The hyperspectral imaging system (HIS) using a Fourier transform infrared (FTIR) spectrometer is one of the key technologies for detection and identification of chemical warfare agents (CWAs). Recently, various detection algorithms based on machine learning techniques have been studied. These algorithms are robust against performance degradation caused by noise signatures generated by FTIR instruments. However, interference signatures from background materials degrade detection performance. In this paper, we propose an efficient algorithm that uses a support vector machine (SVM) classifier to detect CWAs. In contrast to the conventional algorithms that use measured spectra to train the SVM classifier, the proposed algorithm trains the SVM classifier using CWA signatures obtained by removing background signatures from measured spectra. Therefore, the proposed algorithm is robust against the performance degradation induced by interference signatures from background materials. Experimental results verify that the algorithm can detect CWA clouds effectively.
A passive Fourier transform infrared (FTIR) spectrometer is an instrument that can detect and identify chemical contaminants. An FTIR spectrometer exploits the infrared radiation of the surrounding terrain as a light source and receives a mixed signal of background signal, gas signal, and noise. The performance of most detection algorithms for detecting gaseous plumes, such as the normalized matched filter (NMF) and adaptive subspace detector (ASD), deteriorates due to the noise generated by an FTIR spectrometer. In this paper, a noise reduction algorithm based on the maximum noise fraction (MNF) transform to improve the performance of hazardous gas detection algorithms is proposed. We apply the MNF transform to the measured spectra and remove the high noise fraction component. Then the noise-reduced spectra are restored by conducting the inverse MNF transform. The experimental results show that the proposed algorithm reduces the noise and enhances the gas detection performance.
Recently, a hyperspectral imaging system (HIS) with a Fourier Transform InfraRed (FTIR) spectrometer has been widely used due to its strengths in detecting gaseous fumes. Even though numerous algorithms for detecting gaseous fumes have already been studied, it is still difficult to detect target gases properly because of atmospheric interference substances and unclear characteristics of low concentration gases. In this paper, we propose detection algorithms for classifying hazardous gases using a deep neural network (DNN) and a convolutional neural network (CNN). In both the DNN and CNN, spectral signal preprocessing, e.g., offset, noise, and baseline removal, are carried out. In the DNN algorithm, the preprocessed spectral signals are used as feature maps of the DNN with five layers, and it is trained by a stochastic gradient descent (SGD) algorithm (50 batch size) and dropout regularization (0.7 ratio). In the CNN algorithm, preprocessed spectral signals are trained with 1 × 3 convolution layers and 1 × 2 max-pooling layers. As a result, the proposed algorithms improve the classification accuracy rate by 1.5% over the existing support vector machine (SVM) algorithm for detecting and classifying hazardous gases.
A hyperspectral imaging system (HIS) with a Fourier transform infrared (FTIR) spectrometer is an excellent method for the detection and identification of gaseous fumes. Various detection algorithms can remove background spectra from measured spectra and determine the degree of spectral similarity between the extracted signature and reference signatures of target compounds. However, given the interference signatures caused by FTIR instruments, it is impossible to extract the spectral signatures of target gases perfectly. Such interference signatures degrade the detection performance. In this paper, a detection algorithm for gaseous fumes using a multiclass support vector machine (SVM) classifier is proposed. The proposed algorithm has a training step and a test step. In the training step, the spectral signatures are extracted from measured spectra which are labeled. Then, hyperplanes which classify gas spectra are trained and the multiclass SVM classifier outcomes are calculated using the hyperplanes. In the test step, spectral signatures extracted from unknown measured spectra are substituted to the SVM classifier, after which the detection result is obtained. This multiclass SVM classifier robustly responds to performance degradation caused by unremoved interference signatures because it trains not only gaseous signatures but also the related interference signatures. The experimental results verify that the algorithm can effectively detect hazardous clouds.
In this paper, we propose multiple demeaning filters for small target detection in infrared (IR) images. The use of
a demeaning filter is a promising method which detects a small object by removing the background components
with a mean filter. The main factors in the design of a demeaning filter are two types of demeaning methods
and the size of its window. We compare two demeaning methods, the sliding window method and the grid
method, and we analyze the trade-off between the window size and the performance of the demeaning filters and
present limitations related to their use. To overcome the drawbacks of a conventional demeaning filter, the use
of multiple demeaning filters with filters of various sizes is considered. The proposed method not only has the
advantage of being able to detect a small object in a densely cluttered environment, but it also can be used with
low complexity with an integral image. Experimental results demonstrate the robustness and stability of the
proposed multiple demeaning filters with low computational complexity compared with conventional methods.
In this paper, we propose a denoising method for hyperspectral images using a joint bilateral filter. The joint bilateral
filter with the fused image of hyperspectral image bands is applied on the noisy image bands. This fused image is a
single grayscale image that is obtained by the weighted summation of hyperspectral image bands. It retains the features
and details of each hyperspectral image band. Therefore the joint bilateral filter with the fused image is powerful in
reducing noise while preserving the characteristics of the individual spectral bands. We evaluated the performance of the
proposed noise reduction method on hyperspectral imaging systems, which we developed for visible and near-infrared
spectral regions. Experimental results show that the proposed method outperforms the conventional approaches, such as
the basic bilateral filter.
This paper presents an extended bilateral filter with spectral angles and a visualization scheme for hyperspectral image
data. The conventional bilateral filter used to be implemented using a position vector and a luminance value at each pixel
in the scene. Since hyperspectral image data can provide a spectrum vector that has hundreds of bands at each pixel, we
propose an extended bilateral filter using spectral angles. The proposed bilateral filter can be used for extracting and
preserving the spectrum edges of the hyperspectral image. A visualization scheme for hyperspectral images exploiting
the proposed bilateral filter has been also proposed. When objects that have the similar tristimulus intensity but different
spectrums are mixed, they can be separated through the proposed visualization scheme. The resulting images show that
the proposed scheme facilitates anomaly detection in the hyperspectral scenes.
In this paper, we propose an overall target tracking scheme performing image stabilization, detection, tracking,
and classification in the IR sensored image. Firstly, in the image stabilization stage, a captured image is
stabilized from visible frame-to-frame jitters caused by camera shaking. After that, the background of the
image is modeled as Gaussian. Based on the results of the background modeling, the difference image between a
Gaussian background model and a current image is obtained, and regions with large differences are considered as
targets. The block matching method is adopted as a tracker, which uses the image captured from the detected
region as a template. During the tracking process, positions of the target are compensated by the Kalman filter.
If the block matching tracker fails to track targets as they hide themselves behind obstacles, a coast tracking
method is employed as a replacement. In the classification stage, key points are detected from the tracked image
by using the scale-invariant feature transform (SIFT) and key descriptors are matched to those of pre-registered
The demeaning filter detects a small object by removing a background with a mean filter as well as the covariance of an
object and backgrounds. The factors considered in the design of the demeaning filter are the method of demeaning, which
involves subtracting the local mean value from all pixel values, and the acquisition of templates for both the object and the
background. This study compares the sliding window method and the grid method as a demeaning method, and studies the
method of acquisition of an object template. Moreover, a method involving the use of previous frames, a mean filter, and an
opening operation are studied in an effort to acquire a background template. Based on the results of this study, a practical
design of a demeaning filter that is able to detect a small object in an IR image in real time is proposed. Experiment results
demonstrate the superiority of the proposed design in detecting a small object following a 2-D Gaussian distribution even
under severe zero-mean Gaussian noise.
In this paper, a new condition for the target is proposed to increase the robustness of the facet-based detection method
for zero-mean Gaussian noise. In the proposed algorithm, the pixels detected from the maximum extremum condition are
checked further to discern if they are false maximum points in the proposed scheme. The experimental results show that
the proposed algorithm is much more robust for zero-mean Gaussian noise than the conventional detection method.
Recently, multi-sensor image fusion systems and related applications have been widely investigated. In an image fusion
system, robust and accurate multi-modal image registration is essential. In the conventional method, the image registration
process starts with manually-pointed corresponding pairs in both sensored images. Using these corresponding pairs, a
transform matrix is initialized and refined through an optimization process. In this paper, we propose a new automatic
extraction method for such corresponding pairs. The Harris corner detector is employed to extract feature points in both
EO/IR images individually. Patches around the detected feature points are matched with a probabilistic criterion, mutual information
(MI), which is a preferred measure for image registration due to its robust and accurate performance. Simulation
results show that the proposed scheme has a low time complexity and extracts corresponding pairs well.
Target segmentation plays an important role in the entire target
tracking process. This process decides whether the current pixel
belongs to the target region or not. In the previous works, the
target region was extracted according to whether the intensity of
each pixel is larger than a certain value. But simple binarization
using one feature, i.e. intensity, can easily fail to track as
condition changes. In this paper, we employ more features such as
intensity, deviation over time duration, matching error, etc.
rather than intensity only and each feature is weighted by the
weighting logic, which compares the characteristics in the target
region with that in the background region. The weighting logic
gives a higher weight to the feature which has a large difference
between the target region and the background region. So the
proposed segmentation method can control the priority of features
adaptively and is robust to the condition changes of various
We present a fast parameter estimation method for image segmentation using the maximum likelihood function. The segmentation is based on a parametric model in which the probability density function of the gray levels in the image is assumed to be a mixture of two Gaussian density functions. For the more accurate parameter estimation and segmentation, the algorithm is formulated as a compact iterative scheme. In order to reduce computation time and make convergence fast, histogram information is combined into the algorithm. In the iterative computation, the performance of the algorithm greatly depends on the initial values and properly selected initial estimates make convergence fast. A reasonable approach about the computation of initial parameter is also proposed.
A real-time adaptive segmentation method based on new distance features is proposed for the centroid tracker. These novel features are distances from the center point of a predicted target to each pixel by a tracking filter in extraction of a moving target. The proposed method restricts clutters with target-like intensity from entering the tracking window with low computational complexity for real- time applications compared with other complex feature-based methods. Comparative experiments show that the proposed method is superior to the other segmentation methods based on the intensity feature only in target detection and tracking.