Rapid and non-destructive authentication of herbal medicines is important for quality control and market supervision. This study established an interpretable spectral evidence learning framework for visible and near-infrared (Vis/NIR) imaging-based authentication of Codonopsis Radix (CR) and Aurantii Fructus (AF). Compact 31-band mean gray-value spectra were analyzed at ROI and sample levels. CR sample-level spectra were obtained by ROI-group averaging, whereas AF records were retained as individual sample spectra with image-group information used for leakage-controlled validation. Raw spectra, Savitzky–Golay smoothing, multiplicative scatter correction, and standard normal variate correction were compared with machine-learning and deep-learning classifiers. A fold-contained lightweight diffusion (LD) module was further introduced to provide class-conditioned spectral augmentation and denoising-error evidence. Under grouped cross-validation, the strongest non-LD Linear SVM models achieved accuracy/macro-F1 values of 0.9231/0.9238 for CR and 0.9025/0.9018 for AF. After LD augmentation, the best LD-augmented SVM models reached macro-F1 values of 0.9427 and 0.9197, respectively. Across all evaluated model–dataset combinations, LD increased the overall mean macro-F1 from 0.7302 to 0.8189. Model-aligned wavelength evidence and top-wavelength subset tests further showed that selected LED-band subsets retained useful discriminative information within the present imaging configuration. These results support the feasibility of compact Vis/NIR image-based authentication of herbal materials under grouped validation.

Down quality is the core evaluation indicator of thermal insulation products, and its grade determination strictly complies with the down content index specified in the national standard GB/T 17685-2016 Feather and Down. Traditional down quality inspection adopts manual sorting and weighing methods, which are plagued by low efficiency, strong subjectivity and high error rates, thereby restricting the intelligent upgrading of the down industry. This study aims to develop an automatic down detection and quantitative grading method conforming to national standards based on deep learning. A down dataset consisting of 632 RGB images is constructed, with each image containing 5–10 individual down samples and covering five categories: mature down clusters, immature down clusters, down filaments, feathers, and yellow-tail down. Three mainstream frameworks including YOLOv8, YOLOv11 and YOLOv26 are trained for performance comparison. Precision, recall, mAP@50 and mAP@50-95 are adopted as evaluation metrics. In addition, this paper proposes a research idea for down content calculation and automatic classification and grading of down quality in accordance with relevant national standards. The experimental results demonstrate that the latest models do not necessarily achieve the optimal performance. The newly released YOLOv26n and YOLOv26m exhibit relatively low accuracy in the down detection task, with mAP@50 values of only 0.98556 and 0.99077, and recall rates of 0.95032 and 0.97848, respectively, failing to outperform their previous-generation counterparts. In contrast, YOLOv11n achieves the best comprehensive performance, with an mAP@50 of 0.99416, a precision of 0.99544, a recall of 0.99722, and an mAP@50-95 of 0.63464. Meanwhile, the model has only 2.58 M parameters, a computational complexity of 6.3 GFLOPs, and a single training time of approximately 6.7 min, achieving an optimal balance between detection accuracy and computational efficiency. All models show the highest detection accuracy for mature down clusters and yellow-tailed down, while slight confusion exists between immature down clusters and down filaments. This study verifies the feasibility of the YOLO series models in down quality inspection in accordance with national standards, and reveals that model architecture iteration does not necessarily lead to performance improvement on specific industrial datasets. The lightweight and robustly designed YOLOv11n presents greater practical value. The intelligent detection scheme proposed in this paper can assist in optimizing the traditional manual quality inspection workflow, alleviating the burden of manual counting and reducing subjective errors. It provides new ideas and technical references for the rapid screening and objective determination of down quality. Furthermore, the proposed research framework for automatic classification and grading of down quality is expected to promote the development of down quality inspection toward standardization, intelligence, and automation in the future.

Manual sorting is still the mainstream scheme for component identification in down feather quality evaluation, which suffers from low detection efficiency and poor stability. To address these drawbacks, this paper proposes a fine-grained component detection algorithm based on YOLO12 for down feather quality classification. Five typical down feather components are selected as detection targets, including discolored feathers, down filaments, immature down, feathers and pure down. A dedicated down feather object detection dataset consisting of 1140 images is established accordingly. To improve the feature representation capacity of the model for tiny objects, faint boundary features and subtle distinctions between analogous categories, this work integrates the Convolutional Gated Linear Unit (CGLU) into the A2C2f module of YOLO12. While maintaining the original feature aggregation pathways and residual architecture, the conventional MLP feed-forward branch within ABlock is replaced with convolutional gated transformation. Experimental results demonstrate that the proposed A2C2f-CGLU model achieves precision of 96.50%, recall of 94.49%, mAP50 of 98.05% and mAP50-95 of 57.89% with the optimal weights on the validation set. Compared with the original YOLO12, the mAP50-95 metric is elevated by 3.04 percentage points, and the overall performance surpasses two comparative variants, A2C2f-DFFN and A2C2f-KAN. Visualizations of PR curves, confusion matrices and real test samples validate that the proposed method effectively enhances the recognition stability of tiny down feather targets and similar classes. This research provides a visual inspection foundation for subsequent component proportion calculation, quality grade discrimination and the development of intelligent detection systems.

Geographical authentication of medicinal plant materials remains challenging because multispectral variables are often highly collinear and sample grouping can complicate reliable model validation. Existing correlation-based feature-selection strategies also require careful adaptation to multiclass problems to avoid artificial ordering of class labels and information leakage during model development. Therefore, this study aimed to develop a compact and leakage-controlled multispectral learning framework for geographical-origin discrimination. This study analyzed 800 physical Pinellia ternata samples from Gansu Xihe, Sichuan Neijiang, Sichuan Chengdu, and Chongqing Dianjiang (200 samples per origin). Each physical sample was represented by 31 mean grayscale intensities calculated from Otsu-segmented multispectral regions of interest. A Pearson-correlation-guided deep multilayer perceptron (PCG-DeepMLP) was constructed by estimating one-vs-rest band relevance and inter-band redundancy only within the training data. The key methodological innovation is a unified multiclass-aware, relevance–redundancy spectral-learning framework in which class-specific one-vs-rest Pearson relevance is coupled with inter-band redundancy control and embedded within leakage-controlled grouped model development. By learning the spectral subset exclusively from each training partition before nonlinear classification, the framework produces compact and complementary multispectral signatures while preserving multiclass structure and strict independence of held-out groups. Model and feature-selection settings were chosen by three-fold grouped cross-validation within each training partition. PCG-DeepMLP retained 9–21 bands and achieved the highest mean accuracy (0.9812 ± 0.0135), macro-F1 (0.9812 ± 0.0135), Matthews correlation coefficient (MCC; 0.9752 ± 0.0179), and macro-AUC (0.9994 ± 0.0006) among seven models. Its macro-F1 was higher than that of 1D-CNN, 1D-ResNet, full-band MLP, PLS-DA, and random forest after Holm correction. Performance was estimated through a strict nested group-wise internal validation scheme, with every outer test fold remaining isolated from feature selection, preprocessing, and model optimization. These findings demonstrate that multiclass-aware relevance–redundancy learning can retain complementary Pinellia ternata origin-discriminative information in a compact and stable spectral representation, enabling accurate geographical-origin authentication while providing a principled basis for reduced-channel acquisition and future independent multi-batch validation.

To combat counterfeiting in down products and enhance enterprise credibility through technical means, this paper proposes a blockchain-based down quality traceability framework named DPT (down-product traceability). Built on Hyperledger Fabric, the framework integrates the InterPlanetary File System (IPFS), digital anti-counterfeiting watermarks (DW), QR codes, and category-specific encryption strategies to establish a trusted data chain throughout the entire supply chain. Role-based access control (RBAC) is adopted to ensure the secure submission and query of traceability information by all supply chain participants. A trinity data storage architecture is designed to achieve secure and efficient data management. A full-fledged application system was developed and deployed in cooperation with a leading down products enterprise to validate its practical applicability. Performance evaluation using Hyperledger Caliper 0.6.0, which focuses on throughput, latency, and resource utilization under stress testing, confirms that the DPT framework meets the requirements of real-world production. Furthermore, practical sales data verify that the proposed system effectively enhances consumer trust and mitigates counterfeiting behaviors in the market. Future work will focus on further optimizing write operation performance and evolving the system into a more robust clustered architecture.

Multispectral imaging systems often exhibit wavelength-dependent focus responses because of chromatic aberration, nonuniform spectral sensitivity, and scene-dependent reflectance, making single-band autofocus inadequate for optimizing the complete spectral image cube. This study proposes a nonlinear band-correlation guided deep reinforcement learning framework for autofocus optimization in a self-developed 31-band multispectral imaging system covering 360–980 nm. Through-focus response curves are first extracted from all spectral channels, and a nonlinear inter-band dependency matrix is constructed to characterize complementary and redundant focusing information across wavelengths. A compact subset of representative bands is then selected and embedded into the reinforcement-learning state to guide closed-loop motor control. The agent jointly determines focusing direction, displacement, and stopping time while maximizing a global multispectral focus objective that considers mean sharpness, inter-band consistency, and worst-band degradation. During online autofocus, only the representative bands are acquired at intermediate positions, whereas the complete 31-band cube is captured after convergence. The framework is evaluated using standardized optical targets and spectrally heterogeneous scenes through focal-position error, MTF, full-band focus retention, acquisition time, and robustness. The proposed design provides an efficient computational-photonics solution for broadband multispectral autofocus with reduced band-acquisition and motor-search overhead.