King Fahd University of Petroleum and Minerals

King Abdulaziz City for Science and Technology

Lahouari Ghouti

Mohammad K Ibrahim

Adnan A Gutub

Atef J Al-Najjar

Anas Mohammed Albaghajati

Sami El Ferik

Ahmed Ibrahim Al-Herz

Mohammad Alshayeb

Abdulaziz M Al-Khoraidly

Mohammad Tariq Nasir

Moustafa Elshafei

Md Rafiul Hassan

Muhammad Imtiaz Hossain

Abdulazeez Abdulraheem

Amar Khoukhi

Syed Masiur Rahman

Abdullah Hussein Owaidh

Ahmed Abdullah Bahjat

Waleed Mohammad Al-Zahrani

Tarek Ahmed Helmy El-Basuny

A reinforcement learning method and apparatus includes storing video frames in a video memory, performing a first preprocessing step of retrieving a sequence of n image frames of the stored video frames, and merging the n image frames in a fading-in fashion by incrementally increasing the intensity of each frame up to the most recent frame having full intensity to obtain a merged frame; and performing a training step of inputting the merged frame to the DQN and training the DQN to learn Q-values for all possible actions from a state represented by the merged frame with only a single forward pass through the network. The learning method and apparatus includes a second preprocessing step of removing the background from the merged frame. The method can be applied to any DQN learning method that uses a convolution neural network as its core value function approximator.

Enhanced deep reinforcement learning deep q-network models

Methods and apparatuses are provided for code refactoring. The method includes acquiring a code and identifying, using processing circuitry and based on a Markov decision process model, a refactoring sequence. The refactoring sequence includes a plurality of refactoring steps to be applied to the code. Further, the method includes refactoring, by the processing circuitry, the code according to the refactoring sequence.

Method and apparatus for generating a refactored code

Described herein a robot assisted method of deploying sensors in a geographic region. The method of deploying sensors is posed as a Markovian decision process. The robot assigns each grid cell in a map of the geographic region a reward value based on a surface elevation of the geographic region and a soil hardness factor. Further, the robot determines an action for each grid cell of the plurality of grid cells, wherein the action corresponds to an expected direction of movement of the robot in the grid cell. The robot computes a global path as a concatenation of actions starting from a first grid cell and terminating at a second grid cell. The method monitors the movement of the robot on the computed global path and computes a second path based on a deviation of the robot from the global path.

Seismic sensor deployment with a stereographically configured robot

Apparatus and method for deploying sensors

Apparatus and methodologies for code refactoring

An automated content-based image retrieval method, a system and a computer program product for the classification of breast density from mammographic imagery. Raw digital mammogram images taken of patients are initially pre-processed to remove noise and enhance contrast, then subjected to pectoral muscle segmentation to produce region of interest (ROI) images. The ROI images are then decomposed using non-negative matrix factorization (NMF), where a non-negative sparsity constraint and reconstruction quality measures are imposed on the extracted and retained first few NMF factors. Based on the retained NMF factors, kernel matrix-based support vector machines classify the mammogram images binomially or multinomially to breast density categories. Methods of assessing and comparing the NMF-based breast classification method to principal component analysis or PCA-based methods are also described, and the NMF-based method is found to achieve higher classification accuracy and better handling of invariance in the digital mammogram images because of its parts-based factorization.

Method, system and computer program product for breast density classification using parts-based local features

A method for content-based image retrieval for the classification of breast density from mammographic imagery is described. The breast density is characterized through the Fisher linear discriminants (FLD) extracted from the Principal Component Analysis (PCA). Unlike PCA, the FLD provides a very discriminative representation of the mammographic images in terms of the breast density. Various exemplary methods, systems and computer program products are also disclosed.

Method, system and computer program product for breast density classification using fisher discrimination

An apparatus including circuitry configured to receive a plurality of images and extract at least one iris image from each of the plurality of images. The circuitry is configured to receive a claimed identity iris image corresponding to an identity to be authenticated, normalize the iris images and the claimed identity iris image, and filter the normalized extracted iris images to select a subset of the normalized extracted iris images based on a similarity measurement relative to the normalized claimed identity iris image. The circuitry is configured to divide the normalized claimed identity iris image and each image of the subset of images into a plurality of sub-images, filter the sub-images to select the sub-image having a closest similarity measurement relative to a sub-image of the normalized claimed identity image in a corresponding sub-image position to the selected sub-image, and generate a composite iris image by fusing the selected sub-images.

Apparatus and method for iris image analysis

The hybrid virtual account and token-based digital cash protocols provide a user with many pairs of secret and public keys. One of the pairs is the master pair, which consists of one master user private key and one master user public key. Other pairs are pseudonym pairs, wherein each consists of one pseudonym user private key and one pseudonym user public key. The use of a master key pair and pseudonym key pairs circumvents the need for blind signatures. None of the invention's protocols require blind signatures. The protocols comprise public key protocols, digital signatures and symmetric key protocols.

Hybrid virtual account and token-based digital cash protocols

The virtual account and token-based digital cash token protocols use a combination of blind digital signatures and pseudonym authentication with at least two pairs of public and private keys. A user has one master pair of private and public keys and many pseudonym pairs of private and public keys. The new protocols combine blind digital signature and pseudonym-based authentication. Blind digital signatures based on the master pair of keys are used to withdraw digital cash from the user's bank account under the user's real identity. A pseudonym pair of keys is used for depositing digital cash with a digital cash issuer. Pseudonym key pairs are used for spending the digital cash. Digital cash includes digital cash tokens in virtual accounts, which can be managed from a user's fixed or mobile computing platform.

Virtual account and token-based digital cash protocols

The method of performing fingerprint matching includes a matching algorithm that combines affine moment invariants (AMIs) and translation, rotation and scaling invariants (TRS) based on moments analysis without requiring minutiae detection. Preprocessing normalizes the fingerprint image using the Federal Bureau of Investigation's wavelet scalar quantification (WSQ) compression standard definition. Analysis of the orientation field reliably and accurately determines the reference point. The area within a predetermined range around the detected reference point is used as a region of interest (ROI) for feature extraction. A directional filter bank transform (DFB) obtains directional components of the image. An invariant moment analysis on sub-bands of the filtered images extracts features while limiting the effects of noise and non-linear distortions. Absolute distance is used to match the fingerprints.

Method of performing fingerprint matching

The elliptic curve-based message authentication code is a computational method for improving the security of existing message authentication code (MAC) generating methods through the use of elliptic curve cryptography. Particularly, the message authentication codes and elliptic curve cryptography are based on an elliptic curve discrete logarithm problem, which is well known in mathematics to be a computationally hard problem.

Elliptic curve-based message authentication code system and method

The method of performing XZ-elliptic curve cryptography for use with network security protocols provides a computerized method that allows for the encryption of messages through elliptic polynomial cryptography and, particularly, with the embedding of either a symmetric secret key or a public key in the message bit string. The method of performing XZ-elliptic polynomial cryptography is based on the elliptic polynomial discrete logarithm problem. It is well known that an elliptic polynomial discrete logarithm problem is a computationally 'difficult' or 'hard' problem.

Method of performing XZ-elliptic curve cryptography for use with network security protocols

The method of predicting gas composition in a multistage separator includes solutions to the regression problem of gas composition prediction that are developed using an ensemble of hybrid computational intelligence (CI) models. Three separate homogeneous and one heterogeneous ensemble of hybrid computational intelligence (EHCI) models are developed using a parallel scheme. The homogeneous models have the same types of CI models used as base learners, and the heterogeneous model has of different types of CI models used as base learners. Various popular CI models, including multi-layer perceptron (MLP), support vector regression (SVR) and adaptive neuro-fuzzy inference system (ANFIS), are used as base learners of ensemble models.

Growing community of inventors

Dhahran, Saudi Arabia

Lahouari Ghouti