Advanced Deepfake Image Detection: A Robust Framework using InceptionV3 and Xception
Deep Learning in Dermatology: Distinguishing Tinea Ringworm from Eczema through Differential Diagnosis
Pose Tracking, Analysis and Impact Estimation in Real Time
Environmental Damage Assessment using Image Processing
A Review on Brain Tumor Detection using Machine Learning
Identification of Volcano Hotspots by using Resilient Back Propagation (RBP) Algorithm Via Satellite Images
Data Hiding in Encrypted Compressed Videos for Privacy Information Protection
Improved Video Watermarking using Discrete Cosine Transform
Contrast Enhancement based Brain Tumour MRI Image Segmentation and Detection with Low Power Consumption
Denoising of Images by Wavelets and Contourlets using Bi-Shrink Filter
Deepfake technology, driven by generative artificial intelligence, has emerged as a significant challenge to digital trust by enabling the creation of hyper-realistic yet fraudulent media content. Such manipulations are increasingly used for misinformation, financial fraud, and cybersecurity attacks. To address this, the present study introduces a robust deepfake image detection framework based on Convolutional Neural Networks (CNNs), specifically InceptionV3 and Xception. A dataset of 19,457 images containing authentic and AI-generated forgeries was preprocessed through normalization, augmentation, and class balancing to enhance generalization. Both models were fine-tuned using transfer learning on stratified training, validation, and test sets. The results demonstrate strong detection capabilities, with InceptionV3 achieving 92.19% accuracy and Xception achieving 93.44%, while an ensemble approach further improved accuracy to 94.50%. Comparative evaluation against existing approaches indicates that the proposed framework outperforms several image-based detection systems in terms of accuracy, precision, and recall. This study shows that fine-tuned CNNs offer scalable and reliable solutions for detecting deepfakes, thereby supporting cybersecurity, media authentication, and digital forensics. Future work may extend this framework to multimodal and temporal deepfake detection for enhanced robustness against evolving threats.
Misdiagnosis of dermatological disorders is a common issue among healthcare professionals worldwide, especially when distinguishing between conditions with visually similar presentations. Ringworm (Tinea) and eczema are two skin disorders that are frequently misdiagnosed, leading to inappropriate treatments and potential complications. This study proposes a deep learning-based approach to enhance the differential diagnosis of these conditions using advanced convolutional neural networks (CNNs). Five pre-trained CNN architectures, such as VGG16, ResNet50, DenseNet121, InceptionV3, and EfficientNetB0, were fine-tuned and evaluated on the DermNet dataset, focusing on classifying ringworm and eczema. To improve model generalization, various data augmentation techniques were applied during training. Among the evaluated models, DenseNet121 demonstrated superior performance, achieving the highest classification accuracy. This model's effectiveness highlights its potential to significantly reduce misdiagnosis rates in dermatology. The results suggest that deploying CNN-based diagnostic tools could lead to more accurate and efficient dermatological assessments, improving both diagnosis precision and treatment outcomes. These findings pave the way for further research into AI-assisted healthcare solutions aimed at addressing diagnostic challenges in dermatology.
The present study introduces a user-friendly and highly effective push-up monitoring system that exploits camera technology and computer vision techniques to help in conducting workout routines. The system is able to evaluate the push-up form used by the target user, counts the number of repetitions done, and also provides feedback in real-time to ensure that the users maintain proper form while doing the exercise and avoid some common mistakes. This is a sure exercise program tool that will help the fitness enthusiast, trainer, or physiotherapist in improving exercises. It can be used for all types of environments, such as fitness centers, home workouts, and rehabilitation, offering a smart way to monitor progress and optimize training.
Deforestation remains one of the most pressing environmental challenges, significantly impacting ecosystems, biodiversity, and climate stability. This paper proposes the development of an Environmental Damage Assessment System (EDAS) that utilizes advanced Artificial Intelligence (AI) and Machine Learning (ML) algorithms to process satellite imagery for deforestation monitoring. The system identifies trends, forecasts potential risks, and provides actionable insights to support sustainable forest management. EDAS serves as a valuable tool for policymakers, researchers, and conservationists aiming to mitigate environmental degradation. One of EDAS's key strengths is its ability to forecast areas at high risk of deforestation by analyzing historical imagery data. This enables timely interventions and effective conservation planning by policymakers and environmental organizations. Overall, EDAS serves as a valuable tool for understanding the broader environmental impacts of deforestation, supporting data-driven decisions for sustainable forest management.
Biopsy for brain tumor assessment is invasive and risky, motivating noninvasive alternatives based on MRI. This review synthesizes recent advances toward an automated, robust, and intelligent system for early brain tumor diagnosis and grading. A six-step pipeline is presented that integrates standardized MRI preprocessing with histogram equalization, adaptive gamma correction, and Wiener filtering; metaheuristic optimization using Particle Swarm Optimization, sine–cosine, and Grey Wolf Optimization for parameter and feature selection; volumetric segmentation through 3D U- Net and related architectures; and classification through deep learning, transfer learning, and complementary machine learning models. Privacy-preserving deployment within smart healthcare ecosystems is further highlighted using federated learning and secure optimization. Evidence across recent studies indicates that the combination of principled enhancement, optimization, and modern segmentation/classification markedly improves detection, boundary delineation, and grading robustness. Concluding with challenges in cross-site generalization, rigorous clinical validation, and trustworthy AI, the discussion outlines a path to clinically viable, noninvasive MRI-based tumor diagnostics.
