Noise plays a critical role in determining the performance and quantitative accuracy of imaging using a scanning electron microscope (SEM). In existing SEM literature, SEM noise is often assumed to follow Gaussian or Poisson statistics. However, in practice, true statistical behavior can vary depending on the operating conditions. In this work, we present a systematic analysis of SEM noise distributions using experimental images acquired under different accelerating voltages and dwell times with the in-lens and in-chamber secondary electron (SE) detectors, as well as the back-scattered electron (BSE) detector. Statistical tests, histogram analysis, and distribution fitting were employed to identify the underlying noise distributions. The results show that, at short dwell times, SE noise is best described by a Neyman Type-A distribution, whereas BSE noise is better described by a Gamma distribution. To automate the classification of these noise distributions, a deep learning framework based on convolutional neural networks (CNN) with a Convolutional Block Attention Module (CBAM) was implemented. Pretrained architectures were evaluated and compared with the proposed model. Among the pretrained models, MobileNetV2 achieved the highest accuracy of 88%, whereas the proposed CNN-CBAM model achieved the best overall performance, with F1-scores up to 0.98 and an accuracy of 97%.
Accurate instance segmentation of electrode microstructures is essential for establishing processing–structure– property linkages and enabling automated quality assurance in lithium-ion batteries (LIBs) manufacturing. However, dense packing, strong adhesion, and multiscale morphology in LIBs ternary materials scanning electron microscopy (SEM) images often lead existing methods to oversegmentation or undersegmentation and blurred boundaries. We present LI-SEMIS (LIBs ternary materials SEM Instance Segmentation), a unified framework that augments the Segment Anything Model (SAM) with a UNETR style decoder for end to end, instance level prediction. The decoder jointly produces three complementary maps: foreground probability, distance to center, and distance to boundary, which are integrated via a watershed algorithm to yield pixel wise instance labels. Training adopts a joint optimization of Automatic Instance Segmentation (AIS) and Interactive Iterative Segmentation (IIS): simulated point/box prompts (with occasional reuse of previous masks) are progressively injected to co adapt the image/prompt encoders and the decoder. Inference integrates two key modules. Test Time Augmentation (TTA) fuses multi scale and flipped predictions to enhance global robustness, while the Corner Aware Fine Refinement (CAFRe) module performs high resolution re-prediction at image corners and employs a two dimensional Hann window feathering to smoothly blend local and global outputs, suppressing block artifacts and sharpening boundaries. On a dataset of 239 SEM images of LIBs ternary materials (split into 143/47/49 for train/val/test), LI-SEMIS consistently surpasses SAM and μSAM across ViT T/B/L backbones under SA50, SA75, and mSA. Ablations show CAFRe delivers substantial boundary fidelity and separation gains, TTA adds consistent robustness, and their combination is most effective; IIS further yields monotonic improvements with few user interactions exceeding the AIS baseline. LI-SEMIS provides a robust, generalizable solution for high precision microstructural analysis and scalable, automated inspection in LIBs ternary materials manufacturing.
Tenebrionoidea beetles are characterized by distinctive sperm ultrastructure, including high sperm counts per cyst and the unique cyst organization with antiparallel spermatid orientation. The sperm ultrastructure in Melandryidae, Pyrochroidae, and Zopheridae has revealed these conserved features alongside specific features. Melandrya caraboides exhibits a quite long three-layered acrosome and accessory bodies with variable shape along spermiogenesis. Notably, Pyrochroidae species exhibited large, equal mitochondrial derivatives contrasting with the other families, which show unequal size of the two mitochondrial derivatives. The tail tip has the unusual microtubule loss in the two families Melandryidae and Zopheridae. These findings enhance understanding of sperm morphology diversity within Tenebrionoidea and support its utility in phylogenetic assessments, encouraging further studies across additional families to refine evolutionary relationships.
The functional reliability of semiconductor micro/nanostructures, from photonic crystals to MEMS, relies on nanometer-scale topographic precision. While atomic force microscopy (AFM) is vital for their characterization, its metrological accuracy is often compromised by tip-convolution artifacts that blur critical features. To address this, we propose the Topography-Constrained Attention UNet (TCA-UNet). Unlike conventional pixel-wise image restoration models, TCA-UNet incorporates a specialized topographical constraint system comprising Mean Absolute Error, Vertical Total Variation, and a Laplacian curvature-consistency term. This system is designed to enforce strict geometric fidelity with the sample's intrinsic surface morphology and roughness characteristics, thus preventing the generation of over-smoothed or non-physical surface predictions. Architecturally, we optimize the encoder-decoder framework to preserve high-frequency spatial details and integrate a gated channel attention module, enabling the network to adaptively focus on morphological features severely distorted by tipsurface interactions. Evaluated on synthetic surface datasets, TCA-UNet achieved an SSIM of 0.98-0.999 depending on the structure class, with an average PSNR improvement of 4.2-6.1 dB over a standard UNet and 8.2-12.8 dB over the EDN baseline across the three test datasets. In practical AFM measurement, evaluated over five repeated scans, it reduced the deviation of a standard grating linewidth from a sharp-tip reference from 44 nm to 5 nm. The topographical reconstruction error (MSE) of complex cone-array patterned sapphire substrates (PSS) was reduced by 52.3% over five regions, from 8.968 +/- 0.008-4.274 +/- 0.003 nm2. The method is scale-agnostic, in that the network operates in pixel space, but shape-specific: the profile family of the sample must have been covered during training. Our framework provides a geometry-aware approach to suppressing routine imaging artifacts, supporting quantitative characterization of semiconductor devices via AFM within the stated applicability envelope.
