In this paper, a method of spindle extraction of target in inverse synthetic aperture radar (ISAR) image is proposed which depends on Radon Transform. Firstly, utilizing Radon Transform to detect all straight lines which are collinear with these line segments in image. Then, using Sobel operator to detect image contour. Finally, finding all intersections of each straight line and image contour, the two intersections which have maximum distance between them is the two ends of this line segment and the longest line segment of all line segments is spindle of target. According to the proposed spindle extraction method, one hundred simulated ISAR images which are respectively rotated 0 degrees, 10 degrees, 20 degrees, 30 degrees and 40 degrees in counterclockwise are used to do experiment and the proposed method and the detection results are more close to the real spindle of target than the method based on Hough Transform .
In this paper, we present a novel automatic image segmentation method, which combines the active contour method and
the saliency map method. The saliency map which is obtained by inversing the spectral residual of the image brings a
priori knowledge to bear on the image segmentation. The initial level set function is constructed from saliency map. In
this way, an automatic initialization of the level set evolution can be obtained. This method can minimize the iterations
of the level set evolution. The efficiency and accuracy of the method are demonstrated by the experiments on the
synthetic and real images.
Traditional learning-based boundary extraction algorithms classify each pixel edge separately and then get boundaries from the local decisions of a classifier. However, we propose a supervised learning method for boundary extraction by using edgelets as boundary elements. First, we extract edgelets by clustering probabilities of boundary. Second, we use features of edgelets to train a classifier that determines whether an edgelet belongs to a boundary. The classifier is trained by utilizing edgelet features, including local appearance, multiscale features, and global scene features such as saliency maps. Finally, we use the classifier to decide the probability that the edgelet belongs to the boundary. The experimental results in the Berkeley Segmentation Dataset demonstrate that our algorithm can improve the performance of boundary extraction.
Based on mean preserving bi-histogram equalization (BBHE), an adaptive image histogram equalization algorithm for
contrast enhancement is proposed. The threshold is gotten with adaptive iterative steps and used to divide the original
image into two sub-images. The proposed Iterative of Brightness Bi-Histogram Equalization overcomes the
over-enhancement phenomenon in the conventional histogram equalization. The simulation results show that the
algorithm can not only preserve the mean brightness, but also keep the enhancement image information effectively from
visual perception, and get a better edge detection result.
Under the support vector machine framework, the support value analysis-based image fusion has been studied, where the
salient features of the original images are represented by their support values. The support value transform (SVT)-based
image fusion approach have demonstrated some advantages over the existing methods in multisource image fusion. In
this paper, the directional support value transform (DSVT) is applied to the denoising of some standard images
embedded in white noise and the X-ray images. This directional transform is not norm-preserving and, therefore, the
variance of the noisy support values will depend on the scales. And then we use the hard-thresholding rule for estimating
the unknown support values in different scales and the thresholding is scale-dependent. The peak signal noise ratio
(PSNR) is used as an "objective" measure of performance, and our own visual capabilities are used to identify artifacts
whose effects may not be well-quantified by the PSNR value. The experimental results demonstrate that simple
thresholding of the support values in the proposed method is very competitive with techniques based on wavelets,
including thresholding of decimated or undecimated wavelet transforms.
In many spatial analyses and visualizations related to terrain, a high resolution and accurate digital surface model (DSM)
is essential. To develop a robust interpolation and smoothing solutions for airborne light detection and ranging (LIDAR)
point clouds, we introduce the weighted adaptive mapping LS-SVM to fit the LIDAR data. The SVM and the weighted
LS-SVM are introduced to generate DSM for the sub-region in the original LIDAR data, and the generated DSM for this
region is optimized using the points located within this region and additional points from its neighborhood. The fitting
results are adaptively optimized by the local standard deviation and the global standard deviation, which decide whether
the SVM or the weighted LS-SVM is applied to fit the sub-region. The smooth fitting results on synthesis and actual
LIDAR data set demonstrate that the proposed smooth fitting method is superior to the standard SVM and the weighted
LS-SVM in robustness and accuracy.
Image fusion is an important tool in remote sensing, since many Earth observation satellites provide both high-resolution panchromatic (Pan) and low-resolution multispectral (MS) images. To date, many image fusion techniques have been developed. However, the available algorithms can hardly produce a satisfactory fusion result for IKONOS and QuickBird images. Among the existing fusion algorithms, the IHS technique is the most widely used one, and the wavelet fusion is the most frequently discussed one in recent publications because of its advantages over other fusion techniques. But the color distortion of these two techniques is often obvious. The support value fusion technique demonstrates some advantages over the conventional methods. This study presents a new fusion approach that integrates the advantages of both the IHS and the support value techniques to reduce the color distortion of QuickBird fusion results. Different QuickBird images have been fused with this new approach. Visual and statistical analyses prove that the concept of the proposed extended fast IHS (eFIHS) and support value integration is promising, and it does significantly improve the fusion quality compared to conventional IHS (eFIHS) and wavelet fusion techniques.
Airborne light detection and ranging (LIDAR) data filtering is the most time-consuming and expensive part in applications related to laser scanning. This paper proposed a fast facet-based LIDAR data filtering method. LIDAR point clouds are interpolated onto a regular grid, and the filtering of nonground points is implemented on the grid-based data. The simple, quadratic, and cubic facet models, which are respectively, based on the zero, second, and third orders of orthogonal polynomials, are used to estimate the underlying elevation surface trend, which is considered as the approximation of ground surface. As the ground measurements are generally below the objects, the nonground points are filtered by removing the points that are higher than the estimated elevation surface trend. The resulting holes are filled with the nearest remaining measurements. Iteratively filtering in this way, the estimated elevation surface trend converges at the real ground surface. The nonground points that are higher than the finally approximated ground surface are filtered and the ground points are extracted from the LIDAR data. Experimental results on the test data released from the International Society for Photogrammetry and Remote Sensing (ISPRS) demonstrate that the proposed approach is efficient and provides at least comparable performance with the accuracy reports published by ISPRS.
Automatic feature extraction for road information plays a central role in applications related to terrains. In this paper, we propose a new road extraction method using the one-class support vector machine (SVM). For a manually segmented seed road region, only a part of pixels are really road, some pixels locating on the sideway, shadows of the building, and the cars etc., are not really road pixels. The one-class SVM is used to estimate a decision function that takes the value +1 in a small feature region capturing most of the data points in the seed road area, and -1 elsewhere. Since the road pixels in the satellite image have the similar properties, such as the spectral feature in multi-spectral image, the novelty pixel is discriminated by the estimated decision function for road segmentation. Many computation experiments are undertaken on the IKONOS high resolution image. The results demonstrate that the proposed method is effective and has much higher computation efficiency than the standard pixel-based SVM classification method.
A novel method for fusion detection of small infrared targets based on support vector machines (SVM) in the wavelet domain is presented. Target detection task plays an important role in automatic target recognition (ATR) systems because overall ATR performance depends closely on detection results. SVM is a powerful methodology for solving problems in nonlinear classification, function estimation and density estimation. Least-squares support vector machines (LS-SVMs) are reformulations to standard SVMs. The proposed algorithm can be divided into four steps. First, each frame of the image sequence is decomposed by the discrete wavelet frame (DWF). Second, the components with low frequency are performed by regression based on LS-SVM. The one-order partial derivatives in row and column directions are derived. Therefore, feature images of the gradient strength can be obtained. Third, feature images of five consecutive frames are fused to accumulate the energy of target of interest and greatly reduce false alarms. Finally, the segmentation method based on contrast between target and background is utilized to extract the target. In terms of connectivity of moving targets, the majority of residual clutter and false alarms that survive are removed based on 3-D morphological dilation across three consecutive frames along the motion direction of the moving targets. Actual infrared image sequences in backgrounds of sea and sky are applied to validate the proposed approach. Experimental results demonstrate the robustness of the proposed method with high performance.
Detection of dim moving small targets at low signal noise ratio is a very important issue and difficult problem in infrared searching and tracking system. Based on analysis of the character of infrared images, a new double energy accumulating method is proposed. Firstly, images are denoised by wavelet transformation with soft threshold. Then, object motion area is detected according to difference images and the target intensity is well enhanced by accumulating energy two times with addition and product operation. Finally, target candidates are separated from background by thresholding process with the selected threshold. Computer experiments are carried out with an infrared image sequence and the experimental results illustrate that the proposed method is effective and efficient.
Based on statistical learning theory, support vector machine (SVM) is a novel type of learning machine, and it contains polynomial, neural network and radial basis function (RBF) as special cases. The mapped least squares support vector machine (MLS-SVM) is a special least square SVM (LS-SVM), which extends the application of the SVM to the image processing. Based on the MLS-SVM, a family of filters for the approximation of partial derivatives of the digital image surface is designed. Prior information (e.g., local dominant orientation) are incorporated in a two dimension weighted function. The weighted MLS-SVM with the radial basis function kernel is applied to design the proposed filters. Exemplary application of the proposed filters to fingerprint image segmentation is also presented.
Star acquisition is one of the most time-consuming routines in star-tracker operation. In the star image, a star point spread function (PSF) represents a near-Gaussian distribution. The star extraction consists in finding the highest-intensity pixel among the PSFs, collecting the adjacent pixels, and then calculating the star centroids in the star image plane. The candidate highest-intensity pixels are the maximum extremum points of the underlying intensity function of a digital star image. To extract star from the star image, the cubic facet model is applied to fit the underlying intensity surface in star acquisition procedure. A new extraction approach, using surface-fitting methods to approximate locally the image intensity function, and then using the partial derivatives of the fitted surface to make decisions regarding the maximum extremum points, is proposed. A number of experiments are carried out on simulated star images. The experimental results demonstrate that the proposed method is efficient and robust.
We describe a novel interpolation algorithm to find the optimal image intensity function generating an optimal gray-level estimation of interpolated pixels of digital images. The new approach is based on the proposed image block mapping method and least-square support vector machines (LSSVM) with Gaussian radial basis function (RBF) kernels. With the mapping technique, the interpolation procedure of the LSSVM is actually accomplished in the same input vector space. A number of different scale interpolation experiments are carried out. The experimental results demonstrate that the performance of the proposed algorithm is competitive with many other existing methods, such as cubic, spline, and linear methods. The peak signal-to-noise ratio of the image reconstructed by the proposed algorithm is higher than those obtained by the spline. And the estimated accuracy of the proposed algorithm is similar to that of the cubic algorithm, while the computational requirement is lower than the latter.
A new general method of the automatic selection of guide star, which based on a new dynamic Visual Magnitude Threshold (VMT) hyper-plane and the Support Vector Machines (SVM), is introduced. The high dimensional nonlinear VMT plane can be easily obtained by using the SVM, then the guide star sets are generated by the SVM classifier. The experiment results demonstrate that the catalog obtained by the proposed algorithm has a lot of advantages including, fewer total numbers, smaller catalog size and better distribution uniformity.
In this paper, an efficient VLSI architecture for biorthogonal 9/7 wavelet transform by lifting scheme is presented. The proposed architecture has many advantages including, symmetrical forward and inverse wavelet transform as a result of adopting pipeline parallel technique, as well as area and power efficient because of the decrease in the amount of memory required together with the reduction in the number of read/write accesses on account of using embedded boundary data-extension technique. We have developed a behavioral Verilog HDL model of the proposed architecture, which simulation results match exactly that of the Matlab code simulations. The design has been synthesized into XILINX xcv50e-cs144-8, and the estimated frequency is 100MHz.