Showing papers on "Phase correlation published in 2015"
TL;DR: A novel subpixel phase correlation method using singular value decomposition (SVD) and the unified random sample consensus (RANSAC) algorithm and the pixel locking effect was found to be significantly weakened by the proposed method, as compared with the original Hoge's method.
Abstract: Subpixel translation estimation using phase correlation is a fundamental task for numerous applications in the remote sensing community. The major drawback of the existing subpixel phase correlation methods lies in their sensitivity to corruption, including aliasing and noise, as well as the poor performance in the case of practical remote sensing data. This paper presents a novel subpixel phase correlation method using singular value decomposition (SVD) and the unified random sample consensus (RANSAC) algorithm. In the proposed method, SVD theoretically converts the translation estimation problem to one dimensions for simplicity and efficiency, and the unified RANSAC algorithm acts as a robust estimator for the line fitting, in this case for the high accuracy, stability, and robustness. The proposed method integrates the advantages of Hoge's method and the RANSAC algorithm and avoids the corresponding shortfalls of the original phase correlation method based only on SVD. A pixel-to-pixel dense matching scheme on the basis of the proposed method is also developed for practical image registration. Experiments with both simulated and real data were carried out to test the proposed method. In the simulated case, the comparative results estimated from the generated synthetic image pairs indicate that the proposed method outperforms the other existing methods in the presence of both aliasing and noise, in both accuracy and robustness. Moreover, the pixel locking effect that commonly occurs in subpixel matching was also investigated. The degree of pixel locking effect was found to be significantly weakened by the proposed method, as compared with the original Hoge's method. In the real data case, experiments using different bands of ZY-3 multispectral sensor-corrected images demonstrate the promising performance and feasibility of the proposed method, which is able to identify seams of the image stitching between sub-charge-coupled device units.
116 citations
TL;DR: Through simulations of the synthetic data and the real radar data, the effectiveness of the fast parameter estimation algorithm is verified and the GSCFT and the NUFFT has a wider applicability in ISAR imaging applications.
Abstract: In inverse synthetic aperture radar (ISAR) imaging of nonuniformly rotating targets, such as highly maneuvering airplanes and ships fluctuating with oceanic waves, azimuth echoes have to be modeled as cubic phase signals (CPSs) after the range migration compensation and the translational-induced phase error correction. For the CPS model, the chirp rate and the quadratic chirp rate, which deteriorate the azimuth focusing quality due to the Doppler frequency shift, need to be estimated with a parameter estimation algorithm. In this paper, by employing the proposed generalized scaled Fourier transform (GSCFT) and the nonuniform fast Fourier transform (NUFFT), a fast parameter estimation algorithm is presented and utilized in the ISAR imaging of the nonuniformly rotating target. Compared to the scaled Fourier transform-based algorithm, advantages of the fast parameter estimation algorithm include the following: 1) the computational cost is lower due to the utilization of the NUFFT, and 2) the GSCFT has a wider applicability in ISAR imaging applications. The CPS model and the algorithm implementation are verified with the real radar data of a ship target. In addition, the cross-term, which plays an important role in correlation algorithms, is analyzed for the fast parameter estimation algorithm. Through simulations of the synthetic data and the real radar data, we verify the effectiveness of the fast parameter estimation algorithm and the corresponding ISAR imaging algorithm.
98 citations
TL;DR: In simulations using ISAR images measured in a compact range, the proposed method yielded high classification ratios with small-sized data regardless of the location of the rotation center, whereas the existing method was very sensitive to the locationof it.
Abstract: This paper proposes an efficient method to classify inverse synthetic aperture radar (ISAR) images. The proposed method achieves invariance to translation and rotation of ISAR images by using two-dimensional (2D) Fourier transform (FT) of ISAR images, polar mapping of the 2D FT image, and a simple nearest-neighbor classifier. In simulations using ISAR images measured in a compact range, the proposed method yielded high classification ratios with small-sized data regardless of the location of the rotation center, whereas the existing method was very sensitive to the location of it.
43 citations
25 May 2015
TL;DR: In this paper, the Fourier transform coefficients of partial derivatives of the signal satisfy an annihilation relation, and necessary and sufficient conditions for unique recovery of piecewise constant images using the above annihilation relation are presented.
Abstract: We introduce a Prony-like method to recover a continuous domain 2-D piecewise smooth image from few of its Fourier samples. Assuming the discontinuity set of the image is localized to the zero level-set of a trigonometric polynomial, we show the Fourier transform coefficients of partial derivatives of the signal satisfy an annihilation relation. We present necessary and sufficient conditions for unique recovery of piecewise constant images using the above annihilation relation. We pose the recovery of the Fourier coefficients of the signal from the measurements as a convex matrix completion algorithm, which relies on the lifting of the Fourier data to a structured low-rank matrix; this approach jointly estimates the signal and the annihilating filter. Finally, we demonstrate our algorithm on the recovery of MRI phantoms from few low-resolution Fourier samples.
36 citations
TL;DR: A new, fast and computationally efficient lateral subpixel shift registration algorithm is presented that reduces computation time and memory requirements without sacricing the accuracy associated with the usual FFT approach accuracy.
Abstract: A new, fast and computationally efficient lateral subpixel shift registration algorithm is presented. It is limited to register images that differ by small subpixel shifts otherwise its performance degrades. This algorithm significantly improves the performance of the single-step discrete Fourier transform approach proposed by Guizar-Sicairos and can be applied efficiently on large dimension images. It reduces the dimension of Fourier transform of the cross correlation matrix and reduces the discrete Fourier transform (DFT) matrix multiplications to speed up the registration process. Simulations show that our algorithm reduces computation time and memory requirements without sacricing the accuracy associated with the usual FFT approach accuracy.
31 citations
TL;DR: An improved phase correlation (PC) method based on 2-D plane fitting and the maximum kernel density estimator (MKDE) is proposed, which combines the idea of Stone's method and robust estimator MKDE and is robust to aliasing and noise.
Abstract: In this letter, an improved phase correlation (PC) method based on 2-D plane fitting and the maximum kernel density estimator (MKDE) is proposed, which combines the idea of Stone's method and robust estimator MKDE. The proposed PC method first utilizes a vector filter to minimize the noise errors of the phase angle matrix and then unwraps the filtered phase angle matrix by the use of the minimum cost network flow unwrapping algorithm. Afterward, the unwrapped phase angle matrix is robustly fitted via MKDE, and the slope coefficients of the 2-D plane indicate the subpixel shifts between images. The experiments revealed that the improved method can effectively avoid the impact of outliers on the phase angle matrix during the plane fitting and is robust to aliasing and noise. The matching accuracy can reach 1/50th of a pixel using simulated data. The real image sequence tracking experiment was also undertaken to demonstrate the effectiveness of the proposed PC method with a registration accuracy of root-mean-square error better than 0.1 pixels.
19 citations
TL;DR: It is demonstrated that the proposed method leads to an improvement of the registration performance, and its applicability to real images is shown by providing successful examples of blurred image registration followed by depth-of-field extension and multichannel blind deconvolution.
Abstract: In this paper, we extend our recent registration method designed specifically for registering blurred images. The original method works for unknown blurs, assuming the blurring point-spread function (PSF) exhibits an $N$ -fold rotational symmetry. Here, we also generalize the theory to the case of dihedrally symmetric blurs, which are produced by the PSFs having both rotational and axial symmetries. Such kind of blurs are often found in unfocused images acquired by digital cameras, as in out-of-focus shots the PSF typically mimics the shape of the shutter aperture. This makes our registration algorithm particularly well-suited in applications where blurred image registration must be used as a preprocess step of an image fusion algorithm, and where common registration methods fail, due to the amount of blur. We demonstrate that the proposed method leads to an improvement of the registration performance, and we show its applicability to real images by providing successful examples of blurred image registration followed by depth-of-field extension and multichannel blind deconvolution.
19 citations
TL;DR: The purpose of this paper is to improve an existing implementation of multi-scale retinex (MSR) by utilizing the fast Fourier transforms within the illumination estimation step of the algorithm to improve the speed at which Gaussian blurring filters were applied to the original input image.
Abstract: Efficiency in terms of both accuracy and speed is highly important in any system, especially when it comes to
image processing. The purpose of this paper is to improve an existing implementation of multi-scale retinex
(MSR) by utilizing the fast Fourier transforms (FFT) within the illumination estimation step of the algorithm
to improve the speed at which Gaussian blurring filters were applied to the original input image. In addition,
alpha-rooting can be used as a separate technique to achieve a sharper image in order to fuse its results with
those of the retinex algorithm for the sake of achieving the best image possible as shown by the values of the
considered color image enhancement measure (EMEC).
17 citations
16 Nov 2015
TL;DR: In this article, the wave propagation velocity was determined by manual estimates, cross-correlation, and Radon and Fourier transforms resulting in waves propagation velocities between 1.5 and 9 m/s, with spectral contents in the audible range of the heart tones.
Abstract: We studied the mechanical waves after mitral and aortic valve closure in the septal wall of 34 pigs (18 diabetic for heart failure and 16 controls), using high frame rate imaging on a programmable commercial ultrasound machine. The wave pattern was extracted from the data through a local tissue velocity estimator based on phase correlation. The wave propagation velocity was determined by manual estimates, cross-correlation, and Radon and Fourier transforms resulting in waves propagation velocities between 1.5 and 9 m/s, with spectral contents in the audible range of the heart tones. The Fourier analysis showed that the propagation velocity increases with frequency, which is consistent with asymmetric Lamb waves as can be expected from the geometry of the septal wall surrounded by blood.
15 citations
TL;DR: A frequency offset estimation (FOE) algorithm based on improved fast Fourier transform (FFT) for coherent optical systems that adopts multi-steps interpolation with the increasing number of samples to improve the estimation accuracy gradually is investigated.
Abstract: We investigate a frequency offset estimation (FOE) algorithm based on improved fast Fourier transform (FFT) for coherent optical systems. The algorithm implements FFT operation with a small number of samples and then adopts multi-steps interpolation with the increasing number of samples to improve the estimation accuracy gradually. In a 28-GBd coherent quaternary phase-shift keying system, simulation results show that the proposed algorithm reaches the same estimation accuracy with least-squares FOE algorithm that utilizes 64 time spans (LS-64) under the same total number of samples $L$ . But the number of complex multiplications required by the proposed algorithm is just 7.26% and 6.75% of that required by LS-64 at $L=1024$ and $L=2048$ , respectively.
14 citations
TL;DR: Simulation results of the proposed technique prove its better noise immunity than FFT-based method and its superiority than existing methods based on FRFT.
19 Apr 2015
TL;DR: A Fourier optics setup is proposed such that signal recovery up to a global phase factor is possible with very high probability whenever M ≳ 4k log2(N/k) random Fourier intensity measurements are available.
Abstract: This paper considers the problem of recovering a k-sparse, N-dimensional complex signal from Fourier magnitude measurements. It proposes a Fourier optics setup such that signal recovery up to a global phase factor is possible with very high probability whenever M ≳ 4k log 2 (N/k) random Fourier intensity measurements are available. The proposed algorithm is comprised of two stages: An algebraic phase retrieval stage and a compressive sensing step subsequent to it. Simulation results are provided to demonstrate the applicability of the algorithm for noiseless and noisy scenarios.
1 Oct 2015
TL;DR: In this article, the authors deal with the filtering of microscopic images with moire-like noise patterns, analyzes the magnitude spectrum of the Fast Fourier Transform of the noisy image, determines and removes the undesired components.
Abstract: This paper deals with the filtering of microscopic images with moire like noise patterns. The proposed method analyzes the magnitude spectrum of the Fast Fourier Transform of the noisy image, determines and removes the undesired components. Comparisons with a classical method are provided. The experimental results obtained so far are promising.
21 Dec 2015
TL;DR: An image mosaicing technique based on an improved discrete cosine transform (DCT)-based phase correlation (PC) method is presented, which is an extension of DCT-based PC with added blur invariance property and contains a single peak of normalized phase correlation corresponding to the percentage of overlap.
Abstract: This paper presents an image mosaicing technique based on an improved discrete cosine transform (DCT)-based phase correlation (PC). The improved DCT-based PC method is an extension of DCT-based PC with added blur invariance property. Though fast Fourier transform (FFT)-based methods are efficient for mosaicing of images, they often fail to achieve good quality mosaics when there is illumination variation, blurring or color difference due to imperfect image capturing (image acquisition). Cross correlation (CC) is simple to implement however it has problem when the images have different brightness. Also, the peak of normalized cross correlation is not distinct which gives erroneous results leading to misaligned images in the mosaic. The proposed technique in contrast, is simple, computationally efficient and is able to address the problem of illumination/color variation and blurring. Besides, it contains a single peak of normalized phase correlation corresponding to the percentage of overlap. The experimental analysis for different image sets shows that the method produces better mosaics even for degraded images unlike the other conventional correlation approaches.
TL;DR: It is proved that the well-known algorithm of phase correlation is a special case of the method based on the idea of representing translations as convolutions with unknown shifted delta functions, and performing Wiener deconvolution in order to recover the shift between two images.
Abstract: In this letter, we propose a global method for registering color images with respect to translation. Our approach is based on the idea of representing translations as convolutions with unknown shifted delta functions, and performing Wiener deconvolution in order to recover the shift between two images. We then derive a quaternionic version of the Wiener deconvolution filter in order to register color images. The use of Wiener filter also allows us to explicitly take into account the effect of noise. We prove that the well-known algorithm of phase correlation is a special case of our method, and we experimentally demonstrate the advantages of our approach by comparing it to other known generalizations of the phase correlation algorithm.
1 Jun 2015
TL;DR: This work introduces a novel method for extracting Fourier descriptors, which preserve the phase of Fourier coefficients and have the desired invariance, and proposes specific points, called pseudomirror points, to be used as shape orientation reference.
Abstract: Contour-based Fourier descriptors are established as a simple and effective shape description method for content-based image retrieval. In order to achieve invariance under rotation and starting point change, most Fourier descriptor implementations disregard the phase of the Fourier coefficients. We introduce a novel method for extracting Fourier descriptors, which preserve the phase of Fourier coefficients and have the desired invariance. We propose specific points, called pseudomirror points, to be used as shape orientation reference. Experimental results indicate that the proposed method significantly outperforms other Fourier descriptor based techniques.
TL;DR: This work analyses the complexity of phase maps and the problems caused by it in real applications, and proposes a correspondence finding method based on space conversion, which is successful and effective.
Abstract: Phase correlation is an effective method used for 3D shape measurement. It has a defect in the step of finding corresponding points. This work analyses the complexity of phase maps and the problems caused by it in real applications, proposes a correspondence finding method based on space conversion. Applying space conversion, two sets of phase maps from two cameras are integrated to a unique phase space. Accordingly, searching corresponding point between two images can be carried out in the same image coordinate system of the projector. As a supplementary, two algorithms are given for CC method and VR method. Experimental results show that proposed algorithms are successful and effective.
01 Mar 2015-Nuclear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment
TL;DR: In this paper, an iterative pseudo-polar Fourier reconstruction through total variation minimization (PPF-TVM) was proposed for linear scan CT, which applies a straight line trajectory.
Abstract: In this study, we consider a novel form of computed tomography (CT), that is, linear scan CT (LCT), which applies a straight line trajectory. Furthermore, an iterative algorithm is proposed for pseudo-polar Fourier reconstruction through total variation minimization (PPF-TVM). Considering that the sampled Fourier data are distributed in pseudo-polar coordinates, the reconstruction model minimizes the TV of the image subject to the constraint that the estimated 2D Fourier data for the image are consistent with the 1D Fourier transform of the projection data. PPF-TVM employs the alternating direction method (ADM) to develop a robust and efficient iteration scheme, which ensures stable convergence provided that appropriate parameter values are given. In the ADM scheme, PPF-TVM applies the pseudo-polar fast Fourier transform and its adjoint to iterate back and forth between the image and frequency domains. Thus, there is no interpolation in the Fourier domain, which makes the algorithm both fast and accurate. PPF-TVM is particularly useful for limited angle reconstruction in LCT and it appears to be robust against artifacts. The PPF-TVM algorithm was tested with the FORBILD head phantom and real data in comparisons with state-of-the-art algorithms. Simulation studies and real data verification suggest that PPF-TVM can reconstruct higher accuracy images with lower time consumption.
TL;DR: The overall efficiency for 5D regularization is improved largely by using FFT instead of NFFT to tackling nearly all the frequent and time-consuming matrix-vector multiplication.
Abstract: Yang Hao* and Li Jinsong, Research Institute of Petroleum Exploration and Development, PetroChina Limited Company; Ma Shufang, CNOOC Research Institute Summary Both 5D seismic data interpolation and regularization serve as powerful tools to reconstruct seismic data. The latter has a more attractive advantage than the former in that it avoids binning the input seismic traces into a regular spatial grid and the geometry errors it caused, especially when the spatial sampling of the input data is strong irregular. However, as far as the Fourier-based algorithms are concerned, the computing time of the 5D regularization, though significantly shortened by using Non-uniform Fast Fourier Transform (NFFT), is still dozens of times longer than that of the 5D interpolation which benefits quit a lot from Fast Fourier Transform (FFT). We introduce a fast Fourier inversion strategy for 5D seismic data regularization. Iterative method of preconditioned Conjugate Gradient is adopted to solve an optimization problem, through which the matrix in the iterative loop get a block Toeplitz structure so that the matrix-vector multiplication can be accelerated via FFT algorithm. The overall efficiency for 5D regularization is therefore improved largely by using FFT instead of NFFT to tackling nearly all the frequent and time-consuming matrix-vector multiplication.
Patent•
28 Jan 2015
28 Dec 2015
TL;DR: An automatic corneal subbasal nerve registration system is proposed that can be used to give ophthalmologists a summarized and objective description about a diabetic patient's health status using a sequence of CCM images that have been captured from different imaging devices and/or at different times.
Abstract: Confocal microscopy is employed as a fast and non-invasive way to capture a sequence of images from different layers and membranes of the cornea. The captured images are used to extract useful and helpful clinical information for early diagnosis of corneal diseases such as, Diabetic Peripheral Neuropathy (DPN). In this paper, an automatic corneal subbasal nerve registration system is proposed. The main aim of the proposed system is to produce a new informative corneal image that contains structural and functional information. In addition a colour coded corneal image map is produced by overlaying a sequence of Cornea Confocal Microscopy (CCM) images that differ in their displacement, illumination, scaling, and rotation to each other. An automatic image registration method is proposed based on combining the advantages of Fast Fourier Transform (FFT) and phase correlation techniques. The proposed registration algorithm searches for the best common features between a number of sequenced CCM images in the frequency domain to produce the formative image map. In this generated image map, each colour represents the severity level of a specific clinical feature that can be used to give ophthalmologists a clear and precise representation of the extracted clinical features from each nerve in the image map. Moreover, successful implementation of the proposed system and the availability of the required datasets opens the door for other interesting ideas, for instance, it can be used to give ophthalmologists a summarized and objective description about a diabetic patient's health status using a sequence of CCM images that have been captured from different imaging devices and/or at different times.
Patent•
24 Jun 2015
TL;DR: In this article, a remote sensing satellite multispectral image registration method is proposed, in which the two images are partitioned in blocks, and angular points of each pair of image blocks are extracted, and in order to realize registration of a near infrared image, the main direction phase coincidence characteristic of each block is extracted.
Abstract: The invention discloses a remote sensing satellite multispectral image registration method. The remote sensing satellite multispectral image registration method comprises the following steps that 1, a rough registration image is obtained according to the distance of a multispectral detector; 2, histogram projection is carried out on an image to be registered on the basis of a reference image; 3, the two images are partitioned in blocks, and for each pair of image blocks, angular points of each pair of image blocks are extracted, and in order to realize registration of a near infrared image, the main direction phase coincidence characteristic of each block is extracted, and a positive and negative sub pixel matching point set of each block is obtained on the basis of phase correlation and surface fitting; 4, triangular grid partition is carried out on the reference image and the image to be registered, a transformation coefficient is worked out, and resampling is carried out on the image to be registered to obtain a registered image. Compared with the prior art, the method solves the problem that as the gray difference of features of different reflectivities on different spectral sections is large, registration is difficult, and also effectively solves the problem that large remote sensing images in different areas distort to different extents.
TL;DR: In this paper, a global motion estimation method based on Fourier-Mellin and phase-correlation is presented, which is robust to camera vibration or unwanted movement regardless of foreground object's movement.
Abstract: Global motion estimation aims at estimating the whole image changes causing by camera moving. This paper presents a global motion estimation method based on Fourier-Mellin and Phase-Correlation. Rotational angle and scaling factor are acquired by Phase-Correlation between Fourier-Mellin transform images of the reference and current image. Log-Polar coordinates are used to the Flourier amplitude spectrum. Translational parameter is obtained using sub-block based on PhaseCorrelation. The proposed algorithm is robust to camera vibration or unwanted movement regardless of foreground object’s movement. Experimental results show that the proposed algorithm can efficiently and accurately estimate the parameter between two consecutive frames. Keywords-fourier-mellin; phase-correlation; global motion estimation; log-polar coordinate
Patent•
16 Sep 2015
TL;DR: In this paper, a moving image compensation method and device and a display device is presented. And the method comprises the steps that a first image corresponding to a frame before movement and a second image corresponding with a frame after movement are acquired; the first image and the second image are respectively divided into multiple image blocks; calculation is performed on any one pair of mutually corresponding image blocks in the first and second image on the basis of a phase correlation method so that at least one displacement vector of which phase correlation is greater than a preset threshold value is obtained.
Abstract: The invention provides a moving image compensation method and device and a display device. The method comprises the steps that a first image corresponding to a frame before movement and a second image corresponding to a frame after movement are acquired; the first image and the second image are respectively divided into multiple image blocks; calculation is performed on any one pair of mutually corresponding image blocks in the first image and the second image on the basis of a phase correlation method so that at least one displacement vector of which phase correlation is greater than a preset threshold value is obtained; edge detection is performed on the first image and the second image respectively so that the edge characteristic patterns in the pair of image blocks are obtained; and the displacement vectors matched with the edge characteristic patterns in conversion between the first image and the second image are acquired from the at least one displacement vector according to the rpeset conditions to act as the displacement vectors corresponding to the image blocks in the moving compensation process. Accuracy of the displacement vectors confirmed by an MEMC algorithm can be enhanced so that enhancement of the moving object compensation effect is facilitated.
14 Dec 2015
TL;DR: The results suggest that rigid registration using phase correlation may be fairly robust to gamma correction, quantization and multi-spectral acquisition, but more sensitive to differences in illumination and lighting conditions between the input images.
Abstract: The phase correlation method is a computationally-efficient technique for image alignment. Presently, the method is capable of performing rigid image registration with sub-pixel accuracy, and is fairly robust to noise and long translations. However, there are also cases when the images to be aligned were taken at different times or come from different sensors, and may present differences in intensity values or illumination. Many algorithms exist to deal with these issues; however, most of them are computationally expensive. In this article, we explore the robustness of the phase correlation method to illumination and/or intensity changes by means of a quantitative evaluation using artificially-generated rigid transformations. Our results suggest that rigid registration using phase correlation may be fairly robust to gamma correction, quantization and multi-spectral acquisition, but more sensitive to differences in illumination and lighting conditions between the input images.
TL;DR: A fast algorithm for the registration of shapes implicitly represented by their characteristic functions that is characterised with a better accuracy, a higher convergence speed, robustness at the presence of excessive noise and a better performance for registration over large databases of shapes, in comparison with other state-of-the-art shape registration techniques in the literature.
Abstract: This study presents a fast algorithm for the registration of shapes implicitly represented by their characteristic functions. The algorithm proposed here aims to recover the registration parameters (scaling, rotation and translation) by minimising a dissimilarity term between the two shapes. The proposed algorithm is based on phase correlation and statistical shape moments to compute the registration parameters individually. The registration method proposed here is applied to various registration problems, to address issues such as the registration of shapes with various topologies and registration of complex shapes containing various numbers of sub-shapes. The method proposed here is characterised with a better accuracy, a higher convergence speed, robustness at the presence of excessive noise and a better performance for registration over large databases of shapes, in comparison with other state-of-the-art shape registration techniques in the literature.
2 Apr 2015
TL;DR: This brief presents an architecture for in-place fast Fourier transform (IFFT) computation for real valued signals based on modified radix-2 algorithm, which removes the redundant operations from the flow graph.
Abstract: This brief presents an architecture for in-place fast Fourier transform (IFFT) computation for real valued signals. The proposed computation is based on modified radix-2 algorithm, which removes the redundant operations from the flow graph. The modified flow graph contains only real data paths as opposed to complex data paths in a regular flow graph. A new processing element (PE) is proposed which consists of two radix-2 butterflies that can process four inputs signals in parallel. A new conflict-free memory-addressing scheme is proposed to ensure the continuous operation of the FFT processor. The addressing scheme is also used to support multiple parallel PEs. As the proposed PE processes the four parallel inputs that reduce computation cycles and increase the speed compared to prior work. The number of computation cycles is reduced with increase in the number of PEs. As redundant operations are removed, that reduces hardware cost.
15 Jul 2015
TL;DR: To reduce computational complexity and obtain a less distorted result, a similarity transformation is used to approximate the geometric alignment between two images of the same scene, which is estimated using a phase correlation technique.
Abstract: For aerial imaging with UAVs, where there is a limited transmission rate/bandwidth, using low bit depth images instead of full images offers a number of advantages. For example, reducing data transmission, removing superfluous details, and reducing computational loading of on-board platforms (especially for small or micro-scale UAVs). However, the main drawback of using low bit depth imagery is discarding image details of the scene. Fortunately, this can be reconstructed by fusing a sequence of related low bit depth images, which have been properly aligned. To reduce computational complexity and obtain a less distorted result, a similarity transformation is used to approximate the geometric alignment between two images of the same scene, which is estimated using a phase correlation technique. It is shown that the phase correlation method is capable of registering low bit depth images, without any modification, or any pre and/or post-processing.
Patent•
22 Apr 2015
TL;DR: In this paper, an edge significance-based multi-sensor image registration method and system was proposed, where an edge extraction module was used for extracting a reference image and an edge image of a to-be-registered image, and a calculation module is used for converting the edge image to a log-polar coordinate domain, and calculating phase correlation of the image to obtain a rotation angle and a scale factor.
Abstract: The invention relates to an edge significance-based EFMT multi-sensor image registration method and system. The method comprises the following steps that a, an edge extraction module is used for extracting a reference image and an edge image of a to-be-registered image; b, a calculation module is used for converting the edge image to a log-polar coordinate domain, and calculating phase correlation of the log-polar coordinate domain image to obtain a rotation angle and a scale factor of the image; c, an image conversion module is used for converting the to-be-registered image according to the rotation angle and the scale factor, and the edge extraction module is used for extracting the edge image of the converted image; and d, the calculation module is used for calculating the edge image conversion in the step c, the phase correlation and the rotation angle of the edge image of the reference image in the step a to obtain an image transition parameter. According to the scheme, based on a traditional registration algorithm, the FMT conversion is carried out by utilizing the edge image of images, and the accuracy and speed of multi-sensor image registration are improved.
Patent•
15 Jul 2015
TL;DR: In this paper, a sky image cloud cluster movement velocity computing method based on the phase correlation principle was proposed, where the operation flow is simple and direct, and the cloud cluster displacement prediction consuming time can be greatly shortened; the overall movement condition of a cloud cluster in the image can be recognized more effectively, and due to normalization processing in cross-power spectrum computing, the higher robustness is achieved for global image noise.
Abstract: The invention relates to a sky image cloud cluster movement velocity computing method based on the phase correlation principle. The method comprises the following steps that 1, an initial image and a displacement image of a sky image are acquired; 2, gray matrixes of the initial image and the displacement image are generated respectively; 3, image frequency spectrums of the initial image and the displacement image are acquired through the two-dimensional Fourier transform; 4, cross-power spectrum of the initial image and the displacement image and an inverse Fourier transform response matrix of the cross-power spectrum are computed; 5, spike pulse coordinates of the response matrix are extracted to serve as cloud cluster displacement vectors; 6, the cloud cluster movement velocity is computed according to the cloud cluster displacement and the image time interval. According to the sky image cloud cluster movement velocity computing method based on the phase correlation principle, the operation flow is simple and direct, and the cloud cluster displacement prediction consuming time can be greatly shortened; the overall movement condition of a cloud cluster in the image can be recognized more effectively, and due to normalization processing in cross-power spectrum computing, the higher robustness is achieved for global image noise.