An Improved Residual Network for Pork Freshness Detection Using Near-Infrared Spectroscopy

An effective deep learning fusion method for predicting the TVB-N and TVC contents of chicken breasts using dual hyperspectral imaging systems

Total volatile basic nitrogen (TVB-N) and total viable count (TVC) are important freshness indicators of meat. Hyperspectral imaging combined with chemometrics has been proven to be effective in meat detection. However, a challenge with chemometrics is the lack of a universally applicable processing combination, requiring trial-and-error experiments with different datasets. This study proposes an end-to-end deep learning model, pyramid attention features fusion model (PAFFM), integrating CNN, attention mechanism and pyramid structure. PAFFM fuses the raw visible and near-infrared range (VNIR) and shortwave near-infrared range (SWIR) spectral data for predicting TVB-N and TVC in chicken breasts. Compared with the CNN and chemometric models, PAFFM obtains excellent results without a complicated processing combinatorial optimization process. Important wavelengths that contributed significantly to PAFFM performance are visualized and interpreted. This study offers valuable references and technical support for the market application of spectral detection, benefiting related research and practical fields.

Combination of near-infrared spectroscopy with Wasserstein generative adversarial networks for rapidly detecting raw material quality for formula products

Near-infrared spectroscopy (NIRS) has emerged as a key technique for rapid quality detection owing to its fast, non-destructive, and eco-friendly characteristics. However, its practical implementation within the formulation industry is challenging owing to insufficient data, which renders model fitting difficult. The complexity of acquiring spectra and spectral reference values results in limited spectral data, aggravating the problem of low generalization, which diminishes model performance. To address this problem, we introduce what we believe to be a novel approach combining NIRS with Wasserstein generative adversarial networks (WGANs). Specifically, spectral data are collected from representative samples of raw material provided by a formula enterprise. Then, the WGAN augments the database by generating synthetic data resembling the raw spectral data. Finally, we establish various prediction models using the PLSR, SVR, LightGBM, and XGBoost algorithms. Experimental results show the NIRS-WGAN method significantly improves the performance of prediction models, with R2 and RMSE of 0.949 and 1.415 for the chemical components of sugar, respectively, and 0.922 and 0.243 for nicotine. The proposed framework effectively enhances the predictive capabilities of various models, addressing the issue caused by limited training data in NIRS prediction tasks.

BR-Net: Band reweighted network for quantitative analysis of rapeseed protein spectroscopy

Compared with the complexity of chemical methods, near-infrared spectroscopy (NIRS) is widely used in the detection of protein content because of its advantages of being fast and non-destructive. Aiming to tackle the problem that the raw near-infrared spectroscopy contains many redundant wavelengths, which affects the accuracy of quantitative prediction and requires expertise to process, we propose an end-to-end network: Band Reweighted Network (BR-Net) that automates wavelength reweighted and quantitative prediction of protein content in rapeseed. Unlike extracting part of wavelengths by the traditional wavelength selection methods, BR-Net retains all spectral wavelengths and assigns different weights to the wavelengths to express the correlation with the corresponding concentration, which enables wavelength selection without ignoring the information contained in the less relevant wavelengths. We compare BR-Net with traditional selection methods such as SPA, LARS, CARS, and UVE to verify its efficiency and robustness, finding that the R² of the training set and test set are 0.9797 and 0.9215, the RMSEC and RMSEP are 0.4053 and 0.8501, respectively, and the RPD is 3.5686, which prove BR-Net outperforms all the traditional methods. The network described here is universally applicable to a variety of NIR quantitative analyses.

DeepDA-Ace: A Novel Domain Adaptation Method for Species-Specific Acetylation Site Prediction

Protein lysine acetylation is an important type of post-translational modification (PTM), and it plays a crucial role in various cellular processes. Recently, although many researchers have focused on developing tools for acetylation site prediction based on computational methods, most of these tools are based on traditional machine learning algorithms for acetylation site prediction without species specificity, still maintained as a single prediction model. Recent studies have shown that the acetylation sites of distinct species have evident location-specific differences; however, there is currently no integrated prediction model that can effectively predict acetylation sites cross all species. Therefore, to enhance the scope of species-specific level, it is necessary to establish a framework for species-specific acetylation site prediction. In this work, we propose a domain adaptation framework DeepDA-Ace for species-specific acetylation site prediction, including Rattus norvegicus, Schistosoma japonicum, Arabidopsis thaliana, and other types of species. In DeepDA-Ace, an attention based densely connected convolutional neural network is designed to capture sequence features, and the semantic adversarial learning strategy is proposed to align features of different species so as to achieve knowledge transfer. The DeepDA-Ace outperformed both the general prediction model and fine-tuning based species-specific model across most types of species. The experiment results have demonstrated that DeepDA-Ace is superior to the general and fine-tuning methods, and its precision exceeds 0.75 on most species. In addition, our method achieves at least 5% improvement over the existing acetylation prediction tools.

Intelligent Measurement of Coal Moisture Based on Microwave Spectrum via Distance-Weighted kNN

Realizing the rapid measurement of coal moisture content (MC) is of great significance. However, existing measurement methods are time-consuming and damage the original properties of the samples. To address these concerns, a coal MC intelligent measurement system is designed in this study that integrates microwave spectrum analysis and the distance-weighted k-nearest neighbor (DW-kNN) algorithm to realize rapid and non-destructive measurement of coal MC. Specifically, the measurement system is built using portable microwave analysis equipment, which can efficiently collect the microwave signals of coal. To improve the cleanliness of modeling data, an iterative clipping method based on Mahalanobis distance (MD-ICM) is used to detect and eliminate outliers. Based on multiple microwave frequency bands, various machine learning methods are evaluated, and it is found that coal MC measurement using broad frequency signals of 8.05–12.01 GHz yields the best results. Experiments are also carried out on coals from different regions to examine the regional robustness of the proposed method. The results of on-site testing with 27 additional samples show that the method based on the combination of microwave spectrum analysis and DW-kNN has a potential application prospect in the rapid measurement of coal MC.

Deep Residual Network for Smartwatch-Based User Identification through Complex Hand Movements

Wearable technology has advanced significantly and is now used in various entertainment and business contexts. Authentication methods could be trustworthy, transparent, and non-intrusive to guarantee that users can engage in online communications without consequences. An authentication system on a security framework starts with a process for identifying the user to ensure that the user is permitted. Establishing and verifying an individual’s appearance usually requires a lot of effort. Recent years have seen an increase in the usage of activity-based user identification systems to identify individuals. Despite this, there has not been much research into how complex hand movements can be used to determine the identity of an individual. This research used a one-dimensional residual network with squeeze-and-excitation (SE) configurations called the 1D-ResNet-SE model to investigate hand movements and user identification. According to the findings, the SE modules have enhanced the one-dimensional residual network’s identification ability. As a deep learning model, the proposed methodology is capable of effectively identifying features from the input smartwatch sensor and could be utilized as an end-to-end model to clarify the modeling process. The 1D-ResNet-SE identification model is superior to the other models. Hand movement assessment based on deep learning is an effective technique to identify smartwatch users.