The model of local spatial frequency provides a powerful analytical for image analysis. In This paper we explore the application of this representation to long-standing problem in stereovision "the foreshortening problem". We develop phase difference-based algorithm that use an adaptive scale selection process at the corresponding points in the two views. This takes into account surface perspective distortion (foreshortening). Challenges arise from the fact that stereo images are acquired from a slightly different view. Therefore, the projection of the surface in the images is more compressed and occupied a smaller area in one view. Instead of matching intensities directly, a Gabor scale-space expansion (scalogram) is used. The phase difference at corresponding points in the two images is used to estimate the disparity. The suggested algorithm provides an analytical closed-form expression for the effect of perspective foreshortening. It also demonstrates a novel solution to the phase-wraparound problem that has limited the application of other phase-based method. The efficiency and performance is confirmed on the basis of analysis of rectified real images. Hence, our proposed method has a superior performance in comparison with other methods.

Abstract

This paper describes a method to recognize the alpha-was tested to vocabulary of 36 gestures including the American betsfrom a single hand motion using Hidden Markov Models (HMM). Sign Language (ASL) letter spelling alphabet and digits. Yoon In our method, gesture recognition for alphabets is based on three et. al.[6] introduced a hand gesture recognition method which main stages; preprocessing, feature extraction and classification. In preprocessing stage, color and depth information are used to detect used a combined features of locations angle and velocity to both hands and face in connection with morphological operation. determine the discrete vector that is used as input to HMM. After the detection of the hand, the tracking will take place in furtherIn shortly, a gesture is a spatio-temporal pattern [11] and step in order to determine the motion trajectory so-called gesture may be static or dynamic or both as in sign language recogpath. The second stage, feature extraction enhances the gesture path nition. Since HMM are used widely in handwriting, speech which gives us a pure path and also determines the orientation .. . . a between the center of gravity and each point in a pure path. Thereby, recognition, part-of-speech tagging and machne translation, the orientation is quantized to give a discrete vector that used therefore in contrast to Liu et. al.[11], we develop a method as input to HMM. In the final stage, the gesture of alphabets is to recognize the gesture hand graphical from A to Z using recognized by using Left-Right Banded model (LRB) in conjunction Hidden Markov Model take in our account the orientation with Baum-Welch algorithm (BW) for training the parameters of between any point in a pure path and the center of gravity HMM. Therefore, the best path is obtained by Viterbi algorithm th using a gesture database. In our experiment, 520 trained gestures in the graphical gesture. The method depends on the database are used for training and also 260 tested gestures for testing. Our which we built it and Left-Right Banded model (topology of method recognizes the alphabets from A to Z and achieves an average HMM) with 6 states. Each alphabet is based on 30 video recognition rate of 92.3%. (20 for training and 10 for testing) where the input images are captured by a Bumblebee stereo camera that has 6mm Keywords Hidden Markov Model, Gesture recognition, Pattern focaltlen fo aButb2eto e seo at5ra per secon recogition Appicaion focal length for about 2 to 5 second at 15 frames per second recognition, Application.with 240 x 320 pixels image resolution on each frame. The recognition rate is achieved on training and testing gesture with 98.46% and 92.3% respectively. The next sections describe Sign language recognition from hand motion or hand pos-how the method is built and gesture hand graphical is tested.

No abstract

Document analysis tasks, as text recognition, word spotting, or segmentation, are highly dependent on comprehensive and suitable databases for training and validation. However their generation is expensive in sense of labor and time. As a matter of fact, there is a lack of such databases, which complicates research and development. This is especially true for the case of Arabic handwriting recognition, that involves different preprocessing, segmentation, and recognition methods, which have individual demands on samples and ground truth. To bypass this problem, we present an efficient system that automatically turns Arabic Unicode text into synthetic images of handwritten documents and detailed ground truth. Active Shape Models (ASMs) based on 28046 online samples were used for character synthesis and statistical properties were extracted from the IESK-arDB database to simulate baselines and word slant or skew. In the synthesis step ASM based representations are composed to words and text pages, smoothed by B-Spline interpolation and rendered considering writing speed and pen characteristics. Finally, we use the synthetic data to validate a segmentation method. An experimental comparison with the IESK-arDB database encourages to train and test document analysis related methods on synthetic samples, whenever no sufficient natural ground truthed data is available.