No SPIE Account? Create one
Editor Affiliations +
Tianxu Zhang,1 Nong Sang1
1Huazhong Univ. of Science and Technology (China)
1Huazhong Univ. of Science and Technology (China)
Visit My Account to manage your email alerts.
This will count as one of your downloads.
You will have access to both the presentation and article (if available).
This will count as one of your downloads.
You will have access to both the presentation and article (if available).
This PDF file contains the front matter associated with SPIE Proceedings Volume 8918, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.
Visual tracking algorithms based on online boosting generally use a rectangular bounding box to
represent the position of the target, while actually the shape of the target is always irregular. This
will cause the classifier to learn the features of the non-target parts in the rectangle region, thereby
the performance of the classifier is reduced, and drift would happen. To avoid the limitations of the
bounding-box, we propose a novel tracking-by-detection algorithm involving the level set
segmentation, which ensures the classifier only learn the features of the real target area in the
tracking box. Because the shape of the target only changes a little between two adjacent frames and
the current level set algorithm can avoid the re-initialization of the signed distance function, it only
takes a few iterations to converge to the position of the target contour in the next frame. We also
make some improvement on the level set energy function so that the zero level set would have less
possible to converge to the false contour. In addition, we use gradient boost to improve the original
multi-instance learning (MIL) algorithm like the WMILtracker, which greatly speed up the tracker.
Our algorithm outperforms the original MILtracker both on speed and precision. Compared with the
WMILtracker, our algorithm runs at a almost same speed, but we can avoid the drift caused by
background learning, so the precision is better.
Conventional methods often assume that water region is homogeneous and bridge is brighter than background. They usually recognize target by parallel lines detection. But grayscale of bridge has bipolar problem in FLIR images due to interference of complex background and constraints of imaging conditions, which means that it can be greater or lower than river. Furthermore, water is not a homogeneous area as a whole because of the interference of water clutter and shoals. This paper proposes a novel algorithm of bridge recognition based on Gabor filter. Firstly, we obtain target ROI by extracting the horizontal line. And then ROI sub-images are enhanced by Gabor filter and target polarity is determined by bridge body detection. Finally, bridge recognition can be achieved by pier detection according to the target polarity and location of bridge body. Experimental results of nearly 3000 frames show that the proposed algorithm can effectively overcome problems such as bipolar target and low image contrast. It offers a good practicability and accuracy in bridge recognition in FLIR images.
This paper presents a target detection method in synthetic aperture radar (SAR) images with radiometric multiresolution analysis (RMA). The idea is that target saliency can be efficiently computed by comparing the statistics of targets and those of the local background around them. In order to compute reliable statistics of targets, which usually involve a small number of pixels, RMA is adopted. The RMA preprocessing method performs well in stabilizing the statistical characteristics of SAR images. It can effectively restrain the speckle noise while keep the statistical characteristics of the original image. Based on the computed target saliency, adaptive decision thresholds are got by using the constant false alarm rate (CFAR) target detection framework. Our experiments on real SAR images show that the proposed method can achieve better performance compared with the traditional cell average-constant false alarm rate (CA-CFAR) method.
We propose a novel robust texture descriptor, the Gradient Orientation and Modulus Matrix (GOMM), which is based on the fact that human perception of an image pattern depends not only on its intensity, but also on changes in intensity and regularity (such as the gradient modulus and gradient orientation of the image).
A GOMM is constructed in three steps. First, the gradient orientation of each pixel is mapped onto N intervals and the gradient modulus is partitioned into M levels. Next, a block is constructed from the gradient modulus of pixels whose gradient orientations are mapped onto the same interval. Then, each component of the GOMM is given by the sum of the ratios between two terms, namely, the differential gradient modulus grading between the elements in the above-mentioned block, and the distance between the elements. Finally, a six-dimension vector is calculated from each GOMM. By rearranging feature vectors from each GOMM, we can concatenate the vectors to construct a uniform GOMM feature for a given image, irrespective of the angle of the image. Experimental results on the KTH-TIPS2 (The Royal Institute of Technology - Textures under varying Illumination, Pose and Scale) image database show that the GOMM significantly outperforms the other classical descriptors.
A GOMM is constructed in three steps. First, the gradient orientation of each pixel is mapped onto N intervals and the gradient modulus is partitioned into M levels. Next, a block is constructed from the gradient modulus of pixels whose gradient orientations are mapped onto the same interval. Then, each component of the GOMM is given by the sum of the ratios between two terms, namely, the differential gradient modulus grading between the elements in the above-mentioned block, and the distance between the elements. Finally, a six-dimension vector is calculated from each GOMM. By rearranging feature vectors from each GOMM, we can concatenate the vectors to construct a uniform GOMM feature for a given image, irrespective of the angle of the image. Experimental results on the KTH-TIPS2 (The Royal Institute of Technology - Textures under varying Illumination, Pose and Scale) image database show that the GOMM significantly outperforms the other classical descriptors.
In this paper, we propose a specularity-invariant crop extraction method using probabilistic super-pixel markov random field (MRF). Our method is based on the underlying rule that intensity change gradually between highlight areas and its neighboring non-highlight areas. This prior knowledge is embedded into the MRF-MAP framework by modeling the local and mutual evidences of nodes. The marginal probability of each node in the label field is then iteratively computed by Belief Propagation algorithm which leads to the final solution. Comparing experimental results show that our method outperforms the other commonly used extraction methods in yielding highest performance with the lowest standard deviation.
With regard to target detection in complex background in high resolution image sequences attained by Wide Field of View Infrared Surveillance System, a rough-to-meticulous real-time target detection algorithm is proposed. In the rough detection phase, it attains initial high rate target detection by background matching and frame difference algorithm, based on the gray high frequency and moving characteristics of the target in the wide field of view image. In the meticulous recognition phase, focusing on the detected suspected target sliced images, it has further delicate recognition on the basis of targets’ characteristics to exclude those false jamming. The detection result of the test images shows, the algorithm enables stable detection with low-rate false alarm for distant dim targets, and has been applied to the signal processing of the Wide Field of View Infrared Surveillance System.
Aiming at the problem of tracking 3D target in forward-looking infrared (FLIR) image, this paper proposes a high-accuracy robust tracking algorithm based on SIFT and particle filter. The main contribution of this paper is the proposal of a new method of estimating the affine transformation matrix parameters based on Monte Carlo methods of particle filter. At first, we extract SIFT features on infrared image, and calculate the initial affine transformation matrix with optimal candidate key points. Then we take affine transformation parameters as particles, and use SIR (Sequential Importance Resampling) particle filter to estimate the best position, thus implementing our algorithm. The experiments demonstrate that our algorithm proves to be robust with high accuracy.
Aiming at the problem of in-harbor ship detection in forward-looking infrared image, this paper proposes a
method for ship segmentation and false alarm suppressing based on k-means clustering segmentation. We obtain the
simulated model images from visible satellite images and perspective relations. And the harbor area is determined by matching with HOG features. Then we segment the ship out of the harbor area. In order to suppress the false alarm, we apply k-means clustering segmentation to get the ship and the sea area simultaneously. By calculating the external convex polygon, we get rid of the false alarm targets. Experimental results suggest that our method has high detection accuracies and low false alarm rate.
We present a novel contour-based object detector using generalized hough transform where each local part casts a vote for the possible locations of the object center. The angles of line segments as local feature are extracted to describe the contour of the object, then an improved voting tactics is applied to detect the location and attitude of the object. Experimental results demonstrate that the algorithm has an encouraging detection performance.
A novel unsupervised color image segmentation method based on graph cuts with multi-components is proposed, which finds an optimal segmentation of an image by regarding it as an energy minimization problem. First, L*a*b* color space is chosen as color feature, and the multi-scale quaternion Gabor filter is employed to extract texture feature of the given image. Then, the segmentation is formulated in terms of energy minimization with an iterative process based on graph cuts, and the connected regions in each segment are considered as the components of the segment in each iteration. In addition, canny edge detector combined with color gradient is used to remove weak edges in segmentation results with the proposed algorithm. In contrast to previous algorithms, our method could greatly reduce computational complexity during inference procedure by graph cuts. Experimental results demonstrate the promising performance of the proposed method.
Airport runway recognition is of great significance in fields like remote sensing, navigation and traffic monitoring. An airport runway recognition method using the “hypothesize-and-verify” paradigm is proposed. Firstly, local line segments of runway contour are extracted in complex infrared image. Secondly, basing on a new Line Segment Hough Transform, local line segments vote fuzzily in the parameter space to obtain global line segment clustering, and then parallel straight lines are extracted on the basis of parameter space to form hypotheses of potential airport runways. Finally, using contextual information of airport constructions, hypotheses disambiguation and verification of runway is accomplished primarily by extraction of runway markings and segmentation of transportation network, i.e. taxiways and apron. Experimental results demonstrate the good performance of our method on a variety of complex scenes.
For building detection from single very high spatial resolution (VHR) satellite images, we take advantage of visual saliency and Bayesian model to rapidly locate roof-top areas. We firstly generate saliency map of an image by a salient contrast filter using low-level feature. This filter distinguishes salient pixels if a pixel is visually different from its surroundings in color or texture. Secondly, a Bayesian model is proposed to generate all closed rectangular contours as mid-level content in the image. We suggest the area enclosed by contour corresponds to high saliency values. Finally, the roof-top areas are extracted by fusing different level information mentioned above. Experimental results demonstrate the effectiveness of our algorithm.
In this paper, a novel difficult prediction scheme for infrared building target recognition is developed. Our scheme can predict the difficulty of recognizing a designated target in advance, which is desirable in infrared building recognition. The experiment results show that our scheme is efficient to fulfill the prediction task and the prediction is consistent with the real recognition results.
The automatic observation of the field crop attracts more and more attention recently. The use of image processing technology instead of the existing manual observation method can observe timely and manage consistently. It is the basis that extracting the wheat from the field wheat images. In order to improve accuracy of the wheat segmentation, a novel two-stage wheat image segmentation method is proposed. Training stage adjusts several key thresholds which will be used in segmentation stage to achieve the best segmentation results, and counts these thresholds. Segmentation stage compares the different values of color index to determine which class of each pixel is. To verify the superiority of the proposed algorithm, we compared our method with other crop segmentation methods. Experiment results shows that the proposed method has the best performance.
Nowadays ground vehicle detection on airborne platforms is becoming very important for intelligent visual surveillance applications. Object detection using cascade structured classifiers is booming fast in recent decade, and very successful in real-time applications. However, most of them apply a sliding window on multi-scaled images which commonly need heavy computational expense, therefore, are only suitable for using simple features. In this paper, a biologically inspired object detection algorithm is proposed, which exploits image patch based feature learning and visual saliency detection. The image patch based local features are learnt by unsupervised learning to generate an object category specific visual dictionary. Visual saliency detection is performed to extract candidate object regions from a whole image using the learnt local features. Instead of a sliding window, a candidate object region is sent to an object classifier only when its features are salient on the whole image. Since the number of candidate object regions decreases dramatically, it allows to utilize much complex features to represent object images so that it can increase the descriptive capability of the learnt features. The experimental results on practical vehicle image datasets indicate that less computational expense and good detection performance can be achieved.
While Graph Cuts are used for image segmentation, there exist two problems: how to get better initial information of
foreground and background and how to improve the executing efficiency of Graph Cuts algorithm. To solve the first
problem, path morphology and line segment matching algorithm are performed to get initial background information at
the same time as getting initial foreground information, so non-road pixels similar with road pixels in gray value or
texture are avoided being segmented as road points. To cope with the second problem, push-relabel strategy is chosen
and its parallelized version based on NVIDIA CUDA platform is performed in this paper. Our strategy is performed on
dense built-up area and suburban district and proved to be effective in both accuracy and efficiency.
Motion detection plays an important role in intelligent video surveillance. This paper introduces a particular background subtraction technique called ViBe. This technique updates the background model randomly and it has established model with fast, high precision and fast processing speed. ViBe algorithm provides the method of updated background model, but slowly eliminates ghost region. This paper presents an improved ViBe algorithm based on region motion classification. The algorithm considers the difference of the movement directions of feature points in foreground regions on adjacent frame, and defines a criterion function to evaluate the difference so that can quickly eliminate ghost regions. Experimental results show that the proposed algorithm quickly remove the ghost region and improve the detection accuracy.
The Hough transform is a robust tool to extract features, such as straight edges, circles, or ellipses from images and describe them parametrically. However, linear edge in real image is not an ideal line. Tiny orientation change commonly exists in linear edge obtained by edge detector. Standard Hough transform need to discrete Hough space, discreteness must lead to statistical error, which makes it difficult to extract the tiny, short and small line when detecting long line. It makes detecting failure of the tiny line or cannot contain all information of these small lines. In this paper, proposed improved Hough transform can eliminate the statistical error caused by discreteness. Consequently eliminate the bad effect of detecting short and tiny line caused by detecting long line, it make the long line and the short, tiny line can be detected precisely at the same time.
As a method of representing the test sample with few training samples from an overcomplete dictionary, sparse representation has attracted much attention in target classification recently. In this paper, we develop a new SAR vehicle classification method based on sparse representation, in which the correlation between the vehicle’s aspect angle and the sparse representation coefficients is exploited. The detail procedure presented in this paper can be summarized as follows. Initially, the sparse coefficient vector of a test sample is solved by sparse representation algorithm with a pixel based dictionary. Then the coefficient vector is projected onto a sparser one with the constraint of vehicle’s aspect angle. Finally, the vehicle is classified to a certain category that minimizes the reconstruct error with the sparse coefficient vector. We present promising results of applying the proposed method to the MSTAR dataset.
In planetary or lunar landing missions, hazard avoidance is critical for landing safety. Therefore, it is very important to correctly detect hazards and effectively find a safe landing area during the last stage of descent.
In this paper, we propose a passive sensing based HDA (hazard detection and avoidance) approach via descent images to lower the landing risk. In hazard detection stage, a statistical probability model on the basis of the hazard similarity is adopted to evaluate the image and detect hazardous areas, so that a binary hazard image can be generated. Afterwards, a safety coefficient, which jointly utilized the proportion of hazards in the local region and the inside hazard distribution, is proposed to find potential regions with less hazards in the binary hazard image. By using the safety coefficient in a coarse-to-fine procedure and combining it with the local ISD (intensity standard deviation) measure, the safe landing area is determined. The algorithm is evaluated and verified with many simulated descent downward looking images rendered from lunar orbital satellite images.
In this paper, we propose a passive sensing based HDA (hazard detection and avoidance) approach via descent images to lower the landing risk. In hazard detection stage, a statistical probability model on the basis of the hazard similarity is adopted to evaluate the image and detect hazardous areas, so that a binary hazard image can be generated. Afterwards, a safety coefficient, which jointly utilized the proportion of hazards in the local region and the inside hazard distribution, is proposed to find potential regions with less hazards in the binary hazard image. By using the safety coefficient in a coarse-to-fine procedure and combining it with the local ISD (intensity standard deviation) measure, the safe landing area is determined. The algorithm is evaluated and verified with many simulated descent downward looking images rendered from lunar orbital satellite images.
In this paper a method of Fourier spectrum features based edge detection of urban street trees is described. The QuickBird image was first transformed by 2-D discrete Fourier transform. Then the energy of the component in spatial frequency was calculated. The energy distribution of the angle in max energy was used for further study. Different frequency segments was analyzed, the frequency that can best describe the street tree edge was chosen as the cut-off frequency of the street trees edge. Odd Gabor filter in frequency domain with the cut-off frequency and the max-energy angle was applied for the edge detection. The road center line is extracted by a Gabor filter in frequency domain. Then the edge of the street trees is restricted by the road center line. The edge detection result is analyzed by Canny criteria, and the ΣV=1.00, and C=0.89.
In this paper, we propose a novel method to estimate the camera’s ego-motion parameters by directly using the normal flows. Normal flows, the projection of the optical flows along the direction of the gradient of image intensity, could be calculated directly from the image sequence without any artificial assumptions about the captured scene. Different from many traditional approaches which tackle the problem by establishing motion correspondences or by estimating optical flows, our proposed method could obtain the motion parameters directly by using the information of spatio-temporal gradient of the image intensity. Hence, our method requires no specific assumptions about the captured scene, such as the smoothness constraint, continuity constraint, distinct features appearing in the scene and etc.. Our method has been experimentally tested by using both synthetic image data and real image sequences. The experimental results demonstrate that our proposed method is feasible and reliable.
This paper presents a method to segment moving human bodies. A self-adaptive background model is used to update the background image(so-called reference image). By calculating the Euclidean distance of corresponding points in the current and background image, we can check out the foreground objects. And the shadow can be detected and removed according to the characteristics of the shadow regions shown in HSV space. Finally, target tracking is implemented by calculating the relativity of color histogram between the moving areas in two succeeding images.
Outside the classical receptive field (CRF), there exists a broad non-classical receptive field (NCRF). The response of
the central neuron is affected not only by the stimulus inside the CRF, but also modulated by the stimulus surrounding it.
The contextual modulation is mediated by horizontal connections across the visual cortex. In this paper, a contour
detection method inspired by the visual mechanism in the primary visual cortex (V1) is proposed. The method is divided
in three steps. Firstly, the response of every single visual neuron in V1 is computed by local energy. Secondly, the
facilitation and suppression (the contextual influence) on a neuron through horizontal interactions are obtained by
constructing a two neighbor modulating functions. Finally, the total output response of one neuron to complex visual
stimuli is acquired by combing the influence of local visual context on the neuron and energy response by itself. We
tested it on natural image and encouraging results were acquired.
In this paper, an algorithm of image segmentation using region-based MRF combined with boundary information is
proposed. Firstly we obtain the initial over segmentation regions by Meanshift (MS) algorithm and the regions that
include seed points are set to seed regions. Then, starting from the seed regions, the finally result is detected under the
framework of MRF, where an image model is built for the potential function of the regional MRF image segmentation
and combined with edge strength to define a suitable MRF potential function, which is based on the similarity criterion
of the value of the edge. The experimental results indicate that the algorithm can promote the adaptive capacity of the
MRF model and reserve more details as well as region homogeneous.
Precise segmentation is crucial for the feature extraction and classification of ships in SAR imagery. To alleviate the Doppler shift and the cross ambiguity, this paper propose to segment the ship area from its background based on the radon transform. Assuming that the region of interest (ROI) of ship in SAR imagery has been extracted, the detail procedures of the proposed refined segmentation can be summarized as follows. First, the ship’s ROI image is transformed to radon domain, in which pixel intensities are cumulated along different directions. Then, the peak areas are separated to extract the ship’s orientation and the main image area of the ship that orthogonal to the principal axis. Finally, the refined segmentation is achieved in the main image area. Experiments, accomplished over measured medium and high resolution SAR ship images, show the effectiveness of the proposed approach.
Automatic target detection in infrared images is a hot research field of national defense technology. We propose a new saliency-based infrared target detection model in this paper, which is based on the fact that human focus of attention is directed towards the relevant target to interpret the most promising information. For a given image, the convolution of the image log amplitude spectrum with a low-pass Gaussian kernel of an appropriate scale is equivalent to an image saliency detector in the frequency domain. At the same time, orientation and shape features extracted are combined into a saliency map in the spatial domain. Our proposed model decides salient targets based on a final saliency map, which is generated by integration of the saliency maps in the frequency and spatial domain. At last, the size of each salient target is obtained by maximizing entropy of the final saliency map. Experimental results show that the proposed model can highlight both small and large salient regions in infrared image, as well as inhibit repeated distractors in cluttered image. In addition, its detecting efficiency has improved significantly.
Passive millimeter waves (PMMW) image can create interpretable imagery on the objects concealed under clothing, which gives the great advantage to the security system. In conventional detection methods, the object detection methods can be roughly divided into two categories: the edge-based method and the region-based method. In this paper, we propose to combine the two methods for better detecting the concealed object for PMMW imaging. The main idea of the proposed method is to combine the edge-based contrast and region-based center-surround histogram. The proposed method can describe a concealed object locally and regionally, which help us capture more useful information about the edge and region. Experimental results on real images demonstrate that the proposed method can effectively detect the concealed object in the PMMW images.
Selecting key point in airplane target as tracking aimpoint at end term is important for IR imaging missiles to improve guidance accuracy. A new aimpoint selection method proper for engineering application is proposed in this article. Other than tracking the center of plume which is the most marked property of airplanes, some point near engine is selected as aimpoint. Firstly plume and skin are extracted by using different thresholds according to their gray scale statistics and features like circularity, distance ratio and central axis are obtained to classify the image types. Then referring to these image types, the centroid of the segmented sector or a point on the line of central axis of plume sector are selected as aimpoint respectively. The algorithm has the advantage of more efficiency in both space and time consuming. Tests have shown the validity of the algorithm.
In this paper, a new procedure based on least trimmed square for clutter background estimation is proposed. Least trimmed square method identifies multiple outliers in the image, such as noise and target region. Then the clutter background is estimated without these outliers. The performance of this method is compared with the algorithms based on least mean square method, the results show that our method gets higher signal clutter ratio (SCR) gain in target region than other methods which use LMS filter.
Image-spectrum integrated instrument is an infrared scanning system which integrates optics, mechanics, electrics and information processing. Not only can it achieve scene imaging, but also it can detect, track and identify targets of interests in the scene through acquiring their spectra. After having a brief introduction to image-spectrum integrated instrument and analyzing how 2D scanning mirror works, this paper built 3D model of 2D scanning mirror and simulated its motion using two PCs basing on VC++ and ACIS/HOOPS. Two PCs communicate with each other through serial ports. One PC serves as host computer, on which controlling software runs, is responsible for loading image sequence, image processing, target detecting, and generating and sending motion commands to scanning mirror. The other serves as slave computer, on which scanning mirror motion simulation software runs, is responsible for receiving motion commands to control scanning mirror to finish corresponding movements. This method proposed in this paper adopted semi-physical virtual prototype technology and used real scene image sequence to control virtual 2D scanning mirror and simulates motion of real 2D scanning mirror. It has no need for real scanning mirror and is of important practical significance for debugging controlling software of 2D scanning mirror.
This paper focuses on the detection of citrus fruits in the tree canopy under variable illumination and different degree occlusion. We applied a novel segmentation method to detect the visible parts of fruits by fusing the segmentation results of chromatic aberration map, normalized RGB model, and illumination map. This fusion method can detect the highlights, shadows and diffuse zones of fruit targets. The 3-D surface topography of the visible parts of fruits were recovered by the classical algorithm of shade from shading, the fruit targets were recovered by sphere fitting using these point cloud data, and the valid ones were chosen out by validity check. The results showed that the occlusion zones of targets were effectively recovered under various light conditions integrally using the proposed method.
The automatic detection of visually salient information from abundant video imagery is crucial, as it plays an important role in surveillance and reconnaissance tasks for Unmanned Aerial Vehicle (UAV). A real-time approach for the detection of salient objects on road, e.g. stationary and moving vehicle or people, is proposed, which is based on region segmentation and saliency detection within related domains. Generally, the traditional method specifically depends upon additional scene information and auxiliary thermal or IR sensing for secondary confirmation. However, this proposed approach can detect the interesting objects directly from video imagery captured by optical camera fixed on the small level UAV platform. To validate this proposed salient object detection approach, the 25 Hz video data from our low speed small UAV are tested. The results have demonstrated the proposed approach performs excellently in isolated rural environments.
In this paper, we address the problem of object recognition in clutter with incomplete features, focusing on challenging small target. Since the discriminative illumination, occluded or imperfect of the feature extraction algorithm, multi-class features of the object are incomplete which increase the false alarm rate of object recognition. How to use the incomplete features such as region and edge for object recognition, the vote of each fragment region contributes to the recognition result according to its reliability. We cast it as a graph search problem and propose a novel evidence accumulation algorithm to efficiently solve it.
Aerial surveillance is a main functionality of UAV, which is realized via video camera. During the operations, the mission assigned targets always are the kinetic objects, such as people or vehicles. Therefore, object tracking is taken as the key techniques for UAV sensor payload. Two difficulties for UAV object tracking are dynamic background and hardly predicting target’s motion. To solve the problems, it employed the particle filter in the research. Modeling the target by its characteristics, for instance, color features, it approximates the possibility density of target state with weighting sample sets, and the state vector contains position, motion vector and region parameters. The experiments demonstrate the effectiveness and robustness of the proposed method in UAV video tracking.
Automatic target recognition (ATR) usually become difficult when target is blocked by clouds, or low image contrast, or
target repeat mode in a complex optical imaging environment. The ground target recognition method based on landmarks
is a good way for aircraft navigation, which can solve unobvious target recognition problems in a complex wide-field
scene. In combination with the characteristics of selective attention in human visual system, this paper systematic study
the construction rules for the cluster of landmarks, present a landmarks dynamic allocation method in ground target
recognition, which can effectively improve the stability and accuracy of target recognition.
Wide line detection plays an important role in image analysis and computer vision. However, most of the existing algorithms focus on the extraction of the line positions and length, ignoring line thickness and direction which can deepen our understanding of images. This paper presents a novel wide line detector using the ridge distribution feature and layer growth method. Unlike most existing edge and line detectors which use directional derivatives, our proposed method extracts the ridge target point and use the layer growth to find the line completely based on the isotropic nonlinear filter. Ridge points are detected by its distribution symmetry based on the isotropic responses via circular masks and orientation of the ridge is determined roughly. The ridge point is selected as a seed point, then growth layer by layer, to determine the width and orientation of the curvilinear structure accurately. Instead of point by point scanning, we label points in the growth region and adjust the scanning step adaptively which improve the method efficiently. The proposed method can detect the accurate width and direction of lines dynamically. This can provide great convenience for post-processing or for application requirements. A sequence of tests on a variety of image samples demonstrates that the proposed method outperforms state-of-the-art methods.
Subject to limited resolution for targets in many satellite images, low-resolution airplane detection is still difficult and challenging, which plays an important role in remote sensing. In this paper, we propose a new method to detect lowresolution airplanes in satellite images. First, the image is preprocessed by combing the unsharp contrast enhancement (UCE) filtered image and the original image. Second, the Local Edge Distribution (LED), which is susceptible to objects owning clustered edges, e.g., airplane, is calculated to acquire the target candidate regions while restraining large background area. Then, a multi-scale fused gradient feature image is computed to characterize the shapes of targets instead of the original image to overcome the influence from the self-shadow and different coating colors of airplanes. After that, a designed airplane shape filter with a modulated item is used to detect and locate real targets, in which the modulated item can effectively measure the degree of coincidence between the patch region and the airplane shape. Finally, coordinates of target centers are computed in the filtered image. Experimental results demonstrate that the proposed algorithm is effective and robust for detecting low-resolution airplanes in satellite images under various complex backgrounds.
Star image blurred by aircraft vibration decreases location accuracy and probability of the star extraction. In this paper,
first, the influence of aircraft vibration on the star image captured by star sensors is analyzed, and the mathematical
model is deduced and established. Then, in order to overcome the adverse effects of star extraction and stabilize the
accuracy of star sensor in high dynamic environment, a restoration method for blurred star image using Richardson-Lucy
(RL) method is introduced. The experimental results indicate that the proposed method can effectively improve the star
image signal-to-noise ratio and the extraction accuracy.
In order to extract precession frequency, an crucial parameter in ballistic target recognition, which reflected the kinematical characteristics as well as structural and mass distribution features, we developed a dynamic RCS signal model for a conical ballistic missile warhead, with a log-norm multiplicative noise, substituting the familiar additive noise, derived formulas of micro-Doppler induced by precession motion, and analyzed
time-varying micro-Doppler features utilizing time-frequency transforms, extracted precession frequency by measuring the spectrogram’s texture, verified them by computer simulation studies. Simulation demonstrates the excellent performance of the method proposed in extracting the precession frequency, especially in the case of low SNR.
time-varying micro-Doppler features utilizing time-frequency transforms, extracted precession frequency by measuring the spectrogram’s texture, verified them by computer simulation studies. Simulation demonstrates the excellent performance of the method proposed in extracting the precession frequency, especially in the case of low SNR.
In this paper, a new automatic and adaptive aircraft target detection algorithm in high-resolution airport synthetic aperture radar (SAR) images is proposed. Firstly, region segmentation is used to detect the apron area in the images, which provides the potential area where aircrafts may exist and reduce the search range. Secondly, upon the apron area the pre-segmentation is taken to label the possible target points. Thirdly, the constant false alarm rate (CFAR) detector is improved to cope with multi-target detection situation. The clutter pixels in the sliding detection window will be removed automatically based on pre-segmentation result. As a result, more structural features of the targets are preserved. At last, in order to eliminate the detected false targets and solve the problem that the same target is divided into several disconnected areas, a new joint algorithm based on the area recognition factors and distance cluster is presented. The real airborne SAR image data of some airport is used to verify this target detection algorithm, and the result indicates that this algorithm can detect the aircraft target precisely and decrease the false alarm rate.
An efficient automatic small target detection algorithm in infrared image is proposed. Based on non-linear histogram equalization, a coarse-to-fine segmentation is used to segment IR image into target candidates.
Then genuine targets are captured by using contrast-based confidence measure and empirical size constraint. Experimental results demonstrate that the presented method is efficient, accurate and robust.
Then genuine targets are captured by using contrast-based confidence measure and empirical size constraint. Experimental results demonstrate that the presented method is efficient, accurate and robust.
Considering antenna pattern and Doppler frequency-shift separately, the theory model about targets near-field scattering is established from a simple target, using PO (physical optics) method depending on SCTE(Scattering from Complex Targets and Environments)system[1].Then some curves are presented which are in accordance with the results given by references. The calculated results show that the influence of the factors over targets near-field RCS are complex and important. This work has practical engineering value in the modern and high science and technology warfare.
This letter mainly aims at an E-Centrist descriptor for the pedestrian recognition in image sequences with background
moving slowly. Utilizing the motion information detected from the image sequences, pedestrian recognition algorithm is
implemented by combining region of interest(ROI)which probably includes potential pedestrians and an enhanced
descriptor from contour. Experimental results demonstrate that the presented method improves the speed as well as the
accuracy of pedestrian recognition in test sequences.
Facing challenges of nontraditional geometry, multiple resolutions and the same features sensed from different angles, there are more difficulties of robust correspondence matching for ground images along the optic axis.
A method combining SIFT algorithm and the geometric constraint of the ratio of coordinate differences between image point and image principal point is proposed in this paper. As it can provide robust matching across a substantial range of affine distortion addition of change in 3D viewpoint and noise, we use SIFT algorithm to tackle the problem of image distortion. By analyzing the nontraditional geometry of ground image along the optic axis, this paper derivates that for one correspondence pair, the ratio of distances between image point and image principal point in an image pair should be a value not far from 1. Therefore, a geometric constraint for gross points detection is formed. The proposed approach is tested with real image data acquired by Kodak. The results show that with SIFT and the proposed geometric constraint, the robustness of correspondence matching on the ground images along the optic axis can be effectively improved, and thus prove the validity of the proposed algorithm.
A method combining SIFT algorithm and the geometric constraint of the ratio of coordinate differences between image point and image principal point is proposed in this paper. As it can provide robust matching across a substantial range of affine distortion addition of change in 3D viewpoint and noise, we use SIFT algorithm to tackle the problem of image distortion. By analyzing the nontraditional geometry of ground image along the optic axis, this paper derivates that for one correspondence pair, the ratio of distances between image point and image principal point in an image pair should be a value not far from 1. Therefore, a geometric constraint for gross points detection is formed. The proposed approach is tested with real image data acquired by Kodak. The results show that with SIFT and the proposed geometric constraint, the robustness of correspondence matching on the ground images along the optic axis can be effectively improved, and thus prove the validity of the proposed algorithm.
The ship target in the long-range imaging is lack of the features such as size, texture ,shape and so on, so the ship target detection in the context of complex sky and sea has much difficulty. However, the character that the ship target is always located in the sea-sky area will help narrow down the search and suppress the unnecessary noise interference outside the area. The paper proposes a novel method of ship target detection based on sea-sky-line Detection. The method first use a robust and efficient method to extract the sea-sky-line region based on segmentation result, Then, we design a multi-scale filter and adopt an adaptive threshold method of image segmentation to extract the targets from the filtered image. Our filter is adaptive to the variable size of targets and has good ability to overcome the noise. The experimental results show that this method can detect IR target such as warship in a complex sea background.
Detection of infrared dim small target is an important task in many application fields such as automatic target detection,
target search and tracking, and early warning. By combining the block-based background reconstruction and min-cut of
non-balanced graph, a dim small target detection algorithm is presented. First, a background reconstruction based on a
new modeling is presented. Secondly, the background is suppressed though subtracting the reconstructed image from the
original image. Lastly, further segmentation using min-cut for non-balanced graph to the background suppressed image
is proposed in order to obtain the binary image containing target. The optimal segmentation threshold is selected by
heuristic search based on the optimal min-cut. Experimental results show that the proposed method can suppress
background noise and clutter effectively and detect infrared small target accurately.