Results 61 to 70 of about 5,165 (221)
Multiscale Superpixel-Based Sparse Representation for Hyperspectral Image Classification †
Recently, superpixel segmentation has been proven to be a powerful tool for hyperspectral image (HSI) classification. Nonetheless, the selection of the optimal superpixel size is a nontrivial task.
Shuzhen Zhang +3 more
doaj +1 more source
Advancing Photonic Inverse Design with Interpretable Machine Learning
The work applies the interpretable machine learning technique called LIME (local interpretable model‐agnostic explanations) to the inverse design of photonic chips, revealing hidden optimization patterns and guiding better starting designs. Using insights from LIME improves performance of two‐mode multiplexers, showing interpretable methods can ...
Lirandë Pira +5 more
wiley +1 more source
Image segmentation using superpixel ensembles [PDF]
Recently there has been an increasing interest in image segmentation due to the needs of locating objects with high segmentation accuracy as required by many computer vision and image processing tasks.
Gu, Xianbin
core +1 more source
Tablet‐based handwriting tasks (spiral, meander, and wave) are transformed into unified images and analyzed using PD‐MGMA‐DSCNN, a lightweight multiscale gated attention network. Bayesian–genetic optimization improves performance, while SHAP attribution maps provide interpretable handwriting biomarkers for Parkinson's disease screening.
Khosro Rezaee, Ali Khalili Fakhrabadi
wiley +1 more source
Local Binary Patterns and Superpixel-Based Multiple Kernels for Hyperspectral Image Classification
The superpixel-based multiple kernels model uses the average value of all pixels within superpixel as the spatial feature, which results in inaccurate extraction of edge pixels. To solve this problem, a local binary patterns and superpixel-based multiple
Wei Huang +4 more
doaj +1 more source
Pupil Plane Multiplexing for Vectorial Fourier Ptychography
This study proposes a cost‐effective, modality‐adaptive multichannel microscopy framework using pupil‐plane multiplexing. A custom pupil aperture at the Fourier plane encodes channel‐specific transfer functions with spectral or polarization filters, and model‐based reconstruction with channel‐dependent priors decodes them.
Hyesuk Chae +5 more
wiley +1 more source
Lung segmentation of chest X-ray (CXR) images is a fundamental step in many diagnostic applications. Most lung field segmentation methods reduce the image size to speed up the subsequent processing time.
Chien-Cheng Lee +3 more
doaj +1 more source
Correction to: Rooted Spanning Superpixels [PDF]
zbMATH Open Web Interface contents unavailable due to conflicting licenses.
openaire +1 more source
This study presents an interpretable, lightweight hybrid deep learning model for real‐time analysis of breast cancer histopathology in IoMT‐enabled diagnostic systems. By integrating MobileNetV2 and EfficientNet‐B0 with a novel contextual recurrent attention module (CRAM), the framework achieves near‐perfect accuracy while providing transparent Grad ...
Roseline Oluwaseun Ogundokun +4 more
wiley +1 more source

