Electronic Letters on Computer Vision and Image Analysis (ELCVIA - Universitat Autònoma de Barcelona)
Not a member yet
343 research outputs found
Sort by
Genetic Programming for Object Detection: A Two-Phase Approach with an Improved Fitness Function
Get PDFThis paper describes two innovations that improve the efficiency and effectiveness of a genetic programming approach to object detection problems. The approach uses genetic programming to construct object detection programs that are applied, in a moving window fashion, to the large images to locate the objects of interest. The first innovation is to break the GP search into two phases with the first phase applied to a selected subset of the training data, and a simplified fitness function. The second phase is initialised with the programs from the first phase, and uses the full set of training data with a complete fitness function to construct the final detection programs. The second innovation is to add a program size component to the fitness function. This approach is examined and compared with a neural network approach on three object detection problems of increasing difficulty. The results suggest that the innovations increase both the effectiveness and the efficiency of the genetic programming search, and also that the genetic programming approach outperforms a neural network approach for the most difficult data set in terms of the object detection accuracy
Correction for Camera Roll in a Perspectively Distorted Image: Cases for 2 and 3 point perspectives
Get PDFWe propose a method for correction of camera roll in a perspectively distorted image. Here we employ the rectification of the horizon line formed from the vanishing points in order to bring back the image rotation caused from the camera roll. In teleconference systems, surveillance system, aerial images of buildings, mobile robots visual system etc., the range of camera’s view spreads with pan, tilt and roll. It is intended to take omni-directional images for wide range coverage, but this invariably introduces rotation in many images due to the roll of the camera. This paper presents the method for camera roll correction by using the perspective distortion in images captured from a pan/tilt/roll camera control system. The proposed method works on two cases, viz. Calibrated and uncalibrated cameras. Calibrated camera with intrinsic calibration of the camera such as optical center and the focal length of the camera in cases of two point perspectives is considered and in case of three point perspective uncalibrated camera is considered and the camera is partially calibrated for the camera principal point and the roll is performed subsequently. The result comparison made reveals more about the usability of the algorithm in case of calibrated and uncalibrated cases of the camera used
Wavelets and partial differential equations for image denoising
Get PDFIn this paper a wavelet based model for image de-noising is presented. Wavelet coefficients are modelled as waves that grow while dilating along scales. The model establishes a precise link between corresponding modulus maxima in the wavelet domain and then allows to predict wavelet coefficients at each scale from the first one. This property combined with the theoretical results about the characterization of singularities in the wavelet domain enables to discard noise. Significant structures of the image are well recovered while some annoying artifacts along image edges are reduced. Some experimental results show that the proposed approach outperforms the most recent and effective wavelet based denoising schemes
3D Model Based Pose Invariant Face Recognition from a Single Frontal View
Get PDFThis paper proposes a 3D model based pose invariant face recognition method that can recognize a face of a large rotation angle from its single nearly frontal view. The proposed method achieves the goal by using an analytic-to-holistic approach and a novel algorithm for estimation of ear points. Firstly, the proposed method achieves facial feature detection, in which an edge map based algorithm is developed to detect the ear points. Based on the detected facial feature points 3D face models are computed and used to achieve pose estimation. Then we reconstruct the facial feature points\u27 locations and synthesize facial feature templates in frontal view using computed face models and estimated poses. Finally, the proposed method achieves face recognition by corresponding template matching and corresponding geometric feature matching. Experimental results show that the proposed face recognition method is robust for pose variations including both seesaw rotations and sidespin rotations
Improved motion segmentation based on shadow detection
Get PDFIn this paper, we discuss common colour models for background subtraction and problems related to their utilisation are discussed. A novel approach to represent chrominance information more suitable for robust background modelling and shadow suppression is proposed. Our method relies on the ability to represent colours in terms of a 3D-polar coordinate system having saturation independent of the brightness function; specifically, we build upon an Improved Hue, Luminance, and Saturation space (IHLS). The additional peculiarity of the approach is that we deal with the problem of unstable hue values at low saturation by modelling the hue-saturation relationship using saturation-weighted hue statistics. The effectiveness of the proposed method is shown in an experimental comparison with approaches based on RGB, Normalised RGB and HSV
Intelligent CCTV for Mass Transport Security: Challenges and Opportunities for Video and Face Processing
Get PDFCCTV surveillance systems have long been promoted as being effective in improving public safety. However due to the amount of cameras installed, many sites have abandoned expensive human monitoring and only record video for forensic purposes. One of the sought-after capabilities of an automated surveillance system is ";face in the crowd"; recognition, in public spaces such as mass transit centres. Apart from accuracy and robustness to nuisance factors such as pose variations, in such surveillance situations the other important factors are scalability and fast performance. We evaluate recent approaches to the recognition of faces at large pose angles from a gallery of frontal images and propose novel adaptations as well as modifications. We compare and contrast the accuracy, robustness and speed of an Active Appearance Model (AAM) based method (where realistic frontal faces are synthesized from non-frontal probe faces) against bag-of-features methods. We show a novel approach where the performance of the AAM based technique is increased by side-stepping the image synthesis step, also resulting in a considerable speedup. Additionally, we adapt a histogram-based bag-of-features technique to face classification and contrast its properties to a previously proposed direct bag-of-features method. We further show that the two bag-of-features approaches can be considerably sped up, without a loss in classification accuracy, via an approximation of the exponential function. Experiments on the FERET and PIE databases suggest that the bag-of-features techniques generally attain better performance, with significantly lower computational loads. The histogrambased bag-of-features technique is capable of achieving an average recognition accuracy of 89% for pose angles of around 25 degrees. Finally, we provide a discussion on implementation as well as legal challenges surrounding research on automated surveillance
Classification of Objects and Background Using Parallel Genetic Algorithm Based Clustering
Get PDFIn this paper, two novel strategies have been proposed to obtain segmentation of an object and background in a given scene. The first one, known as Featureless(FL) approach, deals with the histogram of the original image where Parallel Genetic Algorithm (PGA) based clustering notion is used to determine the optimal threshold from the discrete nature of the histogram distribution. In this regard, we have proposed a new interconnection model for PGA. The second scheme, the Featured Based(FB) approach, is based on the proposed featured histogram distribution. A feature from the given image is extracted and the histogram corresponding to the derived feature pixels is used to determine the optimal threshold for the original image. The proposed PGA based clustering is used to determine the optimal threshold. The performance of both the schemes is compared with that of Otsu\u27s and Kwon\u27s method and FB method is found to be the best among the three techniques
A Variational Approach to 3D Cylindrical Geometry Reconstruction from Multiple Views
Get PDFIn this paper we present a variational technique for the reconstruction of 3D cylindrical surfaces. Roughly speaking by a cylindrical surface we mean a surface that can be parameterized using the projection on a cylinder in terms of two coordinates, (l, ?), representing the displacement and angle in a cylindrical coordinate system respectively. The starting point for our method is a set of different views of a cylindrical surface, as well as a precomputed disparity map estimation between pair of images. The proposed variational technique is based on an energy minimization where we balance on the one hand the regularity of the cylindrical function given by the distance of the surface points to cylinder axis, and on the other hand, the distance between the projection of the surface points on the images and the expected location following the precomputed disparity map estimation between pair of images. One interesting advantage of this approach is that we regularize the 3D surface by means of a bi-dimensional minimization problem. We show some experimental results for large stereo sequences
A Colour Code Algorithm For Signature Recognition
Get PDFThe paper "A Colour Code Algorithm for Signature Recognition" accounts an image processing application where any user can verify signature instantly. The system deals with a Colour code algorithm, which is used to recognize the signature. The paper deals with the recognition of the signature, as human operator generally make the work of signature recognition. Hence the algorithm simulates human behavior, to achieve perfection and skill through AI. The logic that decides the extent of validity of the signature must implement Artificial Intelligence Pattern recognition is the science that concerns the description or classification of measurements, usually based on underlying model. The measurement or the properties used to classify the objects are called as \u27features\u27, and the types or categories into which they are classified are called as classes. Since most pattern recognition tasks are first done by humans and automated later, the most fruitful source of features has been to asked the people who classify the objects how they tell them a part. The two main approaches to pattern recognition are the statistical (decision theoretic) and the syntactic approaches. Signature recognition is the best example of this fact. The algorithm is tested on various operating systems & we find that it works very well & satisfactory. While implementing the recognition process, we have used quite simpler way. At this stage we are getting accuracy up to about 80% to 90%. These conclusions are made on the basis of testing of 300 person\u27s database
Automatic Detection of Facial Midline And Its Contributions To Facial Feature Extraction
Get PDFWe propose a novel approach for detection of the facial midline from a frontal face image. Using midline as a guide reduces computational cost required for facial feature extraction (FFE) because the midline is capable of restricting multi-dimensional searching process into one-dimensional search. The proposed method detects the facial midline from an edge image as the symmetry axis using the generalized Hough transformation. Experimental results on the FERET database indicate that the proposed algorithm can accurately detect facial midlines over many different scales and rotation. The total computational time for facial feature extraction has been reduced by a factor of 280 using the midline detected by this method