of a typical Mediterranean landscape were measured for 24 hours and their properties were analyzed. The locations were chosen to represent an area with various amount of vegetation. The thermal mean and variance changes around the clock were extracted from these background images and a typical pattern for the variance behavior was identified. A simple model is suggested for describing this behavior. The knowledge of variance dependence on the scene and measurement system parameters could be further used for simulation of IR images or for detection probability estimation.

Polarization signatures of man-made objects in the SW infrared spectral band

Detectability of man made objects in natural environment is strongly affected by the clutter properties of the background. To improve the ability to detect and recognize targets by human observers or by machine vision (ATR algorithm) additional information obtained from polarization properties can be exploited. This paper presents an experimental set up for the measurements of the polarization signature of man made objects based on a commercial FPA IR camera and an externally mounted linear polarizer. Some typical results obtained by this system are presented.

Background properties in arid climates: measurements and analysis

Knowledge of background properties is essential for various applications such as systems engineering and evaluation (e.g. electro-optical sensors or for camouflage design), operational planning and development of ATR algorithms. A series of field tests was conducted in the NEGEV desert in Israel, as a joint effort of the FGAN-FfO (Germany) and EORD (Israel) for characterizing properties of backgrounds in arid climatic regions. Diurnal cycles of background surface temperatures were measured during summer and winter periods in several sites in the NEGEV. The measurement equipment consisted of imaging cameras, most of them calibrated, covering the spectral region from the visible up to the thermal infrared. This paper presents the measurement set- up, the measurement techniques that were used, and some of the first analysis results.

Acquisition and mosaicing of low-contrast background radiometric images for scene simulation

High resolution radiometric images of a large terrain were obtained by scanning the area with a downlooking imaging radiometer attached to a helicopter. A detailed description of the data acquisition process is presented. The output of this process was thousands of small images, which were then combined (mosaiced) to one large radiometric map. For the mosaicing process we use image registration algorithms developed especially for the case where the input images contain no significant features or contrasts, as might be expected from radiometric desert backgrounds at night. In this situation, traditional feature-based or intensity-based registration methods can give unsatisfactory results. We improve these methods using some heuristics. This enables the automatic mosaicing of a few thousand images in a very short time. An example of the use of the resulting radiometric map for the simulation of a thermal sensor searching for targets is described.

Investigation of atmospheric blasts by fast radiometry

Blasts and detonations release large amount of energy in short time duration. Some of this energy is released through radiation in the whole optical spectrum. Measurement of this radiation may serve as a base for investigation of the blast phenomena. A fast multispectral radiometer that operates in proper chosen spectral bands provides extensive information on the physical processes that govern the blast. This information includes the time dependence of the temperature, area of the blast as-well-as of the aerosols and gases that are generated. Analysis of this data indicates the order of the detonation and provides good estimation on the masses and types of the high-explosives (HE) materials and their casing. This paper presents the methodology and instrumentation of fast multispectral radiometry in application to the blast measurement and analysis in a Near-ground Explosion Test (NET). In NET, the flash radiation of the blast was measured for two HE materials: TNT and composition B (CB). The investigation includes charges of different masses (0.25 - 20.0 kg) and of various casing materials (steel, Al, PVC), thickness (2 – 6 mm) and various casing type (open on both face ends and hermetically closed). Analysis of the data demonstrates the power of fast radiometry methodology and reveals the governing characteristics of atmospheric blasts.

Blast investigation by fast multispectral radiometric analysis

Knowledge regarding the processes involved in blasts and detonations is required in various applications, e.g. missile interception, blasts of high-explosive materials, final ballistics and IED identification. Blasts release large amount of energy in short time duration. Some part of this energy is released as intense radiation in the optical spectral bands. This paper proposes to measure the blast radiation by a fast multispectral radiometer. The measurement is made, simultaneously, in appropriately chosen spectral bands. These spectral bands provide extensive information on the physical and chemical processes that govern the blast through the time-dependence of the molecular and aerosol contributions to the detonation products. Multi-spectral blast measurements are performed in the visible, SWIR and MWIR spectral bands. Analysis of the cross-correlation between the measured multi-spectral signals gives the time dependence of the temperature, aerosol and gas composition of the blast. Farther analysis of the development of these quantities in time may indicate on the order of the detonation and amount and type of explosive materials. Examples of analysis of measured explosions are presented to demonstrate the power of the suggested fast multispectral radiometric analysis approach.

Background clutter in the SW infrared spectral band: measurement and analysis

The ability to detect targets, either by human observer or by machine vision algorithms, is strongly affected by the clutter properties of the background. Therefore, the quantitative characterization of the clutter is a must in any target acquisition scenario. This paper will propose a new metric for clutter quantification based on blob analysis algorithm. The algorithm consists of comparison of various calculated properties of the blobs, found in the image, to the properties of the blob representing a target. The algorithm was tested on an experimental database of IR images measured during an air-born field test by a commercially available Focal Plane Array IR camera.