Earticle

Phoneme to Viseme mapping has great application in Visual Speech Recognition, Lip Synchronization, Talking Head Applications, movies, news reading, and film industries. Lot of work has been done in area of various face component detection and recognition. Apart from eye detection, ear detection, iris detection etc, lip tracking and lip detection is one of the favourite topics for researchers. Various algorithms and techniques have been implemented so far to achieve better and better performance. Normalized RGB colour scheme, HSV colour model, Lip detection using HUE segmentation and many more techniques have been implemented and are in the boom. All methods are having their own pros and cons. We are aiming to extract out phonemes from speech as well as we extract visual feature i.e. visemes from face by using hue and saturation values. The reason behind the selection of this algorithm is that, it performs well under various illumination conditions, which is the one of the dimension of difficulty in the area of lip detection. We are aiming to carry out the work on in-house database with varying lighting and noisy conditions.

Fast and robust images matching are of vital importance in low-altitude photogrammetric process. Among the most popular features of images matching are currently SIFT and Harris etc. For time-critical applications such as disaster monitoring, the SIFT features extraction are too slow, and the location accuracy of SIFT and Harris is insufficient in photogrammetric process. In this paper, we present a sub-Harris operator coupled with SIFT for fast images matching in low-altitude photogrammetry. Firstly, the original stereopair is down-sampled to small scaling images, in which rough relative orientation is computed by the corresponding points obtained from SIFT matching. Then sub-pixel level precise Harris corners are extracted within original scale images. Finally, the corresponding points are found in the sets of sub-Harris corners consistent with epipolar geometry obtained from rough images matching. Experimental results show that the proposed method can achieve more excellent performances in accuracy than SIFT and Harris operator based method in the relative orientation, and significantly improve the computational efficiency compared with SIFT.

Sparse representation has been successfully applied to visual tracking to find the target with the minimum reconstruction error from the target templates subspace. Traditional sparsity-based trackers handle corruptions and occlusions of the observation by introducing a set of trivial templates. However, the performance is not so satisfactory in practice. It is because the trivial templates unable to model heavy occlusions effectively, and the likelihood computation and the template update processes do not take full advantage of the occlusion information. In this paper, we propose a novel tracking method taking advantage of local sparse representation to detect occlusions during the tracking sequence. In our method, the target is divided into local patches. We analyze the spatial distribution of the samples employed by the local sparse representation, and determine the occlusion state for each patch respectively. The occluded patches are disregard, only the unoccluded ones are considered for reconstruction and likelihood computation. In addition, a dynamic template update strategy with occlusion handling is introduced to alleviate the drift problem. Experiments on challenging video sequences demonstrate that the proposed tracking algorithm performs favorably against several state-of-the-art methods.

In this paper, we proposed a method to detect the rotation angle for the back side image of a container. Our technique consists of sorting the segmented characters according to the X axis, then detecting the segments belonging to the container number, after that, we divide the segments by groups having the same height, and finally the rotation angle of the image will be the average of the skewed angles related to each group. Our approach is robust, efficient and capable of handling any font and size of characters; regarding its complexity for an image having N lines and M characters, the worst CPU time usage and the worst memory usage is equal to O (NxM) while the network usage and disk usage for one image is O (1) which led –while using an old laptop- to a response time around 0.45 milliseconds to detect the rotation angle of an image when rotated from -45o to +45o with an precision error between ±0.2o. The high accuracy and the fast response time for detecting the rotation angle of container images make our approach suitable for online OCR critical applications.

In this paper, we study spectrum sensing based on dimensionality reduction and random forest (RF) in low signal-to-noise ratio environments. Classifications of three digital modulation types, including BPSK, OFDM and 2FSK, are investigated. From the received radio signal, a set of cyclic spectrum features are first calculated, and the principal component analysis (PCA) is applied to extract the most discriminant feature vector for classification. Furthermore, the detecting signal is classified by the trained random forest, which uses the Gini index as the classification criteria, to test whether the primary user exists. The performance of our proposed PCA combining with RF algorithm is evaluated through simulations and compared with MME, SVM, RF. Experimental results show that with dimensionality reduction, the performance of classification is much better with fewer features than that of without dimensionality reduction.

With the growing demand of high quality multimedia (HD) the data size has increased thus the compression is the essential requirement to process and store data with smaller size. The Multiple Parameter Discrete Fractional Fourier Transform (MPDFRFT) is generalization of the discrete fractional Fourier Transform and can be use for compression of high resolution images with the extra degree of freedom provided by the MPDFRFT and its different fractional orders finally decompressed image can also be recovered. This paper deals with the image compression based on MPDFRFT using Eigen vector decomposition algorithm. The MPDFRFT possesses all the desired properties of discrete fractional Fourier transform. The MPDFRFT converts to the DFRFT when all of its order parameters are the same. We exploit the properties of multiple-parameter DFRFT and propose a novel compression scheme for satellite and medical images more conveniently than urban, rural and natural images. In this scheme image is subdivided and MPDFRFT is applied for the subdivided image to form transformed coefficients and Inverse MPDFRFT is applied for reconstruction of original images. The proposed compression scheme with MPDFRFT significantly shows better results over fractional cosine transform (FRCT), Fourier transforms (FT) and cosine transforms (CT). A comparison has been made between these techniques and observed that a good fidelity of decompressed image can be achieved at different fractional order parameter values of the transforms. The performance of system analyzed based on parameters like Peak Signal-to-Noise Ratio (PSNR), mean square error (MSE) and Compression Ratio (CR). The MPDFRFT provides better mean square error (MSE) and peak signal noise ratio (PSNR) for the same compression ratio (CR) as compared to FRCT, FT, cosine transform and classical lifting scheme based on wavelet, during image processing using MATLAB platform.

Step-Frequency signal is widely used in radar systems. However it suffers from the technique challenges that are the imaging moving-targets. In this paper, a new compound modulation signal is analyzed. It combines phased code modulation during the pulse and Step-Frequency modulation between the pulses time, and it is called as phase-coded Step-Frequency signal. Numeric simulation comparison experiments show that PCSF signal can get the same imaging result with less pulse number, which can be used to solve the conflict problem, such as the number of pulse and transmit average power. So this compound modulation signal can be widely used in the application of engineering project. Finally, the processing method of PCSF is simple analyzed, and numerical simulation proves its effectiveness and accuracy.

In order to realize the high precision real-time measurement for the static objects, three coding-pitches triangular patterns phase-shifting profilometry based on modified Sunzi Theorem (ST) was proposed in this paper. First, according to Sunzi Theorem, the triangular pattern is designed. This Sunzi Theorem was adopted to unwrap three wrapped coordinate acquired from triangular patterns fringe images. The greatest common divisor of coding-pitches is obtained by the used projector. In unwrapping processing, the wrapped coordinate is considered as remainder of the absolute coordinate modulo of the coding-pitch. Generally there are two kinds of rounding policies for the remainder in real number set. If the rounding policy selection is wrong, the rounding result of remainder containing decimal fraction error which can lead the unwrapping result to quite away from the correct one. In order to solve the high sensitivity of the decimal fraction error, the decimal fraction difference is recognized as the criterion of the rounding policy selection for real number erroneous remainder. To verify the presented method in this paper, a 3D shape measurement experimental system is constructed using projector and camera. The experiment results shown that, a maximum standard deviation of measurement error to the rule 3D objects is 0.62mm, and the complex surface reconstruction with different surface reflectivity can be realized very well.