Papers by Niharjyoti Sarangi
Cryptographic hash functions for calculating the message digest of a message has been in practica... more Cryptographic hash functions for calculating the message digest of a message has been in practical use as an effective measure to maintain message integrity since a few decades. This message digest is unique, irreversible and avoids all types of collisions for any given input string. The message digest calculated from this algorithm is propagated in the communication medium along with the original message from the sender side and on the receiver side integrity of the message can be verified by recalculating the message digest of the received message and comparing the two digest values. In this paper we have designed and developed a new algorithm for calculating the message digest of any message and implemented t using a high level programming language. An experimental analysis and comparison with the existing MD5 hashing algorithm, which is predominantly being used as a cryptographic hashing tool, shows this algorithm to provide more randomness and greater strength from intrusion at...
Proceedings of the International Conference on Pattern Recognition Applications and Methods, 2015
Automatically assigning semantically relevant tags to an image is an important task in machine le... more Automatically assigning semantically relevant tags to an image is an important task in machine learning. Many algorithms have been proposed to annotate images based on features such as color, texture, and shape. Success of these algorithms is dependent on carefully handcrafted features. Deep learning models are widely used to learn abstract, high level representations from raw data. Deep belief networks are the most commonly used deep learning models formed by pre-training the individual Restricted Boltzmann Machines in a layer-wise fashion and then stacking together and training them using error back-propagation. In the deep convolutional networks, convolution operation is used to extract features from different sub-regions of the images to learn better representations. To reduce the time taken for training, models that use convex optimization and kernel trick have been proposed. In this paper we explore two such models, Tensor Deep Stacking Network and Kernel Deep Convex Network, for the task of automatic image annotation. We use a deep convolutional network to extract high level features from raw images, and then use them as inputs to the convex deep learning models. Performance of the proposed approach is evaluated on benchmark image datasets.
Tensor Deep Stacking Networks and Kernel Deep Convex Networks for Annotating Natural Scene Images
Lecture Notes in Computer Science, 2015
Image annotation is defined as the task of assigning semantically relevant tags to an image. Feat... more Image annotation is defined as the task of assigning semantically relevant tags to an image. Features such as color, texture, and shape are used by many machine learning algorithms for the image annotation task. Success of these algorithms is dependent on carefully handcrafted features. Deep learning models use multiple layers of processing to learn abstract, high level representations from raw data. Deep belief networks are the most commonly used deep learning models formed by pre-training the individual Restricted Boltzmann Machines in a layer-wise fashion and then stacking together and training them using error back-propagation. However, the time taken to train a deep learning model is extensive. To reduce the time taken for training, models that try to eliminate back-propagation by using convex optimization and kernel trick to get a closed-form solution for the weights of the connections have been proposed. In this paper we explore two such models, Tensor Deep Stacking Network and Kernel Deep Convex Network, for the task of automatic image annotation. We use a deep convolutional network to extract high level features from different sub-regions of the images, and then use these features as inputs to these models. Performance of the proposed approach is evaluated on benchmark image datasets.
Arxiv preprint arXiv:1003.5787, Jan 1, 2010
Cryptographic hash functions for calculating the message digest of a message has been in practica... more Cryptographic hash functions for calculating the message digest of a message has been in practical use as an effective measure to maintain message integrity since a few decades. This message digest is unique, irreversible and avoids all types of collisions for any given input string. The message digest calculated from this algorithm is propagated in the communication medium along with the original message from the sender side and on the receiver side integrity of the message can be verified by recalculating the message digest of the received message and comparing the two digest values. In this paper we have designed and developed a new algorithm for calculating the message digest of any message and implemented it using a high level programming language. An experimental analysis and comparison with the existing MD5 hashing algorithm, which is predominantly being used as a cryptographic hashing tool, shows this algorithm to provide more randomness and greater strength from intrusion attacks. In this algorithm the plaintext message string is converted into binary string and fragmented into blocks of 128 bits after being padded with user defined padding bits. Then using a pseudo random number generator a key is generated for each block and operated with the respective block by a bitwise operator. This process is iterated for the whole message and finally a fixed length message digest is obtained.
Uploads
Papers by Niharjyoti Sarangi