Felipe Lumbreras

Evaluation of methods for ridge and valley detection

Ridges and valleys are useful geometric features for image analysis. Different characterizations have been proposed to formalize the intuitive notion of ridge/valley. In this paper, we review their principal characterizations and propose a new one. Subsequently, we evaluate these characterizations with respect to a list of desirable properties and their purpose in the context of representative image analysis tasks.

3D vehicle sensor based on monocular vision

Determining the position of other vehicles on the road is a key information to help driver assistance systems to increase drivers safety. Accordingly, the work presented in this paper addresses the problem of detecting the vehicles in front of our own one and estimating their 3D position by using a single monochrome camera. Rather than using predefined high level image features as symmetry, shadow search, etc., our proposal for the vehicle detection is based on a learning process that determines, from a training set, which are the best features to distinguish vehicles from non-vehicles. To compute 3D information with a single camera a key point consists of knowing the position where the horizon projects onto the image. However, this position can change in every frame and is difficult to determine. In this paper we study the coupling between the perceived horizon and the actual width of vehicles in order to reduce the uncertainty in their estimated 3D position derived from an unknown horizon.

Multispectral Image Feature Points

This paper presents a novel feature point descriptor for the multispectral image case Far-Infrared and Visible Spectrum images. It allows matching interest points on images of the same scene but acquired in different spectral bands. Initially, points of interest are detected on both images through a SIFT-like based scale space representation. Then, these points are characterized using an Edge Oriented Histogram (EOH) descriptor. Finally, points of interest from multispectral images are matched by finding nearest couples using the information from the descriptor. The provided experimental results and comparisons with similar methods show both the validity of the proposed approach as well as the improvements it offers with respect to the current state-of-the-art.

Combining Priors, Appearance, and Context for Road Detection

Detecting the free road surface ahead of a moving vehicle is an important research topic in different areas of computer vision, such as autonomous driving or car collision warning. Current vision-based road detection methods are usually based solely on low-level features. Furthermore, they generally assume structured roads, road homogeneity, and uniform lighting conditions, constraining their applicability in real-world scenarios. In this paper, road priors and contextual information are introduced for road detection. First, we propose an algorithm to estimate road priors online using geographical information, providing relevant initial information about the road location. Then, contextual cues, including horizon lines, vanishing points, lane markings, 3-D scene layout, and road geometry, are used in addition to low-level cues derived from the appearance of roads. Finally, a generative model is used to combine these cues and priors, leading to a road detection method that is, to a large degree, robust to varying imaging conditions, road types, and scenarios.

Segmentation of petrographical images of marbles

Abstract Microscope images of marble thin sections can be used to determine their geographical origin by means of the shapes, sizes, and spatial distribution of their grains. In this paper, we present a method that is the first step towards an automatic origin determination, namely, the segmentation of grains in digital images of thin marble sections. Each grain has a preferred direction, different from the one of its neighbors, that rules its behavior when illuminated with polarized light. Firstly, we perform an oversegmentation of the image in regions, each one corresponding only to a grain, although this one can be partitioned into several regions. Afterwards, from a sequence of images of the same sample, obtained with polarized light, we compute for each region two parameters which depend on the preferred direction of the grain to which it belongs. Finally, the set of parameter values for all the regions is the input to a region-merging procedure which achieves the final segmentation. We present the results for samples from six quarries, each one with different visual features.

Creaseness from Level Set Extrinsic Curvature

Creases are a type of ridge/valley-like structures of a d dimensional image, characterized by local conditions. As creases tend to be at the center of anisotropic grey-level shapes, creaseness can be considered as a type of medialness. Among the several crease definitions, one of the most important is based on the extrema of the level set curvatures. In 2-d it is used the curvature of the level curves of the image landscape, however, the way it is usually computed produces a discontinuous creaseness measure. The same problem arises in 3-d with its straightforward extension and with other related creaseness measures. In this paper, we first present an alternative method of computing the level curve curvature that avoids the discontinuities. Next, we propose the Mean curvature of the level surfaces as creaseness measure of 3-d images, computed by the same method. Finally, we propose a natural extension of our first alternative method in order to enhance the creaseness measure.

Topological principal component analysis for face encoding and recognition

Abstract Principal component analysis (PCA)-like methods make use of an estimation of the covariances between sample variables. This estimation does not take into account their topological relationships. This paper proposes how to use these relationships in order to estimate the covariances in a more robust way. The new method topological principal component analysis (TPCA) is tested using both face encoding and recognition experiments showing how the generalization capabilities of PCA are improved.

Multispectral piecewise planar stereo using Manhattan-world assumption

This paper proposes a new framework for extracting dense disparity maps from a multispectral stereo rig. The system is constructed with an infrared and a color camera. It is intended to explore novel multispectral stereo matching approaches that will allow further extraction of semantic information. The proposed framework consists of three stages. Firstly, an initial sparse disparity map is generated by using a cost function based on feature matching in a multiresolution scheme. Then, by looking at the color image, a set of planar hypotheses is defined to describe the surfaces on the scene. Finally, the previous stages are combined by reformulating the disparity computation as a global minimization problem. The paper has two main contributions. The first contribution combines mutual information with a shape descriptor based on gradient in a multiresolution scheme. The second contribution, which is based on the Manhattan-world assumption, extracts a dense disparity representation using the graph cut algorithm. Experimental results in outdoor scenarios are provided showing the validity of the proposed framework.

Geographic information for vision-based road detection

Road detection is a vital task for the development of autonomous vehicles. The knowledge of the free road surface ahead of the target vehicle can be used for autonomous driving, road departure warning, as well as to support advanced driver assistance systems like vehicle or pedestrian detection. Using vision to detect the road has several advantages in front of other sensors: richness of features, easy integration, low cost or low power consumption. Common vision-based road detection approaches use low-level features (such as color or texture) as visual cues to group pixels exhibiting similar properties. However, it is difficult to foresee a perfect clustering algorithm since roads are in outdoor scenarios being imaged from a mobile platform. In this paper, we propose a novel high-level approach to vision-based road detection based on geographical information. The key idea of the algorithm is exploiting geographical information to provide a rough detection of the road. Then, this segmentation is refined at low-level using color information to provide the final result. The results presented show the validity of our approach.

Robust Lane Lines Detection and Quantitative Assessment

Detection of lane markings based on a camera sensor can be a low cost solution to lane departure and curve over speed warning. A number of methods and implementations have been reported in the literature. However, reliable detection is still an issue due to cast shadows, wearied and occluded markings, variable ambient lighting conditions etc. We focus on increasing the reliability of detection in two ways. Firstly, we employ a different image feature other than the commonly used edges: ridges, which we claim is better suited to this problem. Secondly, we have adapted RANSAC, a generic robust estimation method, to fit a parametric model of a pair or lane lines to the image features, based on both ridgeness and ridge orientation. In addition this fitting is performed for the left and right lane lines simultaneously, thus enforcing a consistent result. We have quantitatively assessed it on synthetic but realistic video sequences for which road geometry and vehicle trajectory ground truth are known.