Metasurface polarizers are essential optical components in modern integrated optics and play a vital role in many optical applications including Quantum Key Distribution systems in quantum cryptography. However, inverse design of metasurface polarizers with high efficiency depends on the proper prediction of structural dimensions based on required optical response. Deep learning neural networks can efficiently help in the inverse design process, minimizing both time and simulation resources requirements, while better results can be achieved compared to traditional optimization methods. Hereby, utilizing the COMSOL Multiphysics Surrogate model and deep neural networks to design a metasurface grating structure with high extinction rat

Facial emotion recognition finds many real applications in the daily life like human robot interaction, eLearning, healthcare, customer services etc. The task of facial emotion recognition is not easy due to the difficulty in determining the effective feature set that can recognize the emotion conveyed within the facial expression accurately. Graph mining techniques are exploited in this paper to solve facial emotion recognition problem. After determining positions of facial landmarks in face region, twelve different graphs are constructed using four facial components to serve as a source for sub-graphs mining stage using gSpan algorithm. In each group, the discriminative set of sub-graphs are selected and fed to Deep Belief Network (DBN) f

The deep learning algorithm has recently achieved a lot of success, especially in the field of computer vision. This research aims to describe the classification method applied to the dataset of multiple types of images (Synthetic Aperture Radar (SAR) images and non-SAR images). In such a classification, transfer learning was used followed by fine-tuning methods. Besides, pre-trained architectures were used on the known image database ImageNet. The model VGG16 was indeed used as a feature extractor and a new classifier was trained based on extracted features.The input data mainly focused on the dataset consist of five classes including the SAR images class (houses) and the non-SAR images classes (Cats, Dogs, Horses, and Humans). The Conv

The complexity and variety of language included in policy and academic documents make the automatic classification of research papers based on the United Nations Sustainable Development Goals (SDGs) somewhat difficult. Using both pre-trained and contextual word embeddings to increase semantic understanding, this study presents a complete deep learning pipeline combining Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Network (CNN) architectures which aims primarily to improve the comprehensibility and accuracy of SDG text classification, thereby enabling more effective policy monitoring and research evaluation. Successful document representation via Global Vector (GloVe), Bidirectional Encoder Representations from Tra

Face recognition is a crucial biometric technology used in various security and identification applications. Ensuring accuracy and reliability in facial recognition systems requires robust feature extraction and secure processing methods. This study presents an accurate facial recognition model using a feature extraction approach within a cloud environment. First, the facial images undergo preprocessing, including grayscale conversion, histogram equalization, Viola-Jones face detection, and resizing. Then, features are extracted using a hybrid approach that combines Linear Discriminant Analysis (LDA) and Gray-Level Co-occurrence Matrix (GLCM). The extracted features are encrypted using the Data Encryption Standard (DES) for security

    In this article, the research presents a general overview of deep learning-based AVSS (audio-visual source separation) systems. AVSS has achieved exceptional results in a number of areas, including decreasing noise levels, boosting speech recognition, and improving audio quality. The advantages and disadvantages of each deep learning model are discussed throughout the research as it reviews various current experiments on AVSS. The TCD TIMIT dataset (which contains top-notch audio and video recordings created especially for speech recognition tasks) and the Voxceleb dataset (a sizable collection of brief audio-visual clips with human speech) are just a couple of the useful datasets summarized in the paper that can be used to test A