Recent homeland security problems in various countries indicate that fixed surveillance systems at important places are not adequate enough. As the security threats take new dimensions in future, mobile smart security personnel wearing high-tech gear will form the basic infrastructure. See first, listen first, detect first, track first, communicate first with peers, assess the threat and coordinate with security head-quarters are the functions of high-tech gear. This paper proposes a high-tech gear involving (i) hands-free and obtrusion-free textile-based wearable microphone array to
capture users voice and interface with body-worn computer, (ii) microphone arrays embedded in textiles to listen and record others voices from a distance, (iii) miniature cameras embedded in the shirt to provide the user with omni vision (iv) wireless personal display as GUI hidden in textile or natural glasses, (v) GPS and body area network for positional awareness for information in the form of text or textile integrated, (vi) reconfigurable HW/SW for all the above functions configured in the form of a usual belt. The main focus of this paper is how to configure the high-tech gear with
all these sophisticated functions to disappear into the natural wearables of the user giving him normal look in the public.
This project is sponsored by Defence Science & Technology Agency, Ministry of Defence, Singapore. This paper covers multi-discipline technologies at system level, hence not possible to go into details of any subsystem. The main objective of this paper is to share our thoughts and get feedback. Progress and some critical design issues are discussed in this paper.
A new multiscale corner detection method is proposed based on dyadic wavelet transform (WT) of the orientation function of a contour image. As the decomposition of the dyadic WT is complete and its scales are sparse, all the scales are defined as natural scales for corner detection. The points that are wavelet transform modulus maxima (WTMM) at different scales are taken as corner candidates. For each corner candidate, the sum of the corresponding normalized WTMM at all the natural scales is used as significance measure of the "cornerness". The utilization of the complete information makes the performance of the proposed detector independent to the type of input images. The decomposition scales of the WT are restricted by the contour length, which makes the algorithm adaptable for both long contours and short contours. Both subjective and objective evaluation illustrate better performance of the proposed corner detector compared to the conventional methods.
This paper proposes a pose estimation and frontal face detection
algorithm for face recognition. Considering it's application in a
real-world environment, the algorithm has to be robust yet
computationally efficient. The main contribution of this paper is
the efficient face localization, scale and pose estimation using
color models. Simulation results showed very low computational
load when compare to other face detection algorithm. The second
contribution is the introduction of low dimensional statistical
face geometrical model. Compared to other statistical face model
the proposed method models the face geometry efficiently. The
algorithm is demonstrated on a real-time system. The simulation
results indicate that the proposed algorithm is computationally
A Vertical-Strip Least Mean Squared (VSLMS) algorithm is proposed to enhance the detection of small moving targets in IR image sequences. This algorithm is an improvement over the Two-Dimensional LMS (TDLMS), which is designed to detect small targets within highly correlated background of static images. This paper focuses on processing IR image sequences with different background features with layers of sky, sea and land clutter. The VSLMS uses multiple LMS modules and a different scanning method to process individual lines in the IR image sequences. Simulation results show successful enhancement of very small targets in an IR mage sequence.
Robust, real-time, user-friendly, non-restrictive and fully automatic natural like human computer interfaces are required to move away from the present machine-empowered-technologies to future human-empowered-technologies (HET). As one of HET interface technologies, this paper presents a cost-effective stereo face detection and tracking of facial features for determining facial pose. The object features are extracted using max-median filters and a progressive threshold algorithm, the face is verified on 'prominent feature configuration template.' Once face is confirmed, the features are tracked using dynamic programming filter. The results are impressive. Video clips would be shown during presentation in the symposium.
This paper presents a multi-mode fusion algorithm for detection and tracking of dim, point-like target. The key contribution of this paper includes the effective fusion approach to harvest the advantages and complement the disadvantages of various algorithms using conditional voting. From qualitative analysis, these algorithms are separated into two classes, i.e. main and supporting algorithms. In the multi-mode fusion algorithm high confidence is placed on the main algorithms with supporting algorithms used to further reduce the false alarm. The main algorithms trigger a voting process and detection is confirmed true if any of the supporting algorithms report detection. The multi-modal fusion algorithm has lower false alarm and moderate true detection rate compared to any individual algorithm namely, Triple Temporal Filter, Frame Differencing, Continuous Wavelet Transform, Max-median and 2-D Mexican hat filter. Besides, a novel variability filter is proposed to remove strong glint thus reduces false alarm. Kalman filter is used to track the detected targets. A novel track decision algorithm to continue or
terminate the track when target disappears is proposed. Prior knowledge of target in Kalman filter is fed forward to an Adaptive 3-D Matched filter to improve the performance. Three sets of real-world infrared image sequences with very different background and target characteristics were used to test the robustness of the multi-modal fusion algorithm. The algorithm performs satisfactorily in all the image sequences. Video clips will also be presented.
Proc. SPIE. 4728, Signal and Data Processing of Small Targets 2002
KEYWORDS: Target detection, Clouds, Detection and tracking algorithms, Signal processing, Digital filtering, Signal detection, Statistical analysis, 3D acquisition, Electronic filtering, Optical filters
The problem of detecting small target in IR imagery has attracted much research effort over the past few decades. As opposed to early detection algorithms which detect targets spatially in each image and then apply tracking algorithm, more recent approaches have used multiple frames to incorporate temporal as well as spatial information. They often referred to as track before detect algorithms. This approach has shown promising results particularly for detection of dim point-like targets. However, the computationally complexity has prohibited practical usage for such algorithms. This paper presents an adaptive, recursive and computation efficient detection method. This detection algorithm updates parameters and detects occurrence of targets as new frame arrived without storing previous frames, thus achieved recursiveness. Besides, the target temporal intensity change is modeled by two Gaussian distribution with different mean and variance. The derivation of this generalized model has taken account of the wide variation of target speed, therefore detects wider range of targets.
Detection and tracking of low-observable moving targets against heavy clutter in a sequence of infrared images is an important research area. The focus of research in this area is to reliably pick up the most potential targets, track the targets with varying speed and direction, and at the same time reduce the false alarm rate to an acceptable level. However, there is no single method that works equally well in all situations. This paper presents an integrated algorithm based on area-correlation tracker (ACT) and Kalman filter for improving ACT performance for targets with varying speed and direction. Divergence and loss of target when the target is stationary are the two typical problems associated with ACT. In our algorithm, we propose to overcome these shortcomings by introducing an online procedure for updating (or not updating in the case of occlusions) the reference template, in conjunction with linear predictions by using a Kalman filter.
This paper reports on the design considerations and implementations of construction of a receiver system for an infrared communication link using Light-Emitting-Diodes (LEDs). System configuration is non-directed line of sight (LOS) with single element receiver. The speed of the system constructed is 20 MHz (40 Mbps for on-off keying). In our design, the receiver consists of a multi-channel photo-detector array followed by multi-channels transimpedance amplifiers, a summer circuit for combining signals, low-pass filters(LPF) and high pass filters(HPF), and auto-gain control unit(AGC). The spectral match between photodiode and LED, looking angle of photodiode, rise time consideration of the photodiode, trade-off between transit rise time and photodiode effective area, etc, are also be discussed . A receiver module of speed response up to 20 MHz has been successfully constructed and the system is capable of receiving optical signal at a height of 2 m from the ceiling with coverage of a cell of 4 m in diameter.
A Line of Sight, non-directed transceiver system was designed and constructed. The transmitter is composed of two types of LEDs and are tilted at 60 degree(s) to achieve uniform light distribution up to 50 degrees across a plane. The receiver is composed of multiple silicon-pin-photodiodes, transimpedance amplifiers, bandpass filter and AGC, and has a half field of view of 60 degree. Software was written to design uniform distribution on a horizontal plane for non-directed line-of-sight system. Experimental testing was conducted in a room with size 3.4m x 5.1m x 2.6m and measured optical power shows close correspondence to the simulated power distribution. Non-return-to zero signal of certain test codes were transmitted at 34 Mbps, and collected signals indicate that the eye is widely open in the eye diagram.
There are several methods reported in the literature for detecting dim targets against slowly moving clutter. However, each method has its own advantages and disadvantages. The challenge lies in reducing the false alarm rate to an acceptable level. Choosing a threshold for achieving constant false alarm rate is always a tricky problem. Too less a threshold may ensure detection of target pixels. But this will result in with too many false targets, which limit the performance of the post- processor to trace out target paths. Too high threshold results in fewer false alarms but the targets also may miss out, creating a problem in establishing track record of targets. These contradicting issues demand a via-media solution to improve the overall concept of CFAR for the detection of dim point-targets in the presence of the evolving clouds and heavy background clutter. The adaptive threshold is based on random and correlated noises of the incoming image sequence. The incoming frames of data are processed by adaptive threshold and accumulated recursively. The post-processor with built-in flexibility checks for validity of target paths. This paper presents an improvement over our paper presented at SPIE, Denver duing July 1999. The algorithm has been tested with the available database and the results are very promising.
Technological breakthroughs in the field of imaging sensors for missile-seekers and related signal processors helped the military users to achieve `force multiplication'. Fourth generation missile seekers use millimeter-wave and/or infrared-imaging technologies to benefit from the high- resolution capabilities to home on a selected aim-point on a given target using multi-mode signal processing. The drive behind such technologies is to get a first-pass mission- success against the target.
Temporal profiles of point-like dim targets and the extended cloud pixels provide useful information in detecting the target. Among the recent methods that utilize temporal profile of pixel in detecting point target are Triple Temporal Filter (TTF) and Continuous Wavelet Transform (CWT). TTF uses two damped sinusoidal filters, an exponential averaging filter with six appropriate coefficients to deal with different aspects of clutter. TTF is recursive and efficient in detecting point targets without applying any threshold techniques. The performance of CWT is comparable to TTF but all the frames in a sequence need to be stored. Therefore it is computationally complex algorithm.
Temporal profiles of point-like dim targets and the extended cloud pixels provide useful information in detecting the targets. Among the recent methods that utilize temporal profile of pixel in detecting point target are Triple Temporal Filter (TTF) and Continuous Wavelet Transform (CWT). TTF uses two damped sinusoidal filters, an exponential averaging filter with six appropriate coefficients to deal with different aspects of clutter. TTF is recursive and efficient in detecting point targets without applying any threshold techniques. The performance of CWT is comparable to TTF but all the frames in a sequence need to be stored. Therefore it is computationally complex algorithm.
Technological breakthroughs in the field of imaging and non- imaging sensor sand the related signal processors helped the military users to achieve 'force multiplication'. Present day 'smart-weapon systems' are being converted to 'brilliant-weapon systems' to bridge the gap until the most potent new 'fourth generation systems' come on line based on nanotechnology. The recent military tactics have evolved to take advantage of ever improving technologies to improve the quality and performance over time. The drive behind these technologies is to get a first-pass-mission-success against the target with negligible collateral damage, protecting property and the lives of non-combatants. These technologies revolve around getting target information, detection, designation, guidance, aim-point selection, and mission accomplishment. The effectiveness of these technologies is amply demonstrated during recent wars. This paper brings out the emerging trends in visible/IR/radar smart-sensors and the related signal processing technologies that lead to brilliant guided weapon systems. The purpose of this paper is to give an overview to the readers about futuristic systems. This paper also addresses various system configurations including sensor-fusion.
There are several methods reported in the literature for detecting dim targets against slowly moving clutter. However, each method has its won advantages and disadvantages. The challenge lies in reducing the false alarm rate to an acceptable level. 'False alarm rate' defined in case of a significant size of the target in a frame may not be applicable to point-targets. This paper presents a new method for the detection of dim point-targets in the presence of the evolving clouds and heavy background clutter. Choosing a threshold for achieving constant false alarm rate is always a tricky problem. Too less a threshold may ensure detection of target pixels. But this will result in too may false targets, which limit the performance of the post-processor to trace out target paths. Too high a threshold result in fewer false alarms but the targets may also be missing out. Based on off-line studies, it has been found that a 'desirable condition' is required to limit the number of accumulated pixels not to exceed 8 percent of the total image size for post-processing. This paper present a method based on random and correlated noises which in turn selects an auto threshold that leads to the 'desired condition' for the post-processor. An effort has been made to derive an empirical formula based on random and correlated noises to obtain an auto threshold value that achieves the desirable condition. Then the incoming frames of data are then processed by a constant threshold and accumulated as total number of pixels. At the same time the track record of pixels along with frame numbers are recorded. The post-processor to filter out the false alarms uses this information. One advantage of this method is that there is no need to store all the frames to obtain the desired information. The algorithm has been tested with the available database and the results are very promising. It is assumed that most of the targets occupy a couple of pixels. Head-on moving and maneuvering targets are not considered.
This paper deals with the problem of detection and tracking of low observable small-targets from a sequence of IR images against structural background and non-stationary clutter. There are many algorithms reported in the open literature for detection and tracking of targets of significant size in the image plane with good results. However, the difficulties of detecting small-targets arise from the fact that they are not easily discernable from clutter. The focus of research in this area is to reduce the false alarm rate to an acceptable level. Triple Temporal Filter reported by Jerry Silverman et. al., is one of the promising algorithms in this are. In this paper, we investigate the usefulness of Max-Mean and Max-Median filters in preserving the edges of clouds and structural backgrounds, which helps in detecting small-targets. Subsequently, anti-mean and anti-median operations result in good performance of detecting targets against moving clutter. The raw image is first filtered by max-mean/max-median filter. Then the filtered output is subtracted from the original image to enhance the potential targets. A thresholding step is incorporated in order to limit the number of potential target pixels. The threshold is obtained by using the statistics of the image. Finally, the thresholded images are accumulated so that the moving target forms a continuous trajectory and can be detected by using the post-processing algorithm. It is assumed that most of the targets occupy a couple of pixels. Head-on moving and maneuvering targets are not considered. These filters have ben tested successfully with the available database and the result are presented.
This paper deals with the problem of detection and tracking of point-targets from a sequence of IR images against slowly moving clouds as well as structural background. Many algorithms are reported in the literature for tracking sizeable targets with good result. However, the difficulties in tracking point-targets arise from the fact that they are not easily discernible from point like clutter. Though the point-targets are moving, it is very difficult to detect and track them with reduced false alarm rates, because of the non-stationary of the IR clutter, changing target statistics and sensor motion. The focus of research in this area is to reduce false alarm rate to an acceptable level. In certain situations not detecting a true target is acceptable, but declaring a false target as a true one may not be acceptable. Although, there are many approaches to tackle this problem, no single method works well in all the situations. In this paper, we present a multi-mode algorithm involving scene stabilization using image registration, 2D spatial filtering based on continuous wavelet transform, adaptive threshold, accumulation of the threshold frames and processing of the accumulated frame to get the final target trajectories. It is assumed that most of the targets occupy a couple of pixels. Head-on moving and maneuvering targets are not considered. It has been tested successfully with the available database and the results are presented.
Recent progress on the development of 2D staring Focal Plane Array (FPA) sensors has opened up numerous applications ranging from military to commercial. Staring FPA sensors, particularly Infra-Red FPAs, have certain limitations such as non-uniformity, cross-talk and fill-factor. Sensitivity performance of FPAs to these parameters is important particularly for image registration and tracking applications. Each detector in FPA has an active area and dead space around it. The ratio of the active area to the total detector area is called the fill-factor. The dead space around the active detector surface contributes to the loss of information, which in turn can lead to poor performance of the FPA. In this paper, the effect of fill- factor on the performance of the correlation based registration and tracking algorithms is presented. Assuming an input scene, a sampled image output from a FPA is modeled based on a given fill-factor. Sequences of output images resulting from different fill-factors are used for simulation for evaluating the image registration and tracking performance. It is observed that a poor fill-factor results in deteriorated performance.
Proc. SPIE. 3701, Infrared Imaging Systems: Design, Analysis, Modeling, and Testing X
KEYWORDS: Digital signal processing, Signal processing, Infrared search and track, Sensors, Staring arrays, Data conversion, Clocks, Target detection, Detection and tracking algorithms, Analog electronics
Generally, Infrared Search and Track systems use linear focal-plane-arrays with time-delay and integration, because of their high sensitivity. However, the readout is a cumbersome process and needs special effort. This paper describes signal processing and hardware (HW) implementation issues related to front-end electronics, non-uniformity compensation, signal formatting, target detection, tracking and display system. This paper proposes parallel pipeline architecture with dedicated HW for computationally intensive algorithms and SW intensive DSP HW for reconfigurable architecture.
Infrared search and track (IRST) system is a wide field of view surveillance system, meant for autonomous search, detection, acquisition, and cue of potential targets. The first and second generation IRSTs utilized detectors with multiple elements followed by discrete preamplifiers for signal read-out. They have many performance limitations. With the advent of infrared focal plane array (IRFPA) sensors, the present trend is to build IRSTs based on line FPA sensors to achieve higher sensitivity and resolution. However, due to system limitations of line IRFPA sensors, scanning mode of IRST cannot be stopped at any desired position to scan a small sector of interest. They also suffer from more false alarms in target detection. In future, it may be desirable to reduce false alarms, and also to use an IRST system for closed-loop- tracking of a potential target, in addition to its surveillance mode. IRST based on area array sensors may be a better option for this purpose, but it may pose some problems when used in a surveillance mode. This paper addresses this issue. Design considerations of all sub-systems of an IRST based on line/area array sensors, such as scanner assembly, interface electronics with the sensor, nonuniformity correction, signal processor, and the display methodology to cover 360 degrees are also discussed.
Recent reports indicate that cooled and uncooled IR focal plane array sensors are progressing to a field-worthy level for commercial and defense applications. They offer higher sensitivity, amenability to signal processing and mechanical simplicity. However these sensors contain large detector-to- detector dark current (offset) and responsivity (gain) variations. These variations result in a severe problem called fixed pattern noise that can mask/distort the image obtained from the sensor. The correction process is generally termed as nonuniformity compensation. Conventional two-point compensation techniques are accurate enough, but require built-in controllable temperature references along with mechanical and electro-optical shutters. Therefore this compensation technique detracts the mechanical simplicity of using IR focal plane arrays. Scene-based nonuniformity techniques dispenses with the requirement of temperature references and shutters, but are not accurate enough for certain applications. This paper discusses two-point and scene-based nonuniformity compensation algorithms and proposes an empirical formula to automatically calculate the scene constants, which is an essential step towards practical applications. This paper reports the analyzed results of testing the algorithms on a number of IR images. A practical problem of 'artifacts' which arise when using scene-based nonuniformity compensation is also discussed. A common hardware scheme to implement both the algorithms is also presented in this paper.
Area correlation tracking is an effective method of tracking targets or a scene of interest that do not have prominent features or high contrast with respect to the background. The essential step in area correlation tracking is to find the 'best fit' of a target/area reference image in a real- time search image acquired by the imaging sensor. The solutions for these are broadly classified as template and feature level matching algorithms. Various algorithms reported in the open literature have different degrees of performance under different field conditions. For certain autonomous tracking applications where there is relative motion between the sensor and the scene/target, it is essential to continuously register a part of previous scene in the next frame. When the reference is thus updated for each frame, there is a possibility that the system may become divergent, if mis-registration is not detected. A correct registration omitted as a wrong registration is tolerable, but a mis-registration committed as a correct registration is a catastrophe. In view of this, it is essential to improve the reliability of registration, and take a confidence measure whenever it is in doubt. This paper addresses that issue and proposes a scheme to improve the performance of area correlation trackers.
Autonomous navigation of a small, slow speed, low altitude unmanned aerial vehicle (UAV) have many potential applications. UAVs are generally used for (i) remote sensing the areas which are difficult to approach, (ii) surveillance, (iii) target designation or jamming, (iv) weapon delivery or as a weapon by itself, etc. Another potential application would be to use them as cost-effective loitering vehicles near the potential enemy sites, creating nuisance value. In most applications, the solution for autonomous navigation is to install inertial navigation systems (INS) on board the flight vehicle and regularly update the INS as often and as accurately as possible. In this paper, different INS updating techniques are briefly mentioned with their advantages and drawbacks, and then a multi-mode image based navigation is proposed. Using several body mounted focal-plane-array imaging sensors, a bigger image is obtained to get sufficient features for matching. The emphasis in this paper is to get vehicle's speed, direction/attitude, and 'running fixes' by using very reliable 'area correlation' tracking. A combination of feature based scene matching along with area correlation is proposed for updating INS. The effort in this paper is to bring out conceptual ideas of image based navigation to make an UAV to perform better and at the same time cost effective.
Proc. SPIE. 1950, Acquisition, Tracking, and Pointing VII
KEYWORDS: Image registration, Evolutionary algorithms, Simulation of CCA and DLA aggregates, Nickel, Silicon, Infrared imaging, Signal to noise ratio, Bismuth, Detection and tracking algorithms, Statistical analysis
The performance of image registration algorithms for IR images depends on different system parameters such as SNR, gain, offset, image variance etc. in a very complex fashion. A statistical approach has been adopted to predict the performance of image registration algorithms , classical Correlation and Mean Absolute Difference algorithms, in terms of probability of correct registration as a function of different system parameters. Considering the complexity of the problem, some simplifying assumptions have been made regarding the random nature of different images as well as criteria function values which may not be strictly true for real environments. This analysis helps in understanding the sensitivity of different algorithms to different system parameters. This in turn helps to compare the performance of different registration algorithms in differe nt environments. A scheme for choosing adaptive threshold to meet constraints as probability of false hit and probability of miss, has also been proposed
Proc. SPIE. 1697, Acquisition, Tracking, and Pointing VI
KEYWORDS: Detection and tracking algorithms, Image sensors, Optical sensors, Algorithm development, Silicon, Correlation function, Control systems, Simulation of CCA and DLA aggregates, Image registration, Image processing
Area correlation tracking (ACT) is an effective solution for tracking targets that have neither prominent features nor high contrast with the background. The essential step in area correlation tracking is to find the position of best-fit between the reference image of the terrain surrounding the desired target to the real-time scene acquired by an imaging sensor. The image matching is accomplished by the basic mathematical correlation coefficient algorithm (CCA) or mean absolute difference (MAD) algorithm or other algorithms derived from these algorithms. The reference image has to be updated very frequently to circumvent the problems associated with image growth, aspect changes, rotation, etc. Otherwise the image tracking algorithm may become divergent with time thereby losing the track. This paper addresses the real-time implementation of area correlation algorithms for an on-board application. The implementation poses challenges because of the very large number of integer multiplications required by the correlation coefficient algorithm. In view of the space and power constraints in on board applications, the realization becomes all the more complex. To realize the system, a parallel pipelines architecture is adapted. Very high speed arithmetic devices are used for computations, programmable logic devices for the high speed control and a dual microprocessor based system for overall control. This hardware has been evaluated by integrating with a full fledged imaging seeker in captive flight trials. The results are presented.
Real-time hardware developed for image processing applications such as image enhancement, segmentation, image registration, pattern recognition, etc. can not be thoroughly debugged and analyzed for its performance by using conventional test equipment. To cite an example, the hardware for image registration may be failing intermittently or the registration point may be drifting even when the scene and the sensor are static. There is no way of knowing whether the malfunction is due to input data or any noise glitches. Therefore, for evaluating real-time image processing hardware, it is necessary to acquire image sequence data in real-time and also capture the status of the tracker in real-time. Test equipment like logic analyzer, microprocessor development systems do not have the capability to acquire and store image sequences. General purpose data acquisition systems on the IBM PC compatible computer are not suitable as the data rate required is about 500 kbytes/sec. the design of an interface card for achieving this data rate using IEEE-488 (GPIB) interface is described. Results on the evaluation of image processing hardware using this card are also presented.
Autonomous fire and forget weapons have gained importance to achieve accurate first pass kill by hitting the target at an appropriate aim point. Centroid of the image presented by a target in the field of view (FOV) of a sensor is generally accepted as the aimpoint for these weapons. Centroid trackers are applicable only when the target image is of significant size in the FOV of the sensor but does not overflow the FOV. But as the range between the sensor and the target decreases the image of the target will grow and finally overflow the FOV at close ranges and the centroid point on the target will keep on changing which is not desirable. And also centroid need not be the most desired/vulnerable point on the target. For hardened targets like tanks, proper aimpoint selection and guidance up to almost zero range is essential to achieve maximum kill probability. This paper presents a centroid tracker realization. As centroid offers a stable tracking point, it can be used as a reference to select the proper aimpoint. The centroid and the desired aimpoint are simultaneously tracked to avoid jamming by flares and also to take care of the problems arising due to image overflow. Thresholding of gray level image to binary image is a crucial step in centroid tracker. Different thresholding algorithms are discussed and a suitable algorithm is chosen. The real-time hardware implementation of centroid tracker with a suitable thresholding technique is presented including the interfacing to a multimode tracker for autonomous target tracking and aimpoint selection. The hardware uses very high speed arithmetic and programmable logic devices to meet the speed requirement and a microprocessor based subsystem for the system control. The tracker has been evaluated in a field environment.
Proc. SPIE. 1699, Signal Processing, Sensor Fusion, and Target Recognition
KEYWORDS: Image registration, Sensors, Simulation of CCA and DLA aggregates, Signal to noise ratio, Infrared imaging, Infrared radiation, Staring arrays, Detection and tracking algorithms, Curium, Image sensors
Image registration techniques have gained importance for many applications such as area correlation tracking, handing over recognized target scenes from one sensor to another sensor, tracking an area of interest before detection, tracking point targets, and the latest being multi- target handling capability for defense needs, scene stabilization, etc. No single image registration algorithm (IRA) can work satisfactorily for all applications and in all environments. This paper analyzes the suitability of image registration algorithms for infrared images. Infrared images are characterized by low contrast and sensor nonuniformities such as offset-errors, gain variations resulting in fixed pattern noise (FPN). Particularly, in focal plane arrays (FPA), the output of each detector is characterized by `ax + b' where `a' and `b' are gain and dc offset terms respectively, and `x' is photon flux level falling on the detectors. These parameters and especially their variations from detector element to detector element effect the performance of image registrations algorithms. In this paper, the basic IRAs are analyzed in this context of infrared images with low contrast and FPN. Simulated results using real world images are presented. Novel and inexpensive confidence and redundancy measures have been proposed to improve the performance by detecting misregistrations.