Cristina Maduro

Improved adaptive complex diffusion despeckling filter

Despeckling optical coherence tomograms from the human retina is a fundamental step to a better diagnosis or as a preprocessing stage for retinal layer segmentation. Both of these applications are particularly important in monitoring the progression of retinal disorders. In this study we propose a new formulation for a well-known nonlinear complex diffusion filter. A regularization factor is now made to be dependent on data, and the process itself is now an adaptive one. Experimental results making use of synthetic data show the good performance of the proposed formulation by achieving better quantitative results and increasing computation speed.

Estimation of vehicle velocity and traffic intensity using rectified images

This paper presents an automatic method to estimate vehicle velocity and determine the number of vehicles per lane using rectified images. This approach requires the knowledge of two lengths on the ground plane and can be applied to highway scenarios that possess fairly straight lanes in areas near the camera. A scale factor is determined, in order to ascertain vehicle velocity in the sought units. This scale factor associates distances traveled on the image plane with distances on the ground plane and can be obtained by identifying the period and location of the striped highway lanes on the rectified background image. Once the previously referred lines are located it is then possible to locate the highway lane boundaries and, therefore, determine the mean vehicle velocity of each lane.

Synthetic OCT data for image processing performance testing

The use of synthetic images is needed for testing the performance of image processing methods in order to establish a ground truth to test performance metrics. However, these synthetic images do not represent real applications. The aim of this paper is to build a mathematical model to obtain a synthetic noise-free image mimicking a real Optical Coherence Tomography (OCT) B-scan or volume from the human retina, in order to establish a ground truth for filtering performance metrics in this context. Moreover we also suggest a method to add speckle noise to this image based on the speckle noise of the given OCT volume. In this way we establish a repli-cable method to obtain a ground truth for image processing performance metrics that actually mimics a real case.

Ground plane velocity estimation embedding rectification on a particle filter multi-target tracking

This paper presents an integrated solution for vehicles velocity estimation and vehicle counting. The proposed restores the scene geometric properties, building a ground plane rectified image. Moreover, multiple vehicles tracking is performed embedding the concept of region covariance descriptors in a particle filter framework. The results show the effectiveness of the approach here proposed in very clutter scenes.

OCT Noise Despeckling Using 3D Nonlinear Complex Diffusion Filter

An improved despeckling method, based on complex diffusion filtering, is herein presented to enhance structure segmentation in high-definition spectral domain optical coherence tomography (OCT) data. We propose to extend the traditional nonlinear complex diffusion filter concept propose by Gilboa IEEE Trans Pattern Anal Mach Intell 26:1020–1036, 2004) from 2- to 3-dimensions, taking into account the consistency of noise along the entire 3D data volume. Moreover we also propose the extension to 3D of an improved complex diffusion filter (Bernardes et al. Opt Express 18:24,048–24,059, 2010), that was specially built for retinal tissue signal preservation in OCT data and that takes into account an adaptive optimized time step for the finite difference discretization. The extension to 3D of the traditional method compares favorably to existing methods reducing speckle noise and preserving edges and features. As expected, the improved 3D version has better performance than the traditional one. Numerical simulations show the feasibility of the method.

Estimating Vehicle Velocity Using Image Profiles on Rectified Images

In this paper a technique is presented to estimate vehicle velocity using intensity profiles. This technique does not require background estimation or even the identification and tracking of individual vehicles. On the other hand, it requires the estimation of virtual images that represent a bird eye view of the scenario. This is achieved estimating an homography assuming that each lane on the image is parallel on the ground plane. To each rectified lane, an intensity profile is computed along the traffic flow direction for each frame, obtaining an image that represents the displacement as a time function. The main idea is to search for the best matching profile for different times and spaces on each lane, consequently obtaining the velocity profile.

Estimating traffic intensity using profile images on rectified images

In this paper a technique is presented to estimate the traffic intensity for each lane. This method does not require background estimation or even the identification and tracking of individual vehicles. It requires only the identification of each lane and the estimation of a bird eye view of the highway using a rectification method. To each rectified lane, an intensity profile is computed along the traffic flow direction for each frame, obtaining an image that represents the displacement as a time function. The main idea is to search for the best matching profile for different times and spaces on each lane and consequently obtain the velocity profile.