IEEE

CNN-Transformer Based Image Emotion Recognition with Multi-Scale Feature Enhancement

Abstract: Image emotion recognition aims to analyze and understand the emotions conveyed by images. Due to the high similarity of features between different emotion categories, the model is prone to ...
IEEE

Multi-Location and Multi-Feature LMP Forecasting: A 2D Spatiotemporal LSTM-CNN Approach

Abstract: Accurate Locational Marginal Price (LMP) forecasting is crucial for effective energy procurement in electricity markets, as it helps utilities make informed, cost-efficient decisions. This ...
IEEE

Point-Cloud-Based Place Recognition Using CNN Feature Extraction

Abstract: This paper proposes a novel point-cloud-based place recognition system that adopts a deep learning approach for feature extraction. By using a convolutional neural network pre-trained on ...
IEEE

AI-Powered Multimodal Seizure Detection and Classification using CNN and GAN-based EEG Signal Analysis

Abstract: Epilepsy is one of the most common neurological disorders and it still requires very precise and quick detection of seizures in order to provide effective medical treatment. The systems ...
Frontiers

Adaptive self-attention for enhanced segmentation of adult gliomas in multi-modal MRI

Every year there are an estimated 80,000–90,000 new glioma cases, highlighting the need for reliable imaging-based decision support. Although deep learning has improved tumor sub-region segmentation, ...
IEEE

Using Code Coverage to Assess Feature Gaps in MPI Correctness Tool Classification Tests

Abstract: We examine the code generator-based MPI correctness benchmark MPI-BugBench (MBB) by analyzing the code coverage it triggers in three tools: MUST, PARCOACH, and clang-tidy. We present our ...
IEEE

Real-Time Driver Distraction Detection via CNN Feature Extraction and PCA Regularized SVM

Abstract: Driver distraction is a leading contributor to road accidents, highlighting the need for intelligent systems that enable early detection and prevention. This paper presents a model designed ...
IEEE

A Comparative Analysis of Machine Learning Based Power Flow Study with Custom Made Open Source Python Codes

Abstract: Power flow analysis is a cornerstone of power system planning and operation, involving the solution of nonlinear equations to determine the steady-state operating conditions of the power ...
IEEE

Robust Extraction of Masked and Occluded Text using Masked Vision Transformer and CNN Decoder

Abstract: The extraction of text from masked or partially occluded images, such as a road sign covered by a poster, a tree, or the environment (in which case occlusion may be caused by anything but a ...