Tag: data augmentation

A controlled study of 1,196 coffee leaf images found that an explainable Vision Transformer classified nutrient deficiencies with up to ...

A class-conditional latent diffusion model generated high-resolution synthetic retinal images that balanced diabetic retinopathy classes and improved validation accuracy across ...

A retraining-free generative framework called EndoRare synthesizes diverse images of rare gastrointestinal lesions from a single reference image, improving AI ...

A conditional generative adversarial network tuned by an Adaptive Jaya Algorithm generated synthetic forest images that, combined with EfficientNet-B0 feature ...

A new study combines CTGAN-generated synthetic data, hybrid machine learning, SHAP explainability, reliability analysis and multi-objective optimization to design fly ...

Researchers built a low-cost pavement inspection pipeline from smartphones, a denoising diffusion model, and YOLOv11m, showing synthetic images improve alligator ...

YOLOv11 performed best in a single-center comparison using augmented clinical photographs to detect and grade nipple injuries, but the study ...

Researchers in China developed MP-CBA-YOLO, a morphology-preserving, class-balanced object detection method that grades ginseng roots with expert-level accuracy despite scarce ...

Researchers developed GDH-Det, an underwater object detection method that uses a conditional diffusion network to synthesize degraded training images and ...

Researchers used adaptive genetic and particle swarm-optimized neural networks to convert oil-based drilling cuttings into Na-X zeolite with record ammonia ...

Schizophrenia and bipolar disorder have haunted psychiatry with a diagnostic dilemma that is as old as the disciplines themselves. Both ...

A comprehensive new review traces how diffusion models, the generative engines behind modern image synthesis, have become the dominant force ...

A new study shows that synthetic speech generated by large language models significantly improves speech recognition for low-resource Estonian and ...

Researchers combined generative AI data augmentation with an intrinsically interpretable machine-learning model to predict perinatal psychological distress among Aboriginal mothers ...

A systematic 264-experiment study in China found that a Wasserstein GAN augmenting just 76 soil samples improved drone-based soil organic ...

Researchers have developed a deep reinforcement learning framework that generates synthetic patient data and explicitly avoids high-risk clinical states, outperforming ...

Researchers trained AI models on short Spanish first-contact clinical notes and found the algorithms detected suicidal behavior more accurately than ...

Researchers combined machine learning with GAN and SMOGN data augmentation to accurately predict the tensile properties of sub-sized nuclear steel ...

Researchers have built SpudNet5-R3, a lightweight explainable convolutional neural network that detects potato early and late blight with up to ...

Researchers have developed an unsupervised domain adaptation method that lets dialogue summarization models learn from annotated news data and bridge ...