Robust contour matching via the order-preserving assignment problem

A new test to detect impairments of sequential visuospatial memory due to lesions of the temporal lobe

This study investigates visuospatial memory in patients with unilateral lesions of the temporal lobe and the hippocampus resulting from surgery to treat drug-resistant epilepsy. To detect impairments of visuospatial memory in these individuals, a memory test should be specific to episodic memory, the type of memory in which the hippocampus is crucially involved. However, most known visuospatial memory tests do not focus on episodic memory. We hypothesized that a new sequential visuospatial memory test, which has been previously developed and applied only in healthy subjects, might be suitable to fill this gap. The test requires the subject to reproduce a memorized sequence of target locations in ordered recall by typing on a blank graphics tablet. The length of the memorized sequence extended successively after repeated presentation of a sequence of 20 target positions. The test was done twice on day one and again after one week. Visual working memory was tested with the Corsi block-tapping task. The performance in the new test was also related to the performance of the patients in the standard test battery of the neuropsychological examination in the clinical context. Thirteen patients and 14 controls participated. Patients showed reduced learning speed in the new sequential visuospatial memory task. Right-sided lesions induced stronger impairments than left-sided lesions. After one week, retention was reduced in the patients with left-sided lesions. The performance of the patients in commonly used tests of the neuropsychological standard battery did not differ compared to healthy subjects, whereas the new test allowed discrimination between patients and controls at a high correct-decision rate of 0.89. The Corsi block-span of the patients was slightly shorter than that of the controls. The results suggest that the new test provides a specific investigation of episodic visuospatial memory. Hemispheric asymmetries were consistent with the general hypothesis of right hemispheric dominance in visuospatial processing.

Plant Species Identification from Occluded Leaf Images

We present an approach to identify the plant species from the contour information from occluded leaf image using a database of full plant leaves. Although contour based 2D shape matching has been studied extensively in the last couple of decades, matching occluded leaves with full leaf databases is an open and little worked on problem. Classifying occluded plant leaves is even more challenging than full leaf matching because of large variations and complexity of leaf structures. Matching an occluded contour with all the full contours in a database is an NP-hard problem, so our algorithm is necessarily suboptimal. First, we represent the 2D contour points as a β-Spline curve. Then, we extract interest points on these curves via the Discrete Contour Evolution (DCE) algorithm. We use subgraph matching using the DCE points as graph nodes, which produces a number of open curves for each closed leaf contour. Next, we compute the similarity transformation parameters (translation, rotation, and uniform scaling) for each open curve. We then "overlay" each open curve with the inverse similarity transformed occluded curve and use the Fréchet distance metric to measure the quality of the match, retaining the best η matched curves. Since the Fréchet metric is cheap to compute but not perfectly correlated with the quality of the match, we formulate an energy functional that is well correlated with the quality of the match, but is considerably more expensive to compute. The functional uses local and global curvature, Shape Context descriptors, and String Cut features. We minimize this energy functional using a convex-concave relaxation framework. The curve among these best η curves, that has the minimum energy, is considered to be the best overall match with the occluded leaf. Experiments on three publicly available leaf image database shows that our method is both effective and efficient, outperforming other current state-of-the-art methods. Occlusion is measured as the percentage of the overall contour (and not leaf area) that is missing. We show that our algorithm can, even for leaves with a high amounts of occlusion (say 50 percent occlusion), still identify the best full leaf match from the databases.

Toward quantitative three-dimensional microvascular networks segmentation with multiview light-sheet fluorescence microscopy

Three-dimensional (3-D) large-scale imaging of microvascular networks is of interest in various areas of biology and medicine related to structural, functional, developmental, and pathological issues. Light-sheet fluorescence microscopy (LSFM) techniques are rapidly spreading and are now on the way to offer operational solutions for large-scale tissue imaging. This contribution describes how reliable vessel segmentation can be handled from LSFM data in very large tissue volumes using a suitable image analysis workflow. Since capillaries are tubular objects of a few microns scale radius, they represent challenging structures to reliably reconstruct without distortion and artifacts. We provide a systematic analysis of multiview deconvolution image processing workflow to control and evaluate the accuracy of the reconstructed vascular network using various low to high level, metrics. We show that even if low-level structural metrics are sensitive to isotropic imaging enhancement provided by a larger number of views, functional high-level metrics, including perfusion permeability, are less sensitive. Hence, combining deconvolution and registration onto a few number of views appears sufficient for a reliable quantitative 3-D vessel segmentation for their possible use for perfusion modeling.

Deformation Based Curved Shape Representation

In this paper, we introduce a deformation based representation space for curved shapes in . Given an ordered set of points sampled from a curved shape, the proposed method represents the set as an element of a finite dimensional matrix Lie group. Variation due to scale and location are filtered in a preprocessing stage, while shapes that vary only in rotation are identified by an equivalence relationship. The use of a finite dimensional matrix Lie group leads to a similarity metric with an explicit geodesic solution. Subsequently, we discuss some of the properties of the metric and its relationship with a deformation by least action. Furthermore, invariance to reparametrization or estimation of point correspondence between shapes is formulated as an estimation of sampling function. Thereafter, two possible approaches are presented to solve the point correspondence estimation problem. Finally, we propose an adaptation of k-means clustering for shape analysis in the proposed representation space. Experimental results show that the proposed representation is robust to uninformative cues, e.g., local shape perturbation and displacement. In comparison to state of the art methods, it achieves a high precision on the Swedish and the Flavia leaf datasets and a comparable result on MPEG-7, Kimia99 and Kimia216 datasets.

Angular pattern and binary angular pattern for shape retrieval

In this paper, we propose two novel shape descriptors, angular pattern (AP) and binary angular pattern (BAP), and a multiscale integration of them for shape retrieval. Both AP and BAP are intrinsically invariant to scale and rotation. More importantly, being global shape descriptors, the proposed shape descriptors are computationally very efficient, while possessing similar discriminability as state-of-the-art local descriptors. As a result, the proposed approach is attractive for real world shape retrieval applications. The experiments on the widely used MPEG-7 and TARI-1000 data sets demonstrate the effectiveness of the proposed method in comparison with existing methods.

Multiscale distance matrix for fast plant leaf recognition

In this brief, we propose a novel contour-based shape descriptor, called the multiscale distance matrix, to capture the shape geometry while being invariant to translation, rotation, scaling, and bilateral symmetry. The descriptor is further combined with a dimensionality reduction to improve its discriminative power. The proposed method avoids the time-consuming pointwise matching encountered in most of the previously used shape recognition algorithms. It is therefore fast and suitable for real-time applications. We applied the proposed method to the task of plan leaf recognition with experiments on two data sets, the Swedish Leaf data set and the ICL Leaf data set. The experimental results clearly demonstrate the effectiveness and efficiency of the proposed descriptor.

Circular Cone: a novel approach for protein ligand shape matching using modified PCA

Nowadays in modern medicine, computer modeling has already become one of key methods toward the discovery of new pharmaceuticals. And virtual screening is a necessary process for this discovery. In the procedure of virtual screening, shape matching is the first step to select ligands for binding protein. In the era of HTS (high throughput screening), a fast algorithm with good result is in demand. Many methods have been discovered to fulfill the requirement. Our method, called "Circular Cone", by finding principal axis, gives another way toward this problem. We use modified PCA (principal component analysis) to get the principal axis, around which the rotation is like whirling a cone. By using this method, the speed of giving score to a pocket and a ligand is very fast, while the accuracy is ordinary. So, the good speed and the general accuracy of our method present a good choice for HTS.

Rotation-invariant nonrigid point set matching in cluttered scenes

This paper addresses the problem of rotation-invariant nonrigid point set matching. The shape context (SC) feature descriptor is used because of its strong discriminative nature, whereas edges in the graphs constructed by point sets are used to determine the orientations of SCs. Similar to lengths or directions, oriented SCs constructed this way can be regarded as attributes of edges. By matching edges between two point sets, rotation invariance is achieved. Two novel ways of constructing graphs on a model point set are proposed, aiming at making the orientations of SCs as robust to disturbances as possible. The structures of these graphs facilitate the use of dynamic programming (DP) for optimization. The strong discriminative nature of SC, the special structure of the model graphs, and the global optimality of DP make our methods robust to various types of disturbances, particularly clutters. The extensive experiments on both synthetic and real data validated the robustness of the proposed methods to various types of disturbances. They can robustly detect the desired shapes in complex and highly cluttered scenes.

A new method to evaluate order and accuracy of inaccurately and incompletely reproduced movement sequences

Studying imitation learning of long sequences requires the evaluation of inaccurately and incompletely reproduced movement sequences. In order to evaluate the movement reproduction, it has to be assigned to the original stimulus. We developed an assignment algorithm that considers the Spatial Neighborhood and Order of reproduction (SNOA). To evaluate the features of this analysis it was applied to human performance during learning of long pointing sequences under two conditions: stimulus-guided reproduction with high spatial accuracy and imitation learning with low spatial accuracy. The results were compared with a simple assignment considering Spatial Neighborhood only (SNA) and with a Manual Assignment (MA). In the stimulus-guided reproduction the error measures did not differ between the algorithms. In contrast, with imitation learning, SNOA and MA generated higher estimates of order and omission errors than SNA. The results show that SNOA can be used to automatically quantify the similarity of both movement structure and metric information between long target sequences and inaccurate and incomplete movement reproductions.

Probabilistic Models of Object Geometry with Application to Grasping

Robot manipulators typically rely on complete knowledge of object geometry in order to plan motions and compute grasps. However, when an object is not fully in view it can be difficult to form an accurate estimate of the object's shape and pose, particularly when the object deforms. In this paper we describe a generative model of object geometry based on Mardia and Dryden's “Probabilistic Procrustean Shape”, which captures both non-rigid deformations and object variability in a class. We extend their shape model to the setting where point correspondences are unknown using Scott and Nowak's COPAP framework. We use this model to recognize objects in a cluttered image and to infer their complete two-dimensional boundaries with a novel algorithm called OSIRIS. We show examples of learned models from image data and demonstrate how the models can be used by a manipulation planner to grasp objects in cluttered visual scenes.

Automatic extraction of local axis of bone symmetry in CT images

We present an algorithm that extracts local axis of symmetry in CT images automatically. The proposed method combines registration and extraction of bone contours to generate the desired symmetry axis. The method consists of several stages: first extracting the bone contours of the images by using an active contour method, then finding grossly an axis that allows the division of the image into two parts, loosely called 'half' images, but with the understanding that they are not exactly the two halves of the image but rather the two halves of the depicted object. After that, finding a mapping that aligns the first half of the image with the second half and finally, finding the local axis of symmetry from corresponding contours.

Inter-subject comparison of MRI knee cartilage thickness

In this paper, we present the development and application of current image processing techniques to perform MRI inter-subject comparison of knee cartilage thickness based on the registration of bone structures. Each point in the bone surface which is part of the bone-cartilage interface is assigned a cartilage thickness value. Cartilage and corresponding bone structures are segmented and their shapes interpolated to create isotropic voxels. Cartilage thicknesses are computed for each point in the bone-cartilage interfaces and transferred to the bone surfaces. Corresponding anatomic points are then computed for bone surfaces based on shape matching using 3D shape descriptors called shape contexts to register bones with affine and elastic transformations, and then perform a point to point comparison of cartilage thickness values. An alternative technique for cartilage shape interpolation using a morphing technique is also presented. The cartilage segmentation and morphing were validated visually, based on volumetric measurements of porcine knee images which cartilage volumes were measured using a water displacement method, and based on digital thickness values computed with an established technique. Shape matching using 3D shape contexts was validated visually and against manual shape matching performed by a radiologist. The reproducibility of intra- and inter-subject cartilage thickness comparisons was established, as well as the feasibility of using the proposed technique to build a mean femoral shape, cartilage thickness map, and cartilage coverage map. Results showed that the proposed technique is robust, accurate, and reproducible to perform point to point inter-subject comparison of knee cartilage thickness values.