Pinned Repositories
F0-estimation-Noisy-Periodic-Signal-
NLS algorithm for estimation has been implemented successfully and has provided good estimates for all range of frequencies. It has also been tested for different sampling frequencies and this estimator is better compared to autocorrelation method as it doesn’t require any pre-processing. The NLS method is statistically efficient if the noise is modelled as white and Gaussian and much more accurate than the autocorrelation based methods. But a drawback of this algorithm is that it takes time to calculate the cost function matrix as it involves matrix multiplication with an inverse matrix. This method can be made faster if implemented using recursive solver for Toeplitz-plus-Hankel systems for computing NLS cost function.
gans
Generative Adversarial Networks implemented in PyTorch and Tensorflow
HMM-Unsupervised-Machine-learning
Basic implementation of HMM . #Viterbi # Baum Welch
Image-Binarization-Filter-Machine-Learning
Implementation of Image Binarization techniques and enhancement techniques in C#. Medain filter,Convolution filter,Dynamic Gaussian,Gaussian filter
Semantic-Segmentation-using-Fully-Convolutional-Neural-Network-with-TensorFlow-colabs
dataset
TLS-protocol-REST-architecture
TLS protocol implementation ver1.2 according to rfc 5246.
surajbharadwaj's Repositories
surajbharadwaj/F0-estimation-Noisy-Periodic-Signal-
NLS algorithm for estimation has been implemented successfully and has provided good estimates for all range of frequencies. It has also been tested for different sampling frequencies and this estimator is better compared to autocorrelation method as it doesn’t require any pre-processing. The NLS method is statistically efficient if the noise is modelled as white and Gaussian and much more accurate than the autocorrelation based methods. But a drawback of this algorithm is that it takes time to calculate the cost function matrix as it involves matrix multiplication with an inverse matrix. This method can be made faster if implemented using recursive solver for Toeplitz-plus-Hankel systems for computing NLS cost function.
surajbharadwaj/HMM-Unsupervised-Machine-learning
Basic implementation of HMM . #Viterbi # Baum Welch
surajbharadwaj/Image-Binarization-Filter-Machine-Learning
Implementation of Image Binarization techniques and enhancement techniques in C#. Medain filter,Convolution filter,Dynamic Gaussian,Gaussian filter
surajbharadwaj/Semantic-Segmentation-using-Fully-Convolutional-Neural-Network-with-TensorFlow-colabs
dataset
surajbharadwaj/gans
Generative Adversarial Networks implemented in PyTorch and Tensorflow
surajbharadwaj/TLS-protocol-REST-architecture
TLS protocol implementation ver1.2 according to rfc 5246.