Feature Matching Method Based on SURF and Fast Library for Approximate Nearest Neighbor Search

Most visual navigation relies on point matching and many of the matching point pairs are faulty. Generally, they are eliminated by the Random Sample Consensus (RANSAC) algorithms. However, RANSAC algorithms are often time-consuming and their parameters are hard to adjust. For such a purpose, a new method is proposed to eliminate the error matching points of every two consecutive images based on motion information. The proposed method is applicable to the cases that the motion information is available and the interest points are all from a horizontal plane. It does not require iteration and thus time-efficient. To demonstrate the effectiveness, the proposed method is tested on synthetic data of both simulation points and real images. The results show that the proposed method has better performance over RANSAC algorithms.

Nearest neighbor search is a fundamental problem in computer vision and machine learning. The straightforward solution, linear scan, is both computationally and memory intensive in large scale high-dimensional cases, hence is not preferable in practice. Therefore, there have been a lot of interests in algorithms that perform approximate nearest neighbor (ANN) search. In this paper, we propose a novel addition-based vector quantization algorithm, Asymmetric Mapping Quantization (AMQ), to efficiently conduct ANN search. Unlike existing addition-based quantization methods that suffer from handling the problem caused by the norm of database vector, we map the query vector and database vector using different mapping functions to transform the computation of L-2 distance to inner product similarity, thus do not need to evaluate the norm of database vector. Moreover, we further propose Distributed Asymmetric Mapping Quantization (DAMQ) to enable AMQ to work on very large dataset by distributed learning. Extensive experiments on approximate nearest neighbor search and image retrieval validate the merits of the proposed AMQ and DAMQ.

An important aspect of chemoinformatics and material-informatics is the usage of machine learning algorithms to build Quantitative Structure Activity Relationship (QSAR) models. The RANdom SAmple Consensus (RANSAC) algorithm is a predictive modeling tool widely used in the image processing field for cleaning datasets from noise. RANSAC could be used as a “one stop shop” algorithm for developing and validating QSAR models, performing outlier removal, descriptors selection, model development and predictions for test set samples using applicability domain. For “future” predictions (i.e., for samples not included in the original test set) RANSAC provides a statistical estimate for the probability of obtaining reliable predictions, i.e., predictions within a pre-defined number of standard deviations from the true values. In this work we describe the first application of RNASAC in material informatics, focusing on the analysis of solar cells. We demonstrate that for three datasets representing different metal oxide (MO) based solar cell libraries RANSAC-derived models select descriptors previously shown to correlate with key photovoltaic properties and lead to good predictive statistics for these properties. These models were subsequently used to predict the properties of virtual solar cells libraries highlighting interesting dependencies of PV properties on MO compositions.

Nearest neighbor search algorithms have been successful in finding practically useful solutions to computationally difficult problems. In the nearest neighbor search problem, the brute force approach is often more efficient than other algorithms for high-dimensional spaces. A special case exists for objects represented as sparse vectors, where algorithms take advantage of the fact that an object has a zero value for most features. In general, since exact nearest neighbor search methods suffer from the “curse of dimensionality,” many practitioners use approximate nearest neighbor search algorithms when faced with high dimensionality or large datasets. To a reasonable degree, it is known that relying on approximate nearest neighbors leads to some error in the solutions to the underlying data mining problems the neighbors are used to solve. However, no one has attempted to quantify this error or provide practitioners with guidance in choosing appropriate search methods for their task. In this thesis, we conduct several experiments on recommender systems with a goal to find the degree to which approximate nearest neighbor algorithms are subject to these types of error propagation problems. Additionally, we provide persuasive evidence on the trade-off between search performance and analytics effectiveness. Our experimental evaluation demonstrates that a state-of-the-art approximate nearest neighbor search method (L2KNNGApprox) is not an effective solution in most cases. When tuned to achieve high search recall (80% or higher), it provides a fairly competitive recommendation performance compared to an efficient exact search method but offers no advantage in terms of efficiency (0.1x—1.5x speedup). Low search recall (<60%) leads to poor recommendation performance. Finally, medium recall values (60%—80%) lead to reasonable recommendation performance but are hard to achieve and offer only a modest gain in efficiency (1.5x—2.3x).

Orthognathic surgery is applied to restore esthetical facial profile and functional occlusion for patients with dentofacial deformity. Virtual surgical planning (VSP) is indispensable for precise and individualized treatment. Manually designing osteotomy planes is time-consuming and highly experience-dependent. This study aimed to develop and validate an automatic osteotomy plane design method based on deep learning. Methods: A deep learning model, Ortho-OPD (orthognathic osteotomy planes de signer), was proposed, consisting of a segmentation network and the random sample consensus (RANSAC) algorithm. The segmentation network was based on a convolutional neural network (CNN), orthognathic segmenting the craniomaxillofacial (CMF) CT data. Osteotomy planes were then defined by the RANSAC algorithm. Ortho-OPD was trained on 71 samples and tested on 31 cases. The performance was evaluated quantitatively and qualitatively. Results: Ortho-OPD functioned smoothly, and all cases were successfully performed. The 3D boundary-sensitive loss was employed to optimize precision. Evaluation metrics included accuracy and clinical efficiency. The mean dice similarity coefficient (DSC) was 0.920.032 in CMF seg mentation. Ortho-OPD showcased excellent productivity, taking an average of about 9 seconds to complete virtual bimaxillary osteotomy compared to manual work. The angular errors between the predicted planes and ground truth planes, plus the shortest distance from the neural tube or the adjacent apical points to predicted planes, were examined, indicating no significant difference and reliability for preserving vital anatomical structures. Overall, the automatic osteotomy plane design from raw CT data was realized using Ortho-OPD, composed of CNN and RANSAC, providing an efficient and ideal alternative in orthognathic osteotomy planning.

Traditional matching algorithms cannot be directly applied to the fisheye image matching for large distortion existing in fisheye image. Therefore, a matching algorithm based on uncorrected fisheye images is proposed. This algorithm adopts a local feature description method which combines MSER detector with CSLBP descriptor to obtain the image feature. First, the two uncorrected fisheye images captured by binocular vision system are described by the principle of epipolar constraint. Then the region detection is done with MSER and the ellipse fitting is used to the obtained regions. The MSER regions are described by CSLBP subsequently. Finally, in order to exclude the mismatching points of initial match, random sample consensus (RANSAC) algorithm has been adopted to achieve exact match. Experiments show that the method has a good effect on the uncorrected fisheye image matching.

We propose a new algorithm for fast approximate nearest neighbor search based on the properties of ordered vectors. Data vectors are classified based on the index and sign of their largest components, thereby partitioning the space in a number of cones centered in the origin. The query is itself classified, and the search starts from the selected cone and proceeds to neighboring ones. Overall, the proposed algorithm corresponds to locality sensitive hashing in the space of directions, with hashing based on the order of components. Thanks to the statistical features emerging through ordering, it deals very well with the challenging case of unstructured data, and is a valuable building block for more complex techniques dealing with structured data. Experiments on both simulated and real-world data prove the proposed algorithm to provide a state-of-the-art performance.

Acoustic source localization has been an hot research topic with widespread applications in many fields. In the noisy environment or when the reverberation is considerable, the source localization problem becomes challenging and many existing algorithms deteriorate. The paper proposes a robust algorithm which combines the RANdom SAmple Consensus (RANSAC) algorithm, and the Generalized Cross-Correlation (GCC) based Time Difference of Arrival (TDOA). Experiments in real world data show that the proposed algorithm has significantly better performance than the traditional algorithm.

The Iterative Closest Point (ICP) algorithm is one of the most popular methods for geometric alignment of 3-dimensional data points. We focus on how to make it faster for 3D range scanner in intelligent vehicle. The ICP algorithm mainly consists of two parts: nearest neighbor search and estimation of transformation between two data sets. The former is the most time consuming process. Many variants of the k-d trees have been introduced to accelerate the search. This paper presents a remarkably efficient search procedure, exploiting two concepts of approximate nearest neighbor and local search. Consequently, the proposed algorithm is about 24 times faster than the standard k-d tree.

It is important to get reliable measurement data in Electrical impedance tomography (EIT) for getting satisfactory reconstruction result. Because EIT inverse problem is non-linear and ill-posed, incorrect data such as measured data through fault electrode can effect the estimation of resistivity distribution. So, EIT measurement system with methods for checking reliability has been developed. However, most existing methods require extra cost to evaluate reliability of EIT system. This paper presents modified Gauss-Newton (GN) method based on random sample consensus (RANSAC) algorithm for finding fault electrodes and getting good reconstructed image with faulty data for two-phase flow application. Also, suitable residual equation is proposed for determining the threshold to use RANSAC algorithm. Numerical simulations are performed to validate the performance of the proposed method. The results show that the proposed method has a good reconstruction performance compared to the conventional GN method.

Image stitching is one of the branches of computer vision. It combines two or more images for a scene to acquire a high-resolution panoramic image. An invariant local function often uses to stitch two images together. Since the flat plate of a digital radiography (DR) system does not cover all parts of the body, the whole bone structure image cannot seize in a single scan. To solved this problem, image stitching is broadly utilized by medical systems to stitch DR images, which can be helpful for scoliosis or lower extremity deformities in the diagnosis, and pre-operative planning are of great importance. In this paper, the stitching and retrieval of medical images planned. To conquer the background noise in medical images, and improve the recovery of quality and stitching rapidity of medical images, a random sample consensus (RANSAC) algorithm is useful to stitching the images of Chest digital radiography by scale-invariant feature transform (SIFT) and speeded-up robust features (SURF) feature extraction. Down-sampling utilizes to lessen the size of the images and reduction the measure of calculation. In the interim, the phase correlation is engaged to discover the overlapping region. After feature matching and perspective transformation, the stitched image is gotten dependent on the homography. At last, experimentation has finished showing the presentation.

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.

The great advances in deep learning enable human beings to extract the semantic information in ubiquitous unstructured data into high-dimensional vectors, leading to an urgent priority of managing vector data. Nearest neighbor (NN) query is a pivotal question in vector data management. Unfortunately, the NN query has been proved to be intrinsically hard. Researchers, thus, turn to approximate nearest neighbor (ANN) query which targets better efficiency by slightly compromising accuracy. Despite this, existing ANN methods still incur high time and space costs, which are mainly due to the high dimensionality. Specifically, in most in-memory ANN algorithms, the time cost is dominated by the time for comparing the distances among vectors. Each distance comparison operation (DCO) involves hundreds of arithmetic operations as each vector has hundreds of dimensions. Similarly, the space cost is dominated by the space for storing the vectors. Towards the goal of reducing the time and space costs, we present three methods in this thesis. First, we observe that, for a DCO, if an approximate distance can confirm that a vector is not a NN, then it can be discarded without computing exact distances. Based on this observation, we propose a method named ADSampling. For a DCO, it iteratively and incrementally samples a few dimensions and computes an approximate distance. When this distance is accurate enough to decide the result of the DCO, it terminates sampling. Second, to further improve the efficiency of DCOs, we design an efficient distance estimator. We propose a new quantization method named RaBitQ, which maps $D$-dimensional vectors into $D$-bit strings. RaBitQ provides an unbiased and asymptotically optimal estimator with good empirical accuracy. In addition, we introduce efficient implementations of RaBitQ based on bitwise operations or SIMD-based operations. Third, by extending the RaBitQ method, we develop an algorithm which compresses a floating-point vector into a $B$-bit unsigned integer vector. We prove that our algorithm is asymptotically optimal in terms of the space-accuracy trade-off.

When an operator teleoperates a rescue robot, it is necessary for the operator to understand the relationship between the robot and its environment while indirect-vision driving. A LRF (Laser Range Finder) installed on a rotating stage is often used for scanning the environment in order to give the operator spatial information, while there is a tradeoff between its scanning speed and point density. To solve the problem, we implement plane detection by RANSAC (RANdom SAmple Consensus) algorithm, integration of overlapped planes using tables and drawing them as convex hulls. The experimental results show that the scanning speed could be improved because the result of 3D scanning with a low point density shows a good match for that with a high point density by using the implemented methods.

This paper presents an approach to detect sharp edges for estimating point spread function (PSF) of a lens. A category of PSF estimation methods detect sharp edges from low-resolution (LR) images and estimate PSF with the detected edges. Existing techniques usually rely on accurate detection of ending points of the profile normal to an edge. In practice, however, it is often very difficult to localize profiles accurately. Inaccurately localized profiles generate a poor PSF estimation. We employ the Random Sample Consensus (RANSAC) algorithm to rule out outlier points. In RANSAC, prior knowledge about a pattern shape is incorporated, and the edge points lying far away from the pattern shape will be removed. The proposed method is tested on images of saddle patterns. Experimental results show that the proposed method can robustly localize sharp edges from LR saddle pattern images and yield accurate PSF estimation.