This paper presents Slime, a novel non-deep image matching framework which models the scene as rough local overlapping planes. This intermediate representation sits in-between the local affine approximation of the keypoint patches and the global matching based on both spatial and similarity constraints, providing a progressive pruning of the correspondences, as planes are easier to handle with respect to general scenes. Slime decomposes the images into overlapping regions at different scales and computes loose planar homographies. Planes are mutually extended by compatible matches and the images are split into fixed tiles, with only the best homographies retained for each pair of tiles. Stable matches are identified according to the consensus of the admissible stereo configurations provided by pairwise homographies. Within tiles, the rough planes are then merged according to their overlap in terms of matches and further consistent correspondences are extracted. The whole process only involves homography constraints. As a result, both the coverage and the stability of correct matches over the scene are amplified, together with the ability to spot matches in challenging scenes, allowing traditional hybrid matching pipelines to make up lost ground against recent end-to-end deep matching methods. In addition, the paper gives a thorough comparative analysis of recent state-of-the-art in image matching represented by end-to-end deep networks and hybrid pipelines. The evaluation considers both planar and non-planar scenes, taking into account critical and challenging scenarios including abrupt temporal image changes and strong variations in relative image rotations. According to this analysis, although the impressive progress done in this field, there is still a wide room for improvements to be investigated in future research.

This study introduces an extension of the sGLOH2 local image descriptor inspired by RootSIFT ‘square rooting’ as a way to indirectly alter the matching distance used to compare the descriptor vectors. The extended descriptor, named RootsGLOH2, achieved the best results in terms of matching accuracy and robustness among the latest state‐of‐the‐art non‐deep descriptors in recent evaluation contests dealing with both planar and non‐planar scenes. RootsGLOH2 also achieves a matching accuracy very close to that obtained by the best deep descriptors to date. Beside confirming that ‘square rooting’ has beneficial effects on sGLOH2 as it happens on SIFT, experimental evidence shows that classical norm‐based distances, such as the Euclidean and Manhattan distances, only provide suboptimal solutions to the problem of local image descriptor matching. This suggests matching distance design as a topic to investigate further in the near future.

This paper proposes a novel approach for registering the PRNU pattern between different camera acquisition modes by relying on the imaged scene content. First, images are aligned by establishing correspondences between local descriptors: The result can then optionally be refined by maximizing the PRNU correlation. Comparative evaluations show that this approach outperforms those based on brute-force and particle swarm optimization in terms of reliability, accuracy and speed. The proposed scene-based approach for PRNU pattern alignment is suitable for video source identification in multimedia forensics applications.

sGLOH (shifting GLOH) is a histogram-based keypoint descriptor that can be associated to multiple quantized rotations of the keypoint patch without any recomputation. This property can be exploited to define the best distance between two descriptor vectors, thus avoiding computing the dominant orientation. In addition, sGLOH can reject incongruous correspondences by adding a global constraint on the rotations either as an a priori knowledge or based on the data. This paper thoroughly reconsiders sGLOH and improves it in terms of robustness, speed and descriptor dimension. The revised sGLOH embeds more quantized rotations, thus yielding more correct matches. A novel fast matching scheme is also designed, which significantly reduces both computation time and memory usage. In addition, a new binarization technique based on comparisons inside each descriptor histogram is defined, yielding a more compact, faster, yet robust alternative. Results on an exhaustive comparative experimental evaluation show that the revised sGLOH descriptor incorporating the above ideas and combining them according to task requirements, improves in most cases the state of the art in both image matching and object recognition.

This paper introduces a new compositional framework for classifying color correction methods according to their two main computational units. The framework was used to dissect fifteen among the best color correction algorithms and the computational units so derived, with the addition of four new units specifically designed for this work, were then reassembled in a combinatorial way to originate about one hundred distinct color correction methods, most of which never considered before. The above color correction methods were tested on three different existing datasets, including both real and artificial color transformations, plus a novel dataset of real image pairs categorized according to the kind of color alterations induced by specific acquisition setups. Differently from previous evaluations, special emphasis was given to effectiveness in real world applications, such as image mosaicing and stitching, where robustness with respect to strong image misalignments and light scattering effects is required. Experimental evidence is provided for the first time in terms of the most recent perceptual image quality metrics, which are known to be the closest to human judgment. Comparative results show that combinations of the new computational units are the most effective for real stitching scenarios, regardless of the specific source of color alteration. On the other hand, in the case of accurate image alignment and artificial color alterations, the best performing methods either use one of the new computational units, or are made up of fresh combinations of existing units.

We aim at describing a non-parametric approach to evaluate blood cells velocity in oral capillascopic videos. The proposed methodology is based on the application of standard optical flow algorithms and it is part of a general environment to support during the diagnostic process for evaluating peripheral microcirculation in real time. We validated our approach versus handmade measurements provided by physicians. Results on real data pointed out that our system returns an output coherent to these latter observations.

Accurate retinal image registration is essential to track the evolution of eye-related diseases. We propose a semiautomatic method based on features relying upon retinal graphs for temporal registration of retinal images. The features represent straight lines connecting vascular landmarks on the retina vascular tree: bifurcations, branchings, crossings, end points. In the built retinal graph, one straight line between two vascular landmarks indicates that they are connected by a vascular segment in the original retinal image. The locations of the landmarks are manually extracted to avoid the information loss due to errors in a retinal vessels segmentation algorithms. A straight line model is designed to compute a similarity measure to quantify the line matching between images. From the set of matching lines, corresponding points are extracted and a global transformation is computed. The performance of the registration method is evaluated in the absence of ground truth using the cumulative inverse consistency error (CICE).

The estimation of the fundamental matrix from a set of corresponding points is a relevant topic in epipolar stereo geometry [10].Due to the high amount of outliers between the matches, RANSAC-based approaches [7,13,29] have been used to obtain the fundamental matrix.In this paper two new contributes are presented: a new normalized epipolar error measure which takes into account the shape of the features used as matches [17] and a new strategy to compare fundamental matrices.The proposed error measure gives good results and it does not depend on the image scale.Moreover, the new evaluation strategy describes a valid tool to compare different RANSAC-based methods because it does not rely on the inlier ratio, which could not correspond to the best allowable fundamental matrix estimated model, but it makes use of a reference ground truth fundamental matrix obtained by a set of corresponding points given by the user.

Image descriptors are widely adopted structures to match image features. SIFT-based descriptors are collections of gradient orientation histograms computed on different feature regions, commonly divided by using a regular Cartesian grid or a log-polar grid. In order to achieve rotation invariance, feature patches have to be generally rotated in the direction of the dominant gradient orientation. In this paper we present a modification of the GLOH descriptor, a SIFT-based descriptor based on a log-polar grid, which avoids to rotate the feature patch before computing the descriptor since predefined discrete orientations can be easily derived by shifting the descriptor vector. The proposed descriptors, called sGLOH and sGLOH+, have been compared with the SIFT descriptor on the Oxford image dataset, with good results which point out its robustness and stability.

A parallel version of a new automatic Harris-based corner detector is presented. A scheduler to dynamically and homogeneously distribute high computational workload on heterogeneous parallel architectures such as Grid systems has been implemented to speedup the whole procedure. Experimental results show the robustness of the underlying scheduler, which can be easily exploited in various automatic image analysis systems.

This paper introduces a new Harris-affine corner detector algorithm, that does not need parameters to locate corners in images, given an observation scale. Standard detectors require to fine tune the values of parameters which strictly depend on the particular input image. A quantitative comparison between our implementation and a standard Harris-affine implementation provides good results, showing that the proposed methodology is robust and accurate. The benchmark consists of public images used in literature for feature detection.