Showing papers on "Distance transform published in 2015"

Shading-based refinement on volumetric signed distance functions

Abstract: We present a novel method to obtain fine-scale detail in 3D reconstructions generated with low-budget RGB-D cameras or other commodity scanning devices. As the depth data of these sensors is noisy, truncated signed distance fields are typically used to regularize out the noise, which unfortunately leads to over-smoothed results. In our approach, we leverage RGB data to refine these reconstructions through shading cues, as color input is typically of much higher resolution than the depth data. As a result, we obtain reconstructions with high geometric detail, far beyond the depth resolution of the camera itself. Our core contribution is shading-based refinement directly on the implicit surface representation, which is generated from globally-aligned RGB-D images. We formulate the inverse shading problem on the volumetric distance field, and present a novel objective function which jointly optimizes for fine-scale surface geometry and spatially-varying surface reflectance. In order to enable the efficient reconstruction of sub-millimeter detail, we store and process our surface using a sparse voxel hashing scheme which we augment by introducing a grid hierarchy. A tailored GPU-based Gauss-Newton solver enables us to refine large shape models to previously unseen resolution within only a few seconds.

Boundary Preserving Dense Local Regions

Abstract: We propose a dense local region detector to extract features suitable for image matching and object recognition tasks. Whereas traditional local interest operators rely on repeatable structures that often cross object boundaries (e.g., corners, scale-space blobs), our sampling strategy is driven by segmentation, and thus preserves object boundaries and shape. At the same time, whereas existing region-based representations are sensitive to segmentation parameters and object deformations, our novel approach to robustly sample dense sites and determine their connectivity offers better repeatability. In extensive experiments, we find that the proposed region detector provides significantly better repeatability and localization accuracy for object matching compared to an array of existing feature detectors. In addition, we show our regions lead to excellent results on two benchmark tasks that require good feature matching: weakly supervised foreground discovery and nearest neighbor-based object recognition.

A fast algorithm for computation of discrete Euclidean distance transform in three or more dimensions on vector processing architectures

Abstract: In this note, we introduce a function for calculating Euclidean distance transform in large binary images of dimension three or higher in Matlab. This function uses transparent and fast line-scan algorithm that can be efficiently implemented on vector processing architectures such as Matlab and significantly outperforms the Matlab’s standard distance transform function “bwdist” both in terms of the computation time and the possible data sizes. The described function also can be used to calculate the distance transform of the data with anisotropic voxel aspect ratios. These advantages make this function especially useful for high-performance scientific and engineering applications that require distance transform calculations for large multidimensional and/or anisotropic datasets in Matlab. The described function is publicly available from the Matlab Central website under the name “bwdistsc”, “Euclidean Distance Transform for Variable Data Aspect Ratio”.

GDFuzz3D: a method for protein 3D structure reconstruction from contact maps, based on a non-Euclidean distance function

Abstract: MOTIVATION To date, only a few distinct successful approaches have been introduced to reconstruct a protein 3D structure from a map of contacts between its amino acid residues (a 2D contact map). Current algorithms can infer structures from information-rich contact maps that contain a limited fraction of erroneous predictions. However, it is difficult to reconstruct 3D structures from predicted contact maps that usually contain a high fraction of false contacts. RESULTS We describe a new, multi-step protocol that predicts protein 3D structures from the predicted contact maps. The method is based on a novel distance function acting on a fuzzy residue proximity graph, which predicts a 2D distance map from a 2D predicted contact map. The application of a Multi-Dimensional Scaling algorithm transforms that predicted 2D distance map into a coarse 3D model, which is further refined by typical modeling programs into an all-atom representation. We tested our approach on contact maps predicted de novo by MULTICOM, the top contact map predictor according to CASP10. We show that our method outperforms FT-COMAR, the state-of-the-art method for 3D structure reconstruction from 2D maps. For all predicted 2D contact maps of relatively low sensitivity (60-84%), GDFuzz3D generates more accurate 3D models, with the average improvement of 4.87 A in terms of RMSD. AVAILABILITY AND IMPLEMENTATION GDFuzz3D server and standalone version are freely available at http://iimcb.genesilico.pl/gdserver/GDFuzz3D/. CONTACT iamb@genesilico.pl SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

Automatic extraction of blood vessels in the retinal vascular tree using multiscale medialness

Abstract: We propose an algorithm for vessel extraction in retinal images. The first step consists of applying anisotropic diffusion filtering in the initial vessel network in order to restore disconnected vessel lines and eliminate noisy lines. In the second step, a multiscale line-tracking procedure allows detecting all vessels having similar dimensions at a chosen scale. Computing the individual image maps requires different steps. First, a number of points are preselected using the eigenvalues of the Hessian matrix. These points are expected to be near to a vessel axis. Then, for each preselected point, the response map is computed from gradient information of the image at the current scale. Finally, the multiscale image map is derived after combining the individual image maps at different scales (sizes). Two publicly available datasets have been used to test the performance of the suggested method. The main dataset is the STARE project's dataset and the second one is the DRIVE dataset. The experimental results, applied on the STARE dataset, show a maximum accuracy average of around 94.02%. Also, when performed on the DRIVE database, the maximum accuracy average reaches 91.55%.

Multi-feature based automatic face identification on kernel eigen spaces (KES) under unstable lighting conditions

Abstract: Programmed Face Identification (PFI) of images more reliable even under unstable lighting conditions is one of the most important challenges for practical face recognition systems. We tackle this by combining the strengths of robust illumination normalization, local texture-based face representations, distance transform based matching, multiple feature fusion. Additionally we propose Phase Congruency Features which is an approach for detects points of order in the phase spectrum within images. By combining the results produced from both the above mentioned approaches, we try to increase the accuracy of recognizing the given image from group images even with difficult lighting conditions.

Assessment of similarity measurement criteria in Thepade's Sorted Ternary Block Truncation Coding (TSTBTC) for Content Based Video Retrieval

Abstract: Rapid growth in Information technology and Communication networking, have increased the inclination of professionals in storage and archival of multimedia-video data. Efficient and accurate retrieval of archived video data is essential need of many professional groups like researchers, analyst, journalist and historians. Textual metadata based video retrieval is intuitive and subject to human perception. Thus researchers are exploiting the opportunity of extracting the features based on content of video and retrieve the video accordingly. There rises a new horizon of Content Based Video Retrieval. Color content based video retrieval has been attempted using Block Truncation Coding and variations of it. Thepade's Sorted Ternary Block Truncation Coding level 2 (TSTBTC-2) has been attempted for Color Content Based Video Retrieval in this paper. The paper attempts detailed performance comparison of sixteen similarity measures alias Euclidean Distance, City Block Metric, Chebychev Distance, Mean Square Error, Hamming Distance, Minkowski Distance, Soergel Distance, Sorensen Distance, Canberra Distance, Kulczynski Distance, Wave Hedges Distance, Harmonic Mean Distance, Cosine Similarity, Jaccard Distance, Fidelity Distance and Squared Chord Distance across five families L p Minkowski family, L 1 family, Intersection family, Inner Product Family and Fidelity family. TSTBTC-2 based video retrieval experimentation is done on test bed of 500 videos with 500 queries and the average accuracy is computed. After experimentation it has been found that L p Minkowski family performs better and fidelity family's similarity measures are not suitable for applications like Color Content Based Video Retrieval. Among all similarity measures considered Sorensen Distance shows best performance followed by City Block Metric measure.

D$$^{2}$$CO: Fast and Robust Registration of 3D Textureless Objects Using the Directional Chamfer Distance

Abstract: This paper introduces a robust and efficient vision based method for object detection and 3D pose estimation that exploits a novel edge-based registration algorithm we called Direct Directional Chamfer Optimization D$$^{2}$$CO. Our approach is able to handle textureless and partially occluded objects and does not require any off-line object learning step. Depth edges and visible patterns extracted from the 3D CAD model of the object are matched against edges detected in the current grey level image by means of a 3D distance transform represented by an image tensor, that encodes the minimum distance to an edge point in a joint direction/location space. D$$^{2}$$CO refines the object position employing a non-linear optimization procedure, where the cost being minimized is extracted directly from the 3D image tensor. Differently from other popular registration algorithms as ICP, that require to constantly update the correspondences between points, our approach does not require any iterative re-association step: the data association is implicitly optimized while inferring the object position. This enables D$$^{2}$$CO to obtain a considerable gain in speed over other registration algorithms while presenting a wider basin of convergence. We tested our system with a set of challenging untextured objects in presence of occlusions and cluttered background, showing accurate results and often outperforming other state-of-the-art methods.

An Improvement Stereo Vision Images Processing for Object Distance Measurement

Abstract: Human has the ability to roughly estimate the distance of objects because of the stereo vision of human’s eyes. In this paper we proposed an improved stereo vision system to accurately measure the distance of objects in real world. Object distance is very useful for obstacle avoidance and navigation of autonomous vehicles. Recent researches have used stereo cameras for different applications such as 3D image construction, distance measurement, and occlusion detection. The proposed measurement procedure is a three-phase process: object detection, segmentation, and distance calculation. In distance calculation, we proposed a new algorithm to reduce the error. The result shows our measurement system is capable of providing objects distance with less than 5% of measurement error.

Processing apparatus, processing system, processing program, and processing method

Abstract: A processing apparatus includes a distance image acquirer, a moving-object detector, a background recognizer. The distance image acquirer acquires a distance image containing distance information of each pixel. The moving-object detector detects a moving object from the distance image. The background recognizer generates a background model in which a stationary object recognized as background of the moving object is modeled, from the distance image acquired by the distance image acquirer. The moving-object detector changes a method for detecting the moving object, based on a relative positional relation between the moving object and the background model.

BING++: A Fast High Quality Object Proposal Generator at 100fps

Abstract: We are motivated by the need for an object proposal generation algorithm that achieves a good balance between proposal localization quality, object recall and computational efficiency. We propose a novel object proposal algorithm {\em BING++} which inherits the good computational efficiency of BING \cite{BingObj2014} but significantly improves its proposal localization quality. Central to our success is based on the observation that good bounding boxes are those that tightly cover objects. Edge features, which can be computed efficiently, play a critical role in this context. We propose a new algorithm that recursively improves BING's proposals by exploiting the fact that edges in images are typically associated with object boundaries. BING++ improves proposals recursively by incorporating nearest edge points (to proposal boundary pixels) to obtain a tighter bounding box. This operation has linear computational complexity in number of pixels and can be done efficiently using distance transform. Superpixel merging techniques are then employed as post-processing to further improve the proposal quality. Empirically on the VOC2007 dataset, using $10^3$ proposals and IoU threshold 0.5, our method achieves 95.3\% object detection recall (DR), 79.2\% mean average best overlap (MABO), and 68.7\% mean average precision (mAP) on object detection over 20 object classes within an average time of {\bf 0.009} seconds per image.

A novel 3D skeleton algorithm based on neutrosophic cost function

Abstract: This paper proposed a novel algorithm to extract the skeleton for the objects on three dimensional images with or without noise.Neutrosophic cost function is proposed based on neutrosophic set.Neutrosophic cost function is employed to define the cost between each point on skeleton. A skeleton provides a synthetic and thin representation of three dimensional objects, and is useful for shape description and recognition. In this paper, a novel 3D skeleton algorithm is proposed based on neutrosophic cost function. Firstly, the distance transform is used to a 3D volume, and the distance matrix is obtained for each voxel in the volume. The ridge points are identified based on their distance transform values and are used as the candidates for the skeleton. Then, a novel cost function, namely neutrosophic cost function (NCF) is proposed based on neutrosophic set, and is utilized to define the cost between each ridge points. Finally, a shortest path finding algorithm is used to identify the optimum path in the 3D volume with least cost, in which the costs of paths are calculated using the new defined NCF. The optimum path is treated as the skeleton of the 3D volume. A variety of experiments have been conducted on different 3D volume. The experimental results demonstrate the better performance of the proposed method. It can identify the skeleton for different volumes with high accuracy. In addition, the proposed method is robust to the noise on the volume. This advantage will lead it to wide application in the skeleton detection applications in the real world.

Sub-Riemannian Fast Marching in SE(2)

Abstract: We propose a Fast Marching based implementation for computing sub-Riemanninan (SR) geodesics in the roto-translation group SE(2), with a metric depending on a cost induced by the image data. The key ingredient is a Riemannian approximation of the SR-metric. Then, a state of the art Fast Marching solver that is able to deal with extreme anisotropies is used to compute a SR-distance map as the solution of a corresponding eikonal equation. Subsequent backtracking on the distance map gives the geodesics. To validate the method, we consider the uniform cost case in which exact formulas for SR-geodesics are known and we show remarkable accuracy of the numerically computed SR-spheres. We also show a dramatic decrease in computational time with respect to a previous PDE-based iterative approach. Regarding image analysis applications, we show the potential of considering these data adaptive geodesics for a fully automated retinal vessel tree segmentation.

Sub-Riemannian Fast Marching in SE(2)

Abstract: We propose a Fast Marching based implementation for computing sub-Riemanninan (SR) geodesics in the roto-translation group SE(2), with a metric depending on a cost induced by the image data. The key ingredient is a Riemannian approximation of the SR-metric. Then, a state of the art Fast Marching solver that is able to deal with extreme anisotropies is used to compute a SR-distance map as the solution of a corresponding eikonal equation. Subsequent backtracking on the distance map gives the geodesics. To validate the method, we consider the uniform cost case in which exact formulas for SR-geodesics are known and we show remarkable accuracy of the numerically computed SR-spheres. We also show a dramatic decrease in computational time with respect to a previous PDE-based iterative approach. Regarding image analysis applications, we show the potential of considering these data adaptive geodesics for a fully automated retinal vessel tree segmentation.

Hand tracking by extending distance transform and hand model in real-time

Abstract: Tracking a human's hand is not a trivial task. This paper contributes a new approach for hand tracking based on distance transform (DT) and edge points in real-time. In the beginning, we create a hand model geometrically in three dimensions. It is done by utilizing shortened quadrics. After that, the degrees of freedom, shortly called as DOF, for every joint angle correspond to each DOF to use in the later process. The edge likelihood is used for the feature extraction. A Bayesian classifier is utilized adaptively and accurately for the silhouette likelihood. For this reason, it is to cope greatly with any environmental changes visibly. By using these techniques, this method can be performed in real-time. Experimental results are provided.

2d image analyzer

Abstract: The invention relates to a 2D image analyzer comprising an image scaler, an image generator, and a pattern finder. The image scaler is designed to scale an image according to a scaling factor. The image generator is designed to generate an overview image which has a plurality of copies of the received and scaled image. Each copy is scaled by a different scaling factor. In the process, the respective position can be calculated using an algorithm which takes into consideration a distance between the scaled images in the overview image, a distance from the scaled images to one or more of the boundaries of the overview image, and/or other predefined conditions. The pattern finder is designed to carry out a characteristic transformation and classification of the overview image and to output a position where the searched pattern maximally matches the specified pattern. Optionally, a post-processing device can also be provided for smoothing and correcting the position of local maxima in the classified overview image.

Evaluation of Euclidean and Manhanttan metrics in Content Based Image Retrieval system

Abstract: Content-Based Image Retrieval is used nowadays for generating signatures of images in databases and then comparing these stored signatures with the signature of the query image. In this paper color histogram is used as signature of an image and used to compare two images based on Manhattan distance (L1 norm) and Euclidean distance (L2 norm) distance metrics‥In this paper, Corel database is used to evaluate the performance of Manhattan and Euclidean distance metrics. The experimental results showed that Manhattan showed better precision rate than Euclidean distance metric. The evaluation is made using Content based image retrieval application developed using color moments of the Hue, Saturation and Value(HSV) of the image and Gabor descriptors are adopted as texture features.

Image processing device, inspection device, image processing method, and non-transitory recording medium

Abstract: An image processing device includes an acquisitor and a processor. The acquisitor acquires information including a first image of an object and a first relationship. The first image includes first and second image areas. The first image area corresponds to a first position of a first part of the object and has a first pixel value. The second image area corresponds to a second position of a second part of the object and has a second pixel value. The processor performs generating a first generated image including a third and a fourth image areas corresponding to the first and the second positions, determining a third pixel value of the third image area and a fourth pixel value of the fourth image area, and calculating a first distance corresponding to a length of the third image area and a second distance corresponding to a length of the fourth image area.

Constrained distance transforms for spatial atlas registration

Abstract: Spatial frameworks are used to capture organ or whole organism image data in biomedical research. The registration of large biomedical volumetric images is a complex and challenging task, but one that is required for spatially mapped biomedical atlas systems. In most biomedical applications the transforms required are non-rigid and may involve significant deformation relating to variation in pose, natural variation and mutation. Here we develop a new technique to establish such transformations for mapping data that cannot be achieved by existing approaches and that can be used interactively for expert editorial review. This paper presents the Constrained Distance Transform (CDT), a novel method for interactive image registration. The CDT uses radial basis function transforms with distances constrained to geodesics within the domains of the objects being registered. A geodesic distance algorithm is discussed and evaluated. Examples of registration using the CDT are presented. The CDT method is shown to be capable of simultaneous registration and foreground segmentation even when very large deformations are required.

Interactive partial 3D shape matching with geometric distance optimization

Abstract: In this paper, we propose an efficient method for partial 3D shape matching based on minimizing the geometric distance between the source and the target geometry. Unlike existing methods, our method does not use a feature-based distance in order to obtain a matching score. Instead, we use a fast, GPU-based method to approximate the true geometric distance between the source and the target by rendering the source object into a distance field which was built around the target. This function behaves smoothly in the space of transformations and allows for an efficient gradient-based local optimization. In order to overcome local minima, we use single point correspondences between surface points on the source and the target respectively employing simple, yet efficient local features based on the distribution of normal vectors around a reference point. The best correspondences define starting positions for a local optimization. The high efficiency of the distance computation allows for robust determination of the global minima in less than a second, which makes our method usable in interactive applications. Our method works for any kind of input data since it only requires point data with normal information at each point. We also demonstrate the capability of our algorithm to perform global alignment of similar 3D objects.

A Neural Model of Distance-Dependent Percept of Object Size Constancy.

Abstract: Size constancy is one of the well-known visual phenomena that demonstrates perceptual stability to account for the effect of viewing distance on retinal image size. Although theories involving distance scaling to achieve size constancy have flourished based on psychophysical studies, its underlying neural mechanisms remain unknown. Single cell recordings show that distance-dependent size tuned cells are common along the ventral stream, originating from V1, V2, and V4 leading to IT. In addition, recent research employing fMRI demonstrates that an object’s perceived size, associated with its perceived egocentric distance, modulates its retinotopic representation in V1. These results suggest that V1 contributes to size constancy, and its activity is possibly regulated by feedback of distance information from other brain areas. Here, we propose a neural model based on these findings. First, we construct an egocentric distance map in LIP by integrating horizontal disparity and vergence through gain-modulated MT neurons. Second, LIP neurons send modulatory feedback of distance information to size tuned cells in V1, resulting in a spread of V1 cortical activity. This process provides V1 with distance-dependent size representations. The model supports that size constancy is preserved by scaling retinal image size to compensate for changes in perceived distance, and suggests a possible neural circuit capable of implementing this process.

Determining distance between an object and a capture device based on captured image data

Abstract: In various embodiments, methods, systems, and computer program products for determining distance between an object and a capture device are disclosed. The distance determination techniques are based on image data captured by the capture device, where the image data represent the object. These techniques improve the function of capture devices such as mobile phones by enabling determination of distance using a single lens capture device, and based on intrinsic parameters of the capture device, such as focal length and scaling factor(s), in preferred approaches. In some approaches, the distance estimation may be based in part on a priori knowledge regarding size of the object represented in the image data. Distance determination may be based on a homography transform and/or reference image data representing the object, a same type or similar type of object, in more approaches.

Image processing apparatus, image capturing apparatus, image processing method, image capturing method, and non-transitory computer-readable medium for focus bracketing

Abstract: An image processing apparatus includes an obtaining unit and an arranging unit. The obtaining unit obtains a first image, a second image, and a third image. The first image is an image captured with the focus set at a first distance. The second image is an image captured with the focus set at a second distance different from the first distance. The third image is an image captured using an f-number larger than an f-number used in capturing of the first image and the second image. The arranging unit arranges at least a part of the first image and at least a part of the second image at a position based on the third image.

Tumor boundary detection in ultrasound imagery using multi-scale generalized gradient vector flow

Abstract: As a key technology in high-intensity focused ultrasound (HIFU) ablation systems, a precise ultrasound image segmentation method for tumor boundary detection is helpful for ablation of tumors and avoiding tumor recurrence. This study explores a new deformable snake model called multi-scale generalized gradient vector flow (MS-GGVF) to segment ultrasound images in HIFU ablation. The main idea of the technique is dealing with two issues including spurious boundary attenuation and setting the standard deviation of the Gaussian filter. We assign the standard deviation as scales to build the MS-GGVF model and create a signed distance map to use its gradient direction information and magnitude information to refine the multi-scale edge map by attenuating spurious boundaries and highlighting the real boundary. In addition, a fast generalized gradient vector flow computation algorithm based on an augmented Lagrangian method is introduced to calculate the external force vector field to improve the computation efficiency of our model. The experimental segmentations were similar to the ground truths delineated by two medical physicians with high area overlap measure and low mean contour distance. The experimental results demonstrate that the proposed algorithm is robust, reliable, and precise for tumor boundary detection in HIFU ablation systems.

Scalable Parallel Distance Field Construction for Large-Scale Applications

Abstract: Computing distance fields is fundamental to many scientific and engineering applications. Distance fields can be used to direct analysis and reduce data. In this paper, we present a highly scalable method for computing 3D distance fields on massively parallel distributed-memory machines. A new distributed spatial data structure, named parallel distance tree , is introduced to manage the level sets of data and facilitate surface tracking over time, resulting in significantly reduced computation and communication costs for calculating the distance to the surface of interest from any spatial locations. Our method supports several data types and distance metrics from real-world applications. We demonstrate its efficiency and scalability on state-of-the-art supercomputers using both large-scale volume datasets and surface models. We also demonstrate in-situ distance field computation on dynamic turbulent flame surfaces for a petascale combustion simulation. Our work greatly extends the usability of distance fields for demanding applications.

Convolution Filtering of Continuous Signed Distance Fields for Polygonal Meshes

Abstract: Signed distance fields obtained from polygonal meshes are commonly used in various applications. However, they can have C1 discontinuities causing creases to appear when applying operations such as blending or metamorphosis. The focus of this work is to efficiently evaluate the signed distance function and to apply a smoothing filter to it while preserving the shape of the initial mesh. The resulting function is smooth almost everywhere, while preserving the exact shape of the polygonal mesh. Due to its low complexity, the proposed filtering technique remains fast compared to its main alternatives providing C1-continuous distance field approximation. Several applications are presented such as blending, metamorphosis and heterogeneous modelling with polygonal meshes.

A method of complex background estimation in astronomical images

Abstract: In this paper, we present a novel approach to the estimation of strongly varying backgrounds in astronomical images by means of small objects removal and subsequent missing pixels interpolation. The method is based on the analysis of a pixel local neighborhood and utilizes the morphological distance transform. In contrast to popular background estimation techniques, our algorithm allows for accurate extraction of complex structures, like galaxies or nebulae. Moreover, it does not require multiple tuning parameters, since it relies on physical properties of CCD image sensors - the gain and the read-out noise characteristics. The comparison with other widely used background estimators revealed higher accuracy of the proposed technique. The superiority of the novel method is especially significant for the most challenging fluctuating backgrounds. The size of filtered out objects is tunable, therefore the algorithm may eliminate a wide range of foreground structures, including the dark current impulses, cosmic rays or even entire galaxies in deep field images.