- Research Article
250
- 10.1016/j.imavis.2005.09.010
Nonlinear structure tensors
- Nov 30, 2005
- Image and Vision Computing
- Thomas Brox + 3 more +3
Nonlinear structure tensors
In this paper, a novel multi-scale nonlinear structure tensor based corner detection algorithm is proposed to improve effectively the classical Harris corner detector. By considering both the spatial and gradient distances of neighboring pixels, a nonlinear bilateral structure tensor is constructed to examine the image local pattern. It can be seen that the linear structure tensor used in the original Harris corner detector is a special case of the proposed bilateral one by considering only the spatial distance. Moreover, a multi-scale filtering scheme is developed to tell the trivial structures from true corners based on their different characteristics in multiple scales. The comparison between the proposed approach and four representative and state-of-the-art corner detectors shows that our method has much better performance in terms of both detection rate and localization accuracy.
Nonlinear structure tensors
Nonlinear structure tensors
Structure-Adaptive Anisotropic Filter with Local Structure Tensors
We propose a new structure-adaptive anisotropic filtering scheme based on the local structure tensor. We utilize the local structure tensor to measure image local anisotropic features and estimate the orientation of image structures, and these informations are then used to shape and control the anisotropic Gaussian kernel. The proposed filter denoises noisy images while image structures such as corners, junctions and edges are well preserved. Our experimental results clearly show that the proposed scheme outperforms some other adaptive filters such as the adaptive Wiener filter, Weickert’s edge enhancing diffusion (EED) filter and Yang's structure-adaptive anisotropic filter in terms of both mean square errors (MSE) and visual quality, and the one based on the nonlinear structure tensor (NLST) can give much better denoising results than that based on the linear structure tensor (LST), particularly in edge regions.
Read moreAnisotropic filtering with nonlinear structure tensors
We present an anisotropic filtering scheme which uses a nonlinear version of the local structure tensor to dynamically adapt the shape of the neighborhood used to perform the estimation. In this way, only the samples along the orthogonal direction to that of maximum signal variation are chosen to estimate the value at the current position, which helps to better preserve boundaries and structure information. This idea sets the basis of an anisotropic filtering framework which can be applied for different kinds of linear filters, such as Wiener or LMMSE, among others. In this paper, we describe the underlying idea using anisotropic gaussian filtering which allows us, at the same time, to study the influence of nonlinear structure tensors in filtering schemes, as we compare the performance to that obtained with classical definitions of the structure tensor.
Read moreA Novel Structure Tensor Using Nonlocal Total Variation Operator
The structure tensor known as a second moment matrix which integrates the local data information of the image. It has been a well-established tool in the image processing field. To date a variety of nonlinear structure tensors have been emerged. Among them, the non-local structure tensors (NLSTs) is the focus of researching for the reason that it explores the spatial interactions in images. However, the performance of the existing NLST in image analysis is limited. In this paper, we propose a new structure tensor calculation method by using the nonlocal means filter to smooth the matrix-valued data. The resulting nonlocal structure tensor is effective in the orientation estimation and structural analysis of the image. Meanwhile, the nonlocal TV model based this structure tensor has been successfully applied in noise removal. Experimental results show that our model has better performance in preserving the structures, details and textures.
Read moreImproved Rapid Corner Detection Algorithm in Medical Image
Improved Rapid Corner Detection Algorithm in Medical Image
Tensor Processing for Texture and Colour Segmentation
In this paper, we propose an original approach for texture and colour segmentation based on the tensor processing of the nonlinear structure tensor. While the tensor structure is a well established tool for image segmentation, its advantages were only partly used because of the vector processing of that information. In this work, we use more appropriate definitions of tensor distance grounded in concepts from information theory and compare their performance on a large number of images. We clearly show that the traditional Frobenius norm-based tensor distance is not the most appropriate one. Symmetrized KL divergence and Riemannian distance intrinsic to the manifold of the symmetric positive definite matrices are tested and compared. Adding to that, the extended structure tensor and the compact structure tensor are two new concepts that we present to incorporate gray or colour information without losing the tensor properties. The performance and the superiority of the Riemannian based approach over some recent studies are demonstrated on a large number of gray-level and colour data sets as well as real images.KeywordsColour ImageSegmentation MethodGeodesic DistanceStructure TensorTexture SegmentationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Read moreThreshold-Guided Design and Optimization for Harris Corner Detector Architecture
High-speed corner detection is an essential step in many real-time computer vision applications, e.g., object recognition, motion analysis, and stereo matching. Hardware implementation of corner detection algorithms, such as the Harris corner detector (HCD) has become a viable solution for meeting real-time requirements of the applications. A major challenge lies in the design of power, energy and area efficient architectures that can be deployed in tightly constrained embedded systems while still meeting real-time requirements. In this paper, we proposed a bit-width optimization strategy for designing hardware-efficient HCD that exploits the thresholding step in the algorithm to determine interest points from the corner responses. The proposed strategy relies on the threshold as a guide to truncate the bit-widths of the operators at various stages of the HCD pipeline with only marginal loss of accuracy. Synthesis results based on 65-nm CMOS technology show that the proposed strategy leads to power-delay reduction of 35.2%, and area reduction of 35.4% over the baseline implementation. In addition, through careful retiming, the proposed implementation achieves over 2.2 times increase in maximum frequency while achieving an area reduction of 35.1% and power-delay reduction of 35.7% over the baseline implementation. Finally, we performed repeatability tests to show that the optimized HCD architecture achieves comparable accuracy with the baseline implementation (average decrease of repeatability is less than 0.6%).
Read moreNoise-resistant structure-preserving multiscale image decomposition
A challenge for current edge-preserving image decompositions is to deal with noisy images. Gradient- or magnitude difference-based techniques regard the noise boundary as edges, while the local extrema-based method suffers from averaging noisy envelops. We introduce local anisotropy derived from nonlinear local structure tensor to differentiate edges from fine-scale details and noises. Providing low smoothness weights to the places with large local anisotropy rather than a large gradient under the improved weighted least squares optimization framework, we present a noise-resistant, structure-preserving smoothing operator. By either progressively or recursively applying this operator we construct our structure-preserving multiscale image decomposition. Based on the key property of our algorithm, noise resistance, we compare our results with existing edge-preserving image decomposition methods and demonstrate the effectiveness and robustness of our structure-preserving decompositions in the context of image restoration, noisy image abstraction, and noisy image dehazing.
Read moreA Global Sparse Stereo Matching Method under Structure Tensor Constraint
In this paper, a global algorithm based on graph cuts theory is proposed to solve the sparse stereo matching problem. The sparse feature points are extracted by the Harris corner detector. The matching problem is transformed into a labeling problem in the sparse graph which can be solved by energy minimization. In this algorithm, the graph is constructed by sparse feature points instead of pixels, which can lead to simple graph structure. In addition, a structure tensor descriptor, which is invariant to varying illumination, is used as similarity measurement to obtain more accurate result. The experimental results show that this algorithm can obtain accurate matching result.
Read morePipelining Harris corner detection with a tiny FPGA for a mobile robot
With their parallelizable inner structures, field programmable gate array (FPGA) are increasing their popularity in today's embedded systems. In this paper, we present an implemented, unique and pipelined FPGA architecture designed with Verilog HDL to be used on a mobile robot for detecting corners in colored stereo images using Harris corner detection (HCD) algorithm in real time. The architecture consists of 3 pipelined modules and processes RGB555 formatted images in 640×480 resolution. The design is implemented on Xilinx's ML501 board having a XC5VLX50 FPGA, one of the smallest FPGAs of Virtex-5 series. Raw and processed data are stored into a single DDR2 memory of Micron, MT4HTF3264HY on the board, allowing only a single read or write operation at a time. By using less than 75% of FPGA resources and a 100MHz system clock, we achieved a corner detection rate of 0.33 pixels per clock cycle (ppcc) corresponding to a corner detection frequency of 54Hz for the stereo images.
Read moreEnhanced low-complexity pruning for corner detection
Low-complexity corner detection is essential for many real-time computer vision applications that need to be executed on low-cost/low-power embedded platforms such as robots. The widely used Shi---Tomasi and Harris corner detectors become prohibitive in such platforms due to their high computational complexity, which is attributed to the need to apply a complex corner measure on the entire image. In this paper, we introduce a novel and computationally efficient technique to accelerate the Shi---Tomasi and Harris corner detectors. The proposed technique consists of two steps. In the first step, the complex corner measure is replaced with simple approximations to quickly prune away non-corners. In the second step, the complex corner measure is applied to a small corner candidate set obtained after pruning. Evaluations using standard image benchmarks show that the proposed pruning technique achieves up to 75 % speedup on the Nios-II platform, while yielding corners with comparable or better accuracy than the conventional Shi---Tomasi and Harris detectors.
Read moreImproved Non-local Means Algorithm for Image Denosing
In order to solve the problem that Non-local means method (NLM) results in a weaker retention and noise immunity, an improved non-local means (INLM) is proposed. This paper introduces bilateral structure tensor combined with the spatial distance and the gradient distance to weight the structure tensor matrix. It has stronger noise immunity and can extract the texture structure more accurately. Mahalanobis distance is applied as a measure of the similarity of image pixels to suppress noise interference. Analyses and simulation results prove the validity of the proposed algorithm.
Read moreAccelerated Corner-Detector Algorithms
Fast corner-detector algorithms are important for achieving real time in different computer vision applications. In this paper, we present new algorithm implementations for corner detection that make use of graphics processing units (GPU) provided by commodity hardware. The programmable capabilities of modern GPUs allow speeding up counterpart CPU algorithms. In the case of corner-detector algorithms, most steps are easily translated from CPU to GPU. However, there are challenges for mapping the feature selection step to the GPU parallel computational model. This paper presents a template for implementing corner-detector algorithms that run entirely on GPU, resulting in significant speed-ups. The proposed template is used to implement the KLT corner detector and the Harris corner detector, and numerical results are presented to demonstrate the algorithms efficiency.
Read moreA Semi-Automatic Image Registration Scheme based on a Noval Asymmetrical Corner Detector (ACD)
In this paper, a novel Asymmetrical Corner Detector (ACD) algorithm based on the auto-correlation is presented. Based on the study of the fact that asymmetrical corner points are the most common reality in remotely sensed imagery data, the ACD is designed to detect interest points more favourable to slightly asymmetrical points rather than ideal symmetrical points. The experimentation results using images taken by different sensors indicate that the ACD has obtained excellent performance in terms of point localization and computational efficiency. It is more capable of selecting high quality ground control points (GCPs) than some well-established corner detectors, e.g. the Harris Corner Detector [1]. A semi-automatic image co-registration scheme is then proposed, which employed the ACD algorithm to extract evenly distributed GCPs across the overlapped area in the reference image.
Read moreCorner Detection of the Computer VR Microscope Image Based on the 3D Reconstruction Algorithm.
In order to solve the problem of multisolution and ill-formedness of the 3D reconstruction method of a single image (purpose), the author proposes a microscope image segmentation algorithm based on the Harris multiscale corner detection. Separating complex engineering images into several simple basic geometric shapes, rebuild them separately to avoid the ill-conditioned solution problem of directly recovering depth information. In order to improve the registration accuracy of the corner-based image registration algorithm, the idea of multiresolution analysis was introduced into the classic Harris corner detection, and a gray intensity variation formula based on the wavelet transform was constructed, and a scale transformation characteristic was obtained so that the improved Harris corner detection algorithm is invariant to rotation, translation, and scale. Experimental results show that after reconstruction, the error between the length of the object measured based on the point cloud data and the actual length of the object is small, and both remain within the error range of 3 mm. The experiment verifies the fast, accurate, and stable characteristics of the improved algorithm.
Read more