Jin Zeng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jin Zeng has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
Jingjing Xia
8 shared publications- Tian Deng
School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan 430000, China
3 shared publications Libing Liu
School of Computer Science, Hubei University of Technology, Wuhan, P. R. China
2 shared publications- Wenbo Xu
Wuhan University of Technology
2 shared publications Yutong Wang
School of Computer Science, Hubei University of Technology, Wuhan, P. R. China
2 shared publicationsZiyan Li
School of Computer Science, Hubei University of Technology, Wuhan, P. R. China
2 shared publications
Latest Publications
- 1 September 2026
- Published by IOP Publishing in Engineering Research Express
- Vol. 8 (18) , 185218
Multi-scale dual-path cascaded attention network for skin lesion classification
Skin cancer represents a major global health concern, where early detection is critical for improving patient outcomes. Despite considerable promise in dermoscopic image analysis, existing deep learning methods share three persistent limitations: single-stream architectures cannot simultaneously capture global lesion morphology and local fine-grained details; fixed-receptive-field convolutions fail to handle wide intra-class scale variation; and parallel attention modules lack a structured hierarchical refinement logic. To address these issues, we propose the multi-scale dual-path cascaded attention network (MDCAN). A dual-path architecture explicitly decouples global semantic and local detail representations, while a cascaded attention module (MSCA) enforces a sequential ‘channel → spatial → scale’ refinement pipeline within each path. Depthwise separable and dilated convolutions are incorporated to capture information across different receptive fields while limiting additional computational overhead. Evaluated on the ISIC2017, ISIC2019, and HAM10000 datasets over five independent runs, MDCAN achieved mean accuracies of 89.25 ± 0.22%, 91.82 ± 0.32%, and 93.81 ± 0.17%, respectively. With 17.49 M parameters, MDCAN demonstrates competitive classification performance and a favorable trade-off between classification accuracy and model complexity.
- 1 September 2026
- Published by IOP Publishing in Engineering Research Express
- Vol. 8 (18) , 185218
Multi-scale dual-path cascaded attention network for skin lesion classification
Skin cancer represents a major global health concern, where early detection is critical for improving patient outcomes. Despite considerable promise in dermoscopic image analysis, existing deep learning methods share three persistent limitations: single-stream architectures cannot simultaneously capture global lesion morphology and local fine-grained details; fixed-receptive-field convolutions fail to handle wide intra-class scale variation; and parallel attention modules lack a structured hierarchical refinement logic. To address these issues, we propose the multi-scale dual-path cascaded attention network (MDCAN). A dual-path architecture explicitly decouples global semantic and local detail representations, while a cascaded attention module (MSCA) enforces a sequential ‘channel → spatial → scale’ refinement pipeline within each path. Depthwise separable and dilated convolutions are incorporated to capture information across different receptive fields while limiting additional computational overhead. Evaluated on the ISIC2017, ISIC2019, and HAM10000 datasets over five independent runs, MDCAN achieved mean accuracies of 89.25 ± 0.22%, 91.82 ± 0.32%, and 93.81 ± 0.17%, respectively. With 17.49 M parameters, MDCAN demonstrates competitive classification performance and a favorable trade-off between classification accuracy and model complexity.
- 29 August 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (11) , 500
Prior-calibrated spatial-frequency Retinex decomposition for unsupervised low-light image enhancement
- 29 August 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (11) , 500
Prior-calibrated spatial-frequency Retinex decomposition for unsupervised low-light image enhancement
- 10 June 2026
- Published by Springer Nature in Water Resources Management
- Vol. 40 (9)
Improving Water Quality Evaluation Using an Optimized Multi-Output Support Vector Machine
- 10 June 2026
- Published by Springer Nature in Water Resources Management
- Vol. 40 (9)
Improving Water Quality Evaluation Using an Optimized Multi-Output Support Vector Machine
- Open Access
- 1 February 2026
- Published by Springer Nature in Environmental Earth Sciences
- Vol. 85 (3) , 89
Research on water quality evaluation method in the middle reaches of the Yangtze river based on multi-model fusion
- 1 February 2026
- Published by Springer Nature in Environmental Earth Sciences
- Vol. 85 (3) , 89
Open AccessResearch on water quality evaluation method in the middle reaches of the Yangtze river based on multi-model fusion
- 31 October 2025
- Published by Springer Nature in Environmental Geochemistry and Health
- Vol. 47 (12) , 537
Prediction of water quality in the middle area of Yangtze River using efficient machine learning model
The Yangtze River, as the longest river in China and the third-longest in the world, holds immense significance for the country’s ecological security and sustainable development. The water quality in its middle reaches directly impacts millions of people’s drinking water safety, the stability of aquatic ecosystems, and the sustainable development of the middle and lower reaches. This study addresses the need for real-time water quality monitoring in this area by constructing a comprehensive evaluation framework that synergizes the traditional Water Quality Index (WQI) with an entropy-weighted method (EWQI). To improve the prediction model’s efficiency and accuracy, environmental factors were screened using Pearson correlation analysis, and the related environmental factors were selected as the input variables of the prediction model. The performance of the three prediction models were evaluated using root mean square error (RMSE). The results indicated that after reducing the dimensionality of the environmental factors and applying support vector machines (SVM), the maximum RMSE was only 1.61 in the testing process. To address the single-output limitation of SVM, the multi-output support vector machine (MSVM) can be achieved by modifying its internal structure, which enables the concurrent prediction of EWQI and WQI. Furthermore, the grey wolf optimizer (GWO) was integrated to enhance the prediction accuracy of the model. Experimental results demonstrated that the proposed GWO-MSVM model achieved superior prediction performance in the study area, with a maximum RMSE of 1.40, a minimum coefficient of determination (R2) of 0.92, and a maximum mean absolute error (MAE) of 1.04 in the testing process. The research provides a reliable technical framework and strategic support for water quality assessment and ecological conservation in the Yangtze River basin.
- 31 October 2025
- Published by Springer Nature in Environmental Geochemistry and Health
- Vol. 47 (12) , 537
Prediction of water quality in the middle area of Yangtze River using efficient machine learning model
The Yangtze River, as the longest river in China and the third-longest in the world, holds immense significance for the country’s ecological security and sustainable development. The water quality in its middle reaches directly impacts millions of people’s drinking water safety, the stability of aquatic ecosystems, and the sustainable development of the middle and lower reaches. This study addresses the need for real-time water quality monitoring in this area by constructing a comprehensive evaluation framework that synergizes the traditional Water Quality Index (WQI) with an entropy-weighted method (EWQI). To improve the prediction model’s efficiency and accuracy, environmental factors were screened using Pearson correlation analysis, and the related environmental factors were selected as the input variables of the prediction model. The performance of the three prediction models were evaluated using root mean square error (RMSE). The results indicated that after reducing the dimensionality of the environmental factors and applying support vector machines (SVM), the maximum RMSE was only 1.61 in the testing process. To address the single-output limitation of SVM, the multi-output support vector machine (MSVM) can be achieved by modifying its internal structure, which enables the concurrent prediction of EWQI and WQI. Furthermore, the grey wolf optimizer (GWO) was integrated to enhance the prediction accuracy of the model. Experimental results demonstrated that the proposed GWO-MSVM model achieved superior prediction performance in the study area, with a maximum RMSE of 1.40, a minimum coefficient of determination (R2) of 0.92, and a maximum mean absolute error (MAE) of 1.04 in the testing process. The research provides a reliable technical framework and strategic support for water quality assessment and ecological conservation in the Yangtze River basin.
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-9
An Optical Sensing Method for Lung Deposition Surface Area Concentration Measurement
The concentration of lung deposition surface area (LDSA) is a crucial metric for assessing potential health impacts associated with particle size distribution. Traditional LDSA sensors, which rely on particle charging, are limited to measuring particles up to 300 nm in size. However, aerosols in the submicron to micron range are also significant for evaluating exposure risks. In this study, we introduce an optical method for LDSA sensing that leverages light-scattering angular spectrum and deep learning techniques. Our approach utilizes a compact, lens-free sensor to sample angular spectrum signals, which are then processed by a ResNet-like neural network to adjust the conversion factor between LDSA concentration and scattering light intensity. Experimental results demonstrate that our prototype achieves a maximum relative error of 14.78% in measuring the angular spectrum. Additionally, the neural network compensates for variations in particle size distribution and refractive index, refining the conversion factor. When compared to reference LDSA concentrations provided by laboratory instruments, our prototype maintains relative errors within 17.66%. With its simple mechanical design, high accuracy, and cost-effectiveness, this method presents a promising solution for online and field measurements of LDSA concentration.
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-9
An Optical Sensing Method for Lung Deposition Surface Area Concentration Measurement
The concentration of lung deposition surface area (LDSA) is a crucial metric for assessing potential health impacts associated with particle size distribution. Traditional LDSA sensors, which rely on particle charging, are limited to measuring particles up to 300 nm in size. However, aerosols in the submicron to micron range are also significant for evaluating exposure risks. In this study, we introduce an optical method for LDSA sensing that leverages light-scattering angular spectrum and deep learning techniques. Our approach utilizes a compact, lens-free sensor to sample angular spectrum signals, which are then processed by a ResNet-like neural network to adjust the conversion factor between LDSA concentration and scattering light intensity. Experimental results demonstrate that our prototype achieves a maximum relative error of 14.78% in measuring the angular spectrum. Additionally, the neural network compensates for variations in particle size distribution and refractive index, refining the conversion factor. When compared to reference LDSA concentrations provided by laboratory instruments, our prototype maintains relative errors within 17.66%. With its simple mechanical design, high accuracy, and cost-effectiveness, this method presents a promising solution for online and field measurements of LDSA concentration.
- 24 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-8
An Optical Sensor for Measuring the Particle Size Distribution of Aerosol Without Refractive Index Using the Optimized Multioutput Support Vector Machine
Due to the advantages of non-contact and high sensitivity, the optical light-based method has been widely used to measure the particle size distribution (PSD) of aerosols, where the prior information of refractive index is necessary to retrieve the PSD of aerosols. However, the refractive index of aerosols cannot be obtained in situ and on-line. To address this challenge, an optical light-based method is proposed to retrieve the PSD of aerosols using the scattering phase function and optimized machine learning without requiring any prior information of refractive index. The scattering phase function is used to describe the scattering light intensity of aerosols that corresponds to different observation angle. Meanwhile, an optical prototype sensor is designed and evaluated through different kinds of aerosol samples. The retrieved PSD was in good agreement with the reference PSD, and the maximum DKL is 0.08. The experiment results proved that the proposed method in this study can retrieve the PSD of aerosols without requiring the information of refractive index, and show significant potential for measuring the PSD of aerosols in situ and on-line.
- 24 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-8
An Optical Sensor for Measuring the Particle Size Distribution of Aerosol Without Refractive Index Using the Optimized Multioutput Support Vector Machine
Due to the advantages of non-contact and high sensitivity, the optical light-based method has been widely used to measure the particle size distribution (PSD) of aerosols, where the prior information of refractive index is necessary to retrieve the PSD of aerosols. However, the refractive index of aerosols cannot be obtained in situ and on-line. To address this challenge, an optical light-based method is proposed to retrieve the PSD of aerosols using the scattering phase function and optimized machine learning without requiring any prior information of refractive index. The scattering phase function is used to describe the scattering light intensity of aerosols that corresponds to different observation angle. Meanwhile, an optical prototype sensor is designed and evaluated through different kinds of aerosol samples. The retrieved PSD was in good agreement with the reference PSD, and the maximum DKL is 0.08. The experiment results proved that the proposed method in this study can retrieve the PSD of aerosols without requiring the information of refractive index, and show significant potential for measuring the PSD of aerosols in situ and on-line.
- 11 March 2024
- Published by American Chemical Society (ACS) in ACS Sensors
- Vol. 9 (3) , 1208-1217
Light Scattering Method for Aerosol Sizing Based on Machine Learning
Optical scattering has been widely used for aerosol sizing due to its noninvasive and real-time measurement. However, it is crucial to retrieve the particle size distribution (PSD) of aerosols without prior knowledge of the refractive index. Now, it has been a great challenge to measure the refractive index in situ. In this study, a novel PSD sensing method utilizing the light scattering angular spectrum (LSAS) and machine learning techniques is proposed to address this challenge. The complex nonlinear relationship between LSAS and PSD can be constructed while accounting for the refractive index of aerosols. A miniaturized prototype sensor is designed and tested on different sizes of aerosol samples. The experiment results showed that the maximum Kullback-Leibler divergence (DKL) of PSD is 0.07, which indicates that the sensing method can provide the ability for highly accurate aerosol PSD measurement without requiring prior knowledge of the refractive index. The compacted prototype sensor shows great potential for aerosol analysis in conventional field measurements outside the laboratory.
- 11 March 2024
- Published by American Chemical Society (ACS) in ACS Sensors
- Vol. 9 (3) , 1208-1217
Light Scattering Method for Aerosol Sizing Based on Machine Learning
Optical scattering has been widely used for aerosol sizing due to its noninvasive and real-time measurement. However, it is crucial to retrieve the particle size distribution (PSD) of aerosols without prior knowledge of the refractive index. Now, it has been a great challenge to measure the refractive index in situ. In this study, a novel PSD sensing method utilizing the light scattering angular spectrum (LSAS) and machine learning techniques is proposed to address this challenge. The complex nonlinear relationship between LSAS and PSD can be constructed while accounting for the refractive index of aerosols. A miniaturized prototype sensor is designed and tested on different sizes of aerosol samples. The experiment results showed that the maximum Kullback-Leibler divergence (DKL) of PSD is 0.07, which indicates that the sensing method can provide the ability for highly accurate aerosol PSD measurement without requiring prior knowledge of the refractive index. The compacted prototype sensor shows great potential for aerosol analysis in conventional field measurements outside the laboratory.
- Open Access
- 21 June 2022
- Published by MDPI AG in Biosensors
- Vol. 12 (7) , 436
A Simple Optical Aerosol Sensing Method of Sauter Mean Diameter for Particulate Matter Monitoring
Mass concentration is a commonly used but insufficient metric to evaluate the particulate matter (PM) exposure hazard. Recent studies have declared that small particles have more serious impacts on human health than big particles given the same mass concentration. However, state-of-the-art PM sensors cannot provide explicit information of the particle size for further analysis. In this work, we adopt Sauter mean diameter (SMD) as a key metric to reflect the particle size besides the mass concentration. To measure SMD, an effective optical sensing method and a proof-of-concept prototype sensor are proposed by using dual wavelengths technology. In the proposed method, a non-linear conversion model is developed to improve the SMD measurement accuracy for aerosol samples of different particle size distributions and reflective indices based on multiple scattering channels. In the experiment of Di-Ethyl-Hexyl-Sebacate (DEHS) aerosols, the outputs of our prototype sensor demonstrated a good agreement with existing laboratory reference instruments with maximum SMD measurement error down to 7.04%. Furthermore, the simplicity, feasibility and low-cost features of this new method present great potential for distributed PM monitoring, to support sophisticated human exposure hazard assessment.
- 21 June 2022
- Published by MDPI AG in Biosensors
- Vol. 12 (7) , 436
Open AccessA Simple Optical Aerosol Sensing Method of Sauter Mean Diameter for Particulate Matter Monitoring
Mass concentration is a commonly used but insufficient metric to evaluate the particulate matter (PM) exposure hazard. Recent studies have declared that small particles have more serious impacts on human health than big particles given the same mass concentration. However, state-of-the-art PM sensors cannot provide explicit information of the particle size for further analysis. In this work, we adopt Sauter mean diameter (SMD) as a key metric to reflect the particle size besides the mass concentration. To measure SMD, an effective optical sensing method and a proof-of-concept prototype sensor are proposed by using dual wavelengths technology. In the proposed method, a non-linear conversion model is developed to improve the SMD measurement accuracy for aerosol samples of different particle size distributions and reflective indices based on multiple scattering channels. In the experiment of Di-Ethyl-Hexyl-Sebacate (DEHS) aerosols, the outputs of our prototype sensor demonstrated a good agreement with existing laboratory reference instruments with maximum SMD measurement error down to 7.04%. Furthermore, the simplicity, feasibility and low-cost features of this new method present great potential for distributed PM monitoring, to support sophisticated human exposure hazard assessment.
- 7 April 2022
- Published by Springer Nature in Water Resources Management
- Vol. 36 (6) , 2045-2060
Environmental Factors Assisted the Evaluation of Entropy Water Quality Indices with Efficient Machine Learning Technique
- 7 April 2022
- Published by Springer Nature in Water Resources Management
- Vol. 36 (6) , 2045-2060
Environmental Factors Assisted the Evaluation of Entropy Water Quality Indices with Efficient Machine Learning Technique