Speech and non-speech identification along with its classification method that need to be improved in the endpoint detection for speech in noisy environments. The proposed method uses few features to increase the robustness in various noisy environments, and the classification used here KNN technique is applied to effectively combine these multiple features for classification of each speech signal. We evaluate the performance of the proposed method by conducting speech and non-speech classification experiments on noisy speech. We also investigate the importance of various features on speech and non-speech classification in noisy environments and by using this KNN algorithm to obtaining 80% accuracy
This study work is basically aimed at designing and testing of hardware module to perform inversion operation of a matrix in a small time. Here, an approach is made for calculating 3×3 matrix inverse. There are many mathematical methods available for performing matrix inversion and out of them a suitable method, like Adjoint Matrix Method is selected by analysing the computational requirements. The mathematical method of calculating the inverse of matrix is then suitably converted into VHDL code. The code is then tested for simulation using a set of test matrices. After simulation is verified by checking the results of test inputs the code is tested for synthesizability. After the synthesizability is verified then it is finally tested for hardware verification by dumping into FPGA. Altera's DE1 board which consists a Cyclone-II series FPGA EP2C20F484C7 FPGA is used for this study. The test inputs can be fed in either by using on board GPIO or UI or S RAM. The outputs are taken the same way either by GPIO or UI or written to S RAM and are then to be verified by comparing with actual results.
This paper proposed a way of automating the supply chain in industries. For the purpose of demonstration an industrial manufacturing site is modeled as a multi colored zone with a robot representing the vehicles. Navigation of the robot within the zone is realized through line following techniques with intelligence provided by appropriate image processing. The components used are a laptop, a Line Follower, USB camera, serial communication cable.The processes involved are color based segmentation, robot navigation, character differentiation, node counting, transmission of navigation signals
This paper explores the spatial coverage potential of a vision sensor to implement a dual service for a museum security system. Services include object monitoring and invisible visual Multiple-Input and Multiple-Output (MIMO) communication. For object monitoring, the Kalman filter algorithm is used. To implement a secret visual MIMO communication, an Adaptive Row-Column (ARC)-based LSB substitution technique is used along with the Integer Wavelet Transform method. These proposed services are implemented using existing electronic visual displays and surveillance cameras. Experiments are carried out using a sample object and four stego images. The performance of the object tracking system is measured in terms of accuracy by comparing the actual and estimated position of an object. Similarly, the performance of the covert visual MIMO system is measured at two different stages. On the transmitter side, differences between the cover and stego-images are used to measure imperceptibility. Likewise, at the receiver, differences between the original and the reconstructed data images are used to measure the accuracy. Results illustrate that the proposed system has improved imperceptibility, accuracy, and robustness, as compared to existing techniques.
An analytic design of a prototype filter for M-channel maximally decimated cosine-modulated Near Perfect Reconstruction (NPR) filter banks is proposed in this work. The prototype filter is created using the least-square (CLS) method with weighted constraints, which is one-dimensional and requires single-parameter optimization. Compared to existing approaches, this suggested method achieves rapid convergence by analytically determining the optimal step size, ensuring the 3 dB cutoff frequency at π/2 M. The simulation results for design examples outperform the techniques in the available literature in terms of amplitude and aliasing distortion, reaching distortion around 2.4489 × 10-4 and 3.4907 × 10-9, respectively. This optimization algorithm's usefulness is further demonstrated with the sub-band coding of ECG signals. Implementing optimal prototype filters has tangible real-world effects, especially in critical sectors like healthcare and communications, improving diagnostics accuracy, data transmission efficiency, and overall performance.