The preparation of active food packaging materials with both high gas barrier and antioxidant properties remains a great challenge. Hemicellulose is an abundant raw material with excellent oxygen barrier performance and biodegradability, and tea polyphenols (TPs) are natural antioxidants with the function of reducing water sensitivity of polysaccharide-based films. Herein, a synergistic strategy is reported for the preparation of active-controlled food packaging material by incorporating TP into the poly(vinyl alcohol) hemicellulose (PVA/HC) matrix. The obtained film with 5 wt% addition of TP exhibited the highest tensile strength of 41.09 MPa, which was increased by 84% than that of the PVA/HC film. Meanwhile, water vapor permeability was decreased by 25% to 7.39 × 10−12 g⋅cm/(cm2⋅s⋅Pa) compared to a PVA/HC film. Moreover, the addition of TP improved the thermal stability of the (PVA/HC/TP)n film. The outstanding packaging performances were attributed to the robust hydrogen bonding between TP and the PVA/HC matrix as demonstrated by Fourier transform infrared spectroscopy. The DPPH free radical scavenging of (PVA/HC/TP)15 films reached 87.99%, and the film also exhibited enhanced inhibition efficiency against the microorganism E. coli. Compared with the polyethylene plastic wrap, the film showed excellent food packaging applicability by preventing air oxidation of fresh-cut potato. This study sheds light on a design strategy for active packaging films using a natural substance and easily available biomass.
Deep learning (DL) has advanced computer-aided diagnosis, yet the limited data available at local medical centers and privacy concerns associated with centralized AI approaches hinder collaboration. Federated learning (FL) offers a privacy-preserving solution by enabling distributed DL training across multiple medical centers without sharing raw data. This article reviews research conducted from 2016 to 2024 on the use of FL in cancer detection and diagnosis, aiming to provide an overview of the field’s development. Studies show that FL effectively addresses privacy concerns in DL training across centers. Future research should focus on tackling data heterogeneity and domain adaptation to enhance the robustness of FL in clinical settings. Improving the interpretability and privacy of FL is crucial for building trust. This review promotes FL adoption and continued research to advance cancer detection and diagnosis and improve patient outcomes.
Color accuracy in conventional and digital printing processes relies on press characterization to establish the relationship between input device values and output colorimetric or spectral reflectance values. Conventional models, such as Murray–Davies, Clapper–Yule, Yule–Nielsen, and Yule–Nielsen modified spectral Neugebauer, are renowned for providing accurate chromatic and spectral predictions. However, they fall short of accounting for the effects of black ink use and struggle to predict light hues accurately. In order to predict more accurately the color fingerprint, spectral reflectance, of halftone printed images, this study introduces a novel machine-learning-based deep neural network combined with the improved particle swarm optimization algorithm. This enables implementing a spectral reflectance color prediction model for CMYK printing, which eliminates the need to adjust for dot gain during printing. By evaluating this model on a lithographic offset press, we demonstrate its superior performance evidenced by significantly lower root mean square error and color difference (ΔE ∗ 00) values compared to existing methods. This approach minimizes color deviations during printing and reduces material and energy consumption, thereby ultimately enhancing the quality of printed materials.
This study proposes an improved spectral reflectance reconstruction method to convert the response data captured by RGB digital cameras to the object surface spectral reflectance. Additionally, a system noise model is also proposed, which incorporates both signal-dependent and signal-independent components, thus rendering it more closely aligned with real-world conditions. Image data captured with RGB digital cameras under multiple LED light sources, and the inverse distance of the Euclidean distance between the training sample test samples in RGB color space is used as a weighting coefficient to reconstruct the spectral reflectance of the object surface, as accurately as possible. Experimental results show that the proposed method has better reconstruction accuracy than existing methods. The root mean square error values and the reconstructed goodness of fit coefficient higher than 0.9958 indicate that the spectral reconstruction performance has greatly improved compared to existing methods, thus proving the validity of the proposed weighting method.
Deep neural networks (DNNs) have heavily relied on traditional computational units, such as CPUs and GPUs. However, this conventional approach brings significant computational burden, latency issues, and high power consumption, limiting their effectiveness. This has sparked the need for lightweight networks such as ExtremeC3Net. Meanwhile, there have been notable advancements in optical computational units, particularly with metamaterials, offering the exciting prospect of energy-efficient neural networks operating at the speed of light. Yet, the digital design of metamaterial neural networks (MNNs) faces precision, noise, and bandwidth challenges, limiting their application to intuitive tasks and low-resolution images. In this study, we proposed a large kernel lightweight segmentation model, ExtremeMETA. Based on ExtremeC3Net, our proposed model, ExtremeMETA maximized the ability of the first convolution layer by exploring a larger convolution kernel and multiple processing paths. With the large kernel convolution model, we extended the optic neural network application boundary to the segmentation task. To further lighten the computation burden of the digital processing part, a set of model compression methods was applied to improve model efficiency in the inference stage. The experimental results on three publicly available datasets demonstrated that the optimized efficient design improved segmentation performance from 92.45 to 95.97 on mIoU while reducing computational FLOPs from 461.07 MMacs to 166.03 MMacs. The large kernel lightweight model ExtremeMETA showcased the hybrid design’s ability on complex tasks.
This article resolutely uses the concept of feature fusion to establish a deep learning model that can quickly recognize objects and complete an anti-counterfeit label recognition system. The receiver combines the training of the technology acceptance model (TAM) to evaluate the satisfaction of users in completing the anti-counterfeit label classification training. In this study, the fusion-based recognition program was employed to extract the feature sets of different categories of anti-counterfeit labels based on the operation of multilayer convolutional neural networks (CNNs) with different depth models. Using neighborhood components analysis, ten important sets of features from different CNN models were selected and reorganized parallelly into a new small-scale feature fusion dataset. By using naive Bayes and support vector machine methods, efficient classification of wine label image feature datasets after fusion was achieved. The feature fusion anti-counterfeiting label recognition system proposed in this article had a maximum recognition accuracy of 99.29% and a data reduction compression ratio of about 1/50. In addition to reducing training time, it maintained a high level of accuracy. This study established a TAM with the advantage of a feature fusion anti-counterfeit label recognition system. The model was tested on 100 consumers, and a satisfaction evaluation and validation analysis with partial least squares structural equation modeling were completed thereafter. The efficiency of the fusion-based deep learning model met the level of consumer satisfaction. This will be beneficial for educating consumers to use and enhance their willingness to promote and repurchase wine products in the future.
With the advancement of digital printing technology, the rapid detection of inkjet defects have become critical research areas in OnePass inkjet printing. These defects can disrupt the continuity and smoothness of the image, leading to a decrease in print quality. To overcome these issues, this study proposed an inkjet defect detection method based on printing characterization. Image processing technology was used to obtain the printing characterization information of the printer and extract the information of ink point positions. Optimizing the allocation matrix through Sinkhorn algorithm and combining it with Robust Point Matching algorithm to construct the transmission objective function was accomplished to obtain the optimal point set matching model. This model serves two purposes: diagnosing the nozzle function status of each printhead and quantifying the alignment errors between printheads. Experiments demonstrated the high precision of this detection method. We analyzed the impact of related parameters on the model’s performance and assessed changes in image quality under different alignment errors. This approach provides a new solution for optimizing printer maintenance.
The bearing vibration fault monitoring of advanced printing systems is crucial for ensuring system reliability, improving print quality, and enhancing production efficiency. In consideration of the high noise and strong interference attributes of bearing vibration signals of printing equipment caused by complex environmental factors, the original data noise interference is suppressed by a diagnostic approach for rolling bearing vibration signal faults combined Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Dung Beetle Optimization-Support Vector Machine (DBO-SVM) proposed in this article. This method suppresses noise interference in the original data by decomposing and reconstructing vibration signals, and innovatively proposes the DBO-optimized SVM to address the problems of poor anti-interference ability and weak generalization ability of a single intelligent diagnostic method, effectively improving the fault diagnosis rate. First, the decomposition by CEEMDAN effectively overcomes the difficulties of modal aliasing and significant reconstruction errors found in traditional empirical mode decomposition (EMD). The optimal component reconstruction strategy considering the Correlation Coefficient and the Variance Contribution Rate is designed to obtain the vibration signal after noise reduction. Second, to effectively avoid the limitation of heavily relying on expert experience for hyperparameter adjustment, a DBO-SVM model is constructed utilizing a heuristic beetle optimization algorithm, dynamically optimizing the key kernel function parameters and penalty factors of SVM. Finally, the algorithm’s performance was tested using public datasets and self-tested data from Case Western Reserve University. The results indicate that the proposed approach achieves greater diagnostic accuracy and exhibits robust generalization.