IR Image Sequences Research Articles

Evolutionary algorithm for data association and IMM-based target tracking in IR image sequences

Simultaneous tracking of multiple maneuvering and non-maneuvering targets in the presence of dense clutter and in the absence of any a priori information about target dynamics is a challenging problem. A successful solution to this problem is to assign an observation to track for state update known as data association. In this paper, we have investigated tracking algorithms based on interacting multiple model to track an arbitrary trajectory in the presence of dense clutter. The novelty of the proposed tracking algorithms is the use of genetic algorithm for data association, i.e., observation to track fusion. For data association, we examined two novel approaches: (i) first approach was based on nearest neighbor approach and (ii) second approach used all observations to update target state by calculating the assignment weights for each validated observation and for a given target. Munkres’ optimal data association, most widely used algorithm, is based on nearest neighbor approach. First approach provides an alternative to Munkres’ optimal data association method with much reduced computational complexity while second one overcomes the uncertainty about an observation’s source. Extensive simulation results demonstrate the effectiveness of the proposed approaches for real-time tracking in infrared image sequences.

A Particle Filter Based Algorithm for State Estimation of Dim Moving Point Targets in IR Image Sequence

Under the condition of the targets’ initial information is already estimated successfully, This paper presents a real-time target tracking method based on particle filter (PF) update algorithm. According to the particles' transmission characteristics and the measurements from a single frame detection, the algorithm estimate a target’s following moving state. In this way, the dim moving point target could be tracked successfully under the low signal-noise-ratio (SNR) in IR image sequence. But it comes to the multiple targets’ tracking; we should take the data integration into account. In order to reduce the amount of calculation, we use the data fusion method to divide the measurements in the overlapped parts of windows, appoints the measurements for the corresponding target. That is the Nearest Neighbor Standard Filter (NNSF) theory is used to choose the target for a measurement. The paper introduced the related theory and the concrete steps for accomplish the algorithm and also simulated the proposed tracking algorithm on the MATLAB platform. Experimental analysis and results showed that the algorithm achieved real-time, dynamic stability and robustness while track the random moving targets in high clutter environment.

Infrared moving target detection and tracking based on tensor locality preserving projection

This paper presents an algorithm for detecting and tracking dim moving target in IR image sequence with low SNR. The algorithm, which is designed based on the locality preserving projection, accepts tensors as inputs. The justification for the algorithm comes from the role of the generalized eigenvalue problem in providing an optimal embedding for the manifold. Not only does the proposed method inherit the attractive characteristics of the locality preserving projections in terms of exploiting the intrinsic manifold structure, it is also appealing in terms of significant reduction in both space complexity and time complexity. Experimental results on two IR image sequences demonstrate the effectiveness of the proposed algorithm.

English

Lossless image compression has long been recognised as an important need for several applications such as medical imaging, storage of critical IR image sequences, and remote sensing. In this paper, a simple, fast and easy to realisable-on-hardware lossless video compression algorithm is proposed that is well-suited for IR imageries. Context-based median predictor is used for prediction of reference pixels. Three neighboring pixels are used as context for prediction. Inter-frame coding is performed by encoding the redundant pixels in an efficient way, using 1-bit code. Finally, the arithmetic coder is used as entropy coding. The proposed algorithm is able to operate in image compression and video compression mode. The proposed Median Predictor based Lossless Video Compression (MPLVC) algorithm is compared with Joint Pictures Experts Group-Lossless (JPEG-LS) and Fast and Efficient Lossless Image Compression System (FELICS) for compression performance. The results demonstrate that proposed algorithm is superior in encoding rate with added advantage in simplicity and ease in realization on hardware.Defence Science Journal, 2009, 59(2), pp.183-188, DOI:http://dx.doi.org/10.14429/dsj.59.1507

Simultaneous velocity and diffusivity mapping in the case of 3-D transient heat diffusion: heat pulse thermography and IR image sequence analysis

A non uniform heat pulse is used as a non intrusive marker on opaque liquids or highly deformable solid media, in order to estimate both the full velocity and thermal diffusivity fields from thermal image processing. The difficult issue of identifying both the local transport and diffusion properties is addressed by adopting two different inverse approaches. First, a new specific version of the Total Least Squares (TLS) method is implemented, yielding both sensitivity analysis and confidence estimation, with no prior knowledge of the measurement noise variance. A second strategy is proposed from an iterative Digital Image Correlation (DIC) approach, yielding an hybrid DIC/TLS method. Both methods are first validated on simulated data,and calibrated with homogenous materials under uniform displacement. Two different grids are used in order to analyze the effect of the spatial period of the heat pulse on the retrieved parameters. Then, the effective velocity and thermal diffusivity fields are retrieved from an experimental bench in the case of a moving Thermisorel® sample. The Fine grid is found to yield better estimation results than the coarse grid. Thermal diffusivity mapping obtained with the TLS method is more trusted.

Parameter Adjustment for a Dynamic Programming Track-before-Detect-Based Target Detection Algorithm

This paper addresses the problem of tracking a dim moving point target in a sequence of IR images. The proposed tracking system, based on the track-before-detect (TBD) approach, is designed to track and detect dim maneuvering targets froman image sequence under low SNR conditions; a dynamic programming algorithm (DPA) is used to process the frames in the sequence. This paper deals with the practical issues of setting the parameters of our algorithm to maximize tracking capability. In our flexible approach, we are able to accommodate different levels of noise and different target velocities and mobilities. Our algorithm is tested on real data to determine its efficacy.

Improved power-law-detector-based moving small dim target detection in infrared images

A novel algorithm based on an improved power-law detector is proposed to detect moving small dim targets against heavy cluttered background in IR image sequences. As the gray-scale value of a pixel in an image fluctuates when a target passes by, this fluctuation can be viewed as a weak transient signal, and can be detected by the power-law detector, which performs well for transient signal detection. The experimental results illustrate that the proposed algorithm is effective and adaptable for moving small dim target detection in image sequences with heavy clutter

Fast new small-target detection algorithm based on a modified partial differential equation in infrared clutter

To detect and track moving dim targets against the complex cluttered background in IR image sequences is still a difficult problem because the nonstationary structured background clutter usually results in low target detectability and a high probability of false alarm. A new adaptive anisotropic filter based on a modified partial differential equation (AFMPDE) is proposed to detect a small target in such a strong cluttered background. A regularizing operator is employed to adaptively eliminate structured background and simultaneously enhance the target signal. The proposed algorithm's performance is illustrated and compared with the two-dimensional least mean square (TDLMS) adaptive filter on real IR image data. Experimental results demonstrate that the proposed novel method is fast and effective.

Temporal Profile Based Small Moving Target Detection Algorithm in Infrared Image Sequences

A new algorithm is presented which deals with the problem of detecting small moving targets in infrared image sequences that also contain drifting and evolving clutter. Through development of models of the temporal behavior of the static background, target and cloud edge on a single pixel basis, the new algorithm employing the connecting line of the stagnation points (CLSP) of the temporal profile as the baseline is created and tested. The deviation of the temporal profile and its CLSP is analyzed and it is determined that the distribution of the residual temporal profile obtained by subtracting the baseline from the temporal profile can be modeled by a Gaussian distribution. The occurrences of the targets have intensity values significantly different to the distribution of the residual temporal profile. Unlike the conventional 3-D method, this new algorithm operates on the temporal profile in 1-D space, not in 3-D space, thus having a higher computational efficiency. Experiments with real IR image sequences have proved the validity of the new approach.

Archetypes and principal components of an IR image sequence

In the last years, the principal component analysis (PCA) has been used in the thermal NDT/E (TNDT/E) field as a useful tool for signal enhancement and feature extraction. Even though many efforts were addressed to understand the meaning of the principal components, no apparent connections have been found with the thermal processes involved in the test. In this paper, PCA is compared with another matrix factorization method called archetypal analysis (AA). Contrary to PCA, AA does not imply any transformation of the original reference system and moreover the archetypes are intrinsically related to the experimental data. The main drawback is that AA is heavy in terms of computational time. To accelerate the convergence, the pixel purity index (PPI) algorithm is used. Results of PCA and AA applied to thermogram sequences are presented and discussed.

Moving weak point target detection and estimation with three-dimensional double directional filter in IR cluttered background

We assess the performance of a novel three-dimensional double directional filtering 3DDDF algorithm for detecting and tracking weak moving dim targets against a complex cluttered background in infrared IR image sequences. This proposed method increases the tar- get energy accumulation ability further than the three-dimensional direc- tional filter 3DDF method. Prior to the filtering, a new prewhitening method termed a three-dimensional spatialtemporal adaptive prediction filter TDSTAPF is used to suppress the cluttered background. Exten- sive experiment results demonstrate the proposed algorithms' ability to detect weak dim point targets against a complex cloud-cluttered back- ground in real IR image sequence and the performance comparisons of the proposed method and 3DDF. © 2005 Society of Photo-Optical Instrumentation

Airborne threat detection in navy IRST systems

A new algorithm is presented for the detection of airborne targets by means of infrared sensors operating in naval surveillance scenarios. The proposed algorithm consists of two steps: background clutter removal and detection over the residual clutter. The algorithm is fully automatic and its implementation does not require the tuning of any parameter other than the threshold for setting the probability of false alarm. The algorithm performance is investigated by means of a sequence of experimental IR images taken in a typical maritime environment. The results show that the proposed algorithm outperforms a standard detection algorithm specifically tailored to the analysed scenario. A statistical analysis of the experimental data is also performed to validate the hypotheses used to derive the detection algorithm.

Invariant object tracking using fringe-adjusted joint transform correlator

A novel real-time-invariant target tracking algorithm is proposed for forward-looking IR (FLIR) image sequences based on a modified synthetic discriminant function (SDF)-based fringe-adjusted joint transform correlator (JTC) algorithm. The proposed approach utilizes fringe-adjusted JTC-based tracking, which efficiently compensates camera motion and accommodates 3-D distortions. To further increase the tracking accuracy, the target model is continuously updated using a modified version of the synthetic discriminant function (SDF) algorithm. The proposed technique can track small objects comprising only a few pixels and is capable of compensating the high egomotion of the sensor. We demonstrate the robustness of the proposed technique by computer simulation performed on sequences of real-life FLIR imagery taken from an airborne moving platform.

Detecting and tracking dim moving point target in IR image sequence

This paper presents a novel algorithm for detecting and tracking dim moving point target in IR image sequence with low SNR. Original images are preprocessed using temperature non-linear elimination and Top-hat operator, and then a composite frame is obtained by reducing the three-dimensional (3D) spatio-temporal scanning for target to 2D spatial hunting. Finally the target trajectory is tracked under the condition of constant false-alarm probability (CFAR). Based on the experimental results, the algorithm can successfully detect dim moving point target and accurately estimate its trajectory. The algorithm, insensitive to the velocity mismatch and the changes of statistical distribution of background or noise, is adaptable to real-time target detection and tracking.

Statistical analysis of IR thermographic sequences by PCA

Automatic processing of IR sequences is a desirable target in Thermal Non-Destructive Evaluation (TNDE) of materials. Unfortunately, this task is made difficult by the presence of many undesired signals that corrupt the useful information detected by the IR camera. In this paper the Principal Component Analysis (PCA) is used to process IR image sequences to extract features and reduce redundancy by projecting the original data onto a system of orthogonal components. As a thermographic sequence contains information both in space and time, the way of applying the PCA to these data cannot be straightforwardly borrowed from typical applications of the PCA where the information is mainly spatial (e.g. remote sensing, face recognition). This peculiarity has been analysed and the results are reported. Finally, in addition to the use of the PCA as an unsupervised method, its use in a “learning and measuring” configuration is considered.

スターセンサ画像の暗い星検出への繰り返し型最大値フィルタの応用

To autonomously determine the attitude of a spacecraft by a star sensor without a rate sensor, it is necessary to detect and track stars even when the attitude change rate of the spacecraft, i.e., the body rate, is high. A high body rate causes a motion blur and decreases the signal-to-noise ratio (SNR) of stars, making stable detection of many stars difficult. In this paper, in order to overcome this difficulty we propose a star detection method using the recursive maximum filter (RMF). Though the RMF was originally proposed for the detection of small moving targets with low SNR from an IR image sequence, we show that it also works well for images from a star sensor. In a star sensor image, both bright and dim stars exist at the same time. In the proposed method, bright stars are detected and masked in each image as a preprocessing step. Then, by applying the RMF to the masked image sequence, the background noise is suppressed and dim stars are gradually enhanced, until at last they can be detected by thresholding. By using this method, the present upper limit of the body rate (1deg/s) for star tracking will be approximately doubled.

Space–time processing for the detection of airborne targets in IR image sequences

The problem of detecting airborne targets in a sequence of images recorded by a long range infra-red sensor is investigated. The generalised likelihood ratio test detector is derived. The detector structure and its actual implementation are discussed. Approximated expressions of the false alarm and detection probabilities are obtained and validated by means of simulations. Example results obtained on a set of experimental data are reported.

Point target detection in IR image sequences: a hypothesis-testing approach based on target and clutter temporal profile modeling

We approach the problem of point target detection in infrared image sequences by modeling the temporal behavior of clutter and targets on a single-pixel basis. These models, which are experimentally verified, are then used to develop a temporal likelihood-ratio test and derive the corresponding decision rule. We demonstrate the effectiveness of the technique by applying it to real infrared image sequences containing targets of opportunity and evolving cloud clutter. The physical models and resulting hypothesis-testing approach could also be applicable to other image-sequence-processing scenarios, using acquisition systems besides infrared imaging, such as the detection of small moving objects or structures in a biomedical or biological imaging scenario, or the detection of satellites, meteors, or other celestial bodies in night-sky imagery acquired using a telescope.

Knowledge-based analysis of satellite oceanographic images

A knowledge-based program to assist humans interpret satellite infrared images of a region of the Atlantic Ocean is described. the organization of the system and the nature of its knowledge base are discussed. the operation of the system in tracking mesoscale events in a registered sequence of IR images is illustrated. © 1989 Wiley Periodicals, Inc.