Jinyou Hu’s research while affiliated with China Agricultural University and other places

... Many researchers have developed traceability system based on wireless network, which can realize ambient and quality data capture and trace for tilapia cold chain (Cheng, Wang, and Xie 2023;Mu et al. 2021;Zhang, Huang, et al. 2022;Zhang, Sun, et al. 2022). However, optimization and effectively monitoring the whole tilapia cold chain using intelligent sensing devices is the key to improve the quality of tilapia (Zhang et al. 2024;He et al. 2021;Huang, Wang, Zhang, et al. 2023). ...

Reference:

Reliable Quality Traceability for Tilapia Cold Chain Using Blockchain and Machine Learning Techniques

... The study employed random forests, gradient boosting machines, ANNs, and regularised linear regression to predict RR, VT, and eye temperature with 13 predictor variables from three different dimensions, such as production, cowrelated, and environmental factors [132]. In another study, PVDF flexible piezoelectric sensors were used to record respiratory rhythm in Hu sheep, and the technology provides potential technical support for future health monitoring and early prediction of diseases in large farm animals [133]. ...

Reference:

Applications of Artificial Intelligence for Heat Stress Management in Ruminant Livestock

... 48 Light GBM efficiently handles large datasets with fast prediction speeds, while KNN, a non-parametric method, adapts to both linear and nonlinear data structures by classifying or predicting based on proximity. 49 The performance of the prediction models was evaluated using RMSE, the ratio of prediction to deviation (RPD) and the coefficient of determination (R 2 ) obtained in the calibration and prediction or validation set. A calibration model with R 2 > 0.9 and RPD > 2.5 indicates very good prediction and is suitable for quality control applications. ...

Reference:

Non‐destructive measurement of eggshell strength using NIR spectroscopy and explainable artificial intelligence

... Zhang et al. introduced a non-destructive testing system based on flexible bioimpedance for characterizing the quality of salmon through online monitoring of bioimpedance signals, as well as changes in ambient temperature and relative humidity [55]. Xia et al. developed a highly integrated, low-cost wearable electronic system for live fish, measuring the stress response of sturgeon with high accuracy, aligning with stress level assessments [56]. ...

Reference:

Electrical impedance spectroscopy for non-destructive meat freshness assessment

... In addition, it is necessary to use appropriate modeling methods to process multiscale features data for quality grading (L. Zhang et al., 2023b), as shown in Figure 1d. In recent years, methods such as partial least squares discriminant analysis (PLS-DA) (Yuan et al., 2022), support vector machine (SVM) (Wu et al., 2016), and convolutional neural network (CNN) (Guo et al., 2021;Ju et al., 2022) have been developed for defect classification of jujube. ...

Reference:

Abnormal phenotypic defects detection of jujube using explainable machine learning enhanced computer vision

... There are many parameters such as soil electrical conductivity [143], pestici dues in fresh fruits and vegetables [144], toxins [145], phenolic content [146], an quality [147][148][149] that need to be detected and monitored via chemical analysis or contact measurement in smart agriculture. The electronic tongue has integrated chemical sensors to sense and monitor the different parameters in agricultural e ments and food quality and safety inspired by the human gustatory system [150-15 [128]. ...

Reference:

Neuromorphic Computing for Smart Agriculture

... Traditional non-destructive testing methods, i.e. x-ray tomography [6], infrared detection [7], visual inspection [8], ultrasonic detection [9], and acoustic emission (AE) monitoring [10], are commonly used for defect characterization, which helps to identify and evaluate the defects for AM products during or after the manufacturing procedure. However, those methods are difficult to compromise between real-time detection and detection efficiency. ...

Reference:

Spectral characterization method for wire-arc additive manufacturing monitoring with broadband AE signals

... Recent research emphasises the integration of AI with devices to monitor and predict freshness levels, offering significant benefits in precision, scalability, and automation (Feng et al. 2020;Prema and Visumathi 2022;Li et al. 2022). By utilising these technologies, the seafood industry will be able to minimise food waste and ensure consumer safety. ...

Reference:

Quality Determination of Frozen-Thawed Shrimp Using Machine Learning Algorithms Powered by Explainable Artificial Intelligence

... Moreover, AI can optimize water quality parameters based on specific fish species requirements. Researchers have developed AI models for water quality index prediction and classification, employing techniques like artificial neural networks and machine learning algorithms to enhance aquaculture water quality management [37][38][39]. Their findings demonstrated that the proposed models can effectively forecast the Water Quality Index (WQI) and categorize water quality with high reliability [40]. ...

Reference:

Unlocking the Potential: Artificial Intelligence Applications in Aquaculture Greenhouse Development

... With the rapid development of e-commerce, the demand in the logistics industry continues to grow [1,2], which promotes the rapid development of logistics packaging automation technology [3]. As a key link in commodity distribution, the quality and efficiency of logistics packaging directly affect the operational efficiency of the supply chain and the shopping experience of consumers [4,5]. Traditional manual inspection methods are inadequate when facing large-scale and highfrequency logistics packaging and are easily affected by human factors [6][7][8][9]. ...

Reference:

Automated Logistics Packaging Inspection Based on Deep Learning and Computer Vision: A Two-Dimensional Flow Model Approach