Tudor Barbu

Novel automatic video cut detection technique using Gabor filtering

Video shot transition identification constitutes an important computer vision research field, being applied, as an essential step, in many other digital video analysis domains: video scene detection, video compression, video indexing, video content retrieval and video object tracking. This paper approaches the video cut transition detection domain, providing a novel feature-based automatic identification method. We propose a feature extraction technique that uses 2D Gabor filtering, computing tridimensional image feature vectors for the video frames. Most shot cut detection techniques use a thresholding operation to discriminate between the inter-frame difference metric values and thus identify the video break points. Our identification approach is not threshold-based, using an automatic unsupervised distance classification procedure instead of a threshold. Thus, we provide a region-growing based classification approach, that proves to be very efficient in clustering the distances between feature vectors of consecutive frames. The two resulted distance classes determine a satisfactory video shot detection.

Variational Image Denoising Approach with Diffusion Porous Media Flow

A novel PDE-based image denoising approach is proposed in this paper. One designs here a nonlinear filter for image noise reduction based on the diffusion flow generated by the porous media equation , where is a nonlinear continuous function of the form , . With respect to standard 2D Gaussian smoothing and some nonlinear PDE-based filters, this one is more efficient to remove noise from degraded images and also to reduce “staircasing” effects and preserve the image edges.

Pedestrian detection and tracking using temporal differencing and HOG features

We proposed a novel multiple moving object detection technique.Objects representing pedestrians are detected using shape- and skin-based conditions.Our human tracking technique uses HOG-based features and an object matching process. This article proposes a multiple human detection and tracking approach. A moving person identification technique is provided first. The video objects are detected using a novel temporal differencing based procedure and several mathematical morphology-based operations. Then, our technique determines what moving image objects represent pedestrian people, by testing several conditions related to human bodies and detecting the skin regions from the movie frames. A robust human tracking method using a Histogram of Oriented Gradient (HOG) based template matching process is then introduced in our paper. Some person detection and tracking experiments and method comparisons are also described.

Robust Anisotropic Diffusion Scheme for Image Noise Removal

Abstract An anisotropic diffusion-based image enhancement method is proposed in this article. The provided PDE denoising approach is derived from the well-known Perona-Malik nonlinear diffusion model, representing an improved version of it. Thus, we model a novel diffusivity function and explain the mathematical reasons behind it. The conductance parameter of our diffusion technique is automatically detected at each iteration. The proposed PDE-based smoothing technique provides satisfactory noise removal and edge enhancement results, outperforming the most diffusion-based approaches.

A Novel Variational PDE Technique for Image Denoising

A robust variational PDE model for image noise removal is proposed in this paper. One considers an energy functional to be minimized, based on a novel smoothing constraint. Then, the corresponding Euler-Lagrange equation is determined. The obtained PDE model is solved, by using a numerical discretization scheme. Some results of our image denoising experiments and method comparisons are also described in this article.

PDE-based image restoration using variational denoising and inpainting models

A PDE-based image restoration model is proposed in this paper. It aims to restore degraded images that are affected by both noise and missing zones. The considered restoration approach is based on two PDE variational techniques. The first variational method performs an efficient noise reduction, while the second variational model provides the image reconstruction. By using both variational models, one achieves a much better enhancement of the degraded image.

Content-Based Image Retrieval Using Gabor Filtering

This paper provides a content-based digital image retrieval system. Our CBIR system uses the query by example technique and the relevance feedback. A Gabor filter based image feature extraction is proposed first. Thus, 3D image feature vectors using even-symmetric 2D Gabor filters are computed for the images of a large collection and for the input image. At each step an input image is selected, from the output set obtained in the previous step, and the most similar images from the collection are retrieved.

Image Categorization Based on Computationally Economic LAB Colour Features

An easy to compute and small colour feature vector is introduced in this paper, as a tool to be used in the process of retrieval or classification of similarly coloured digital images from very large databases. A particular set of “ab” planes from the LAB colour system is used, along with a specific configuration of colour regions within them. The colour feature vector is low dimensional (only 96 components), computationally economic and performs very well on a carefully selected database of rose images, publicly available.

Comparing various voice recognition techniques

In this paper we describe and compare some speaker recognition techniques. A mel-cepstral short-time analysis for speech signals is presented first. It is used in the voice feature extraction stage. Then speech-dependent and speech-independent speaker recognition approaches are proposed by us in this paper. Minimum mean-distance classification methods are used by the supervised classifier of the voice identification system. Threshold-based procedures are provided for the speaker verification part of the system.

Content-based video recognition technique using a nonlinear metric

This work focuses on the videoclip recognition task. The proposed video recognition approach represents a process composed of two parts: the feature extraction operation and the classification of the obtained video feature vectors. We provide a way to define the similarity between movies. In the featuring stage each clip is characterized by the graphical content, expressed by some feature vectors, of its keyframes. Shot detection and keyframe extraction techniques are also provided. We then propose a special nonlinear metric which is able to compare video feature vectors represented as matrices having different dimensions, and use it in the classification process