Unbiased AI-powered news
4 stories related to this topic, newest first.
9to5mac.comA new convolutional network called MSFNet-PD combines progressive dilation and multi-scale feature fusion to detect surface defects on strip steel. The model aims to improve both detection accuracy and processing speed for industrial inspection lines. Tests on the SD-Saliency-900…
link.springer.comA study published on May 10, 2026, introduced a drilled-to-bossed joint geometry for direct drive friction welding of AA2024-T351 aluminum alloy. The approach achieved tensile strength of 538 MPa and torsional strength of 325 MPa using specific process parameters. Deep learning m…
medium.comResearchers from Saudi Arabia, India and Ethiopia unveiled a new architecture combining Hilbert-Huang transform feature extraction, path-informed adaptive modeling and chaotic optimization that achieved 99 percent validation accuracy on two EV datasets. The paper, received in Dec…
Substrate placeholder — needs reviewNature has released two separate research papers. One employs multiomics and deep learning to analyze regulatory syntax in human development. The other uses satellite imagery to examine volatility in human night-time activity patterns.