Bin Feng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bin Feng has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Lin Li
Northwestern Polytechnical University
3 shared publications - Qing Ye
National University of Defense Technology
2 shared publications - Ru Lei
Key Laboratory of Information Fusion Technology of the Ministry of Education, Northwestern Polytechnical University, Xian, Shaanxi, China
2 shared publications Fan Wang
1 shared publicationHaoyang Zhang
1 shared publicationJinpei Xiao
Northwestern Polytech Univ, Sch Automat, Xian 710072, Peoples R China
1 shared publication
Latest Publications
- 18 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 36407-36422
A Dual-Channel Deep Learning Method for Demosaicing Based on an Optimized Color-Polarization Array
Polarization and color are two essential attributes of color-polarization imaging In recent years, the application of both polarization and color information has gained widespread attention across various image processing research domains. This paper proposes a novel color-polarization array design method by employing the non-dominated sorting genetic algorithm (NSGA-II) to generate mosaicked images with color and polarization information, followed by a dual-channel guided deep-learning reconstruction network to recover these attributes. This optimization method produces a compact 3×3 array with an innovative filter layout. This minimizes polarization angles and unit array size while maintaining the integrity of color-polarization information in the reconstructed images. In addition, the use of deep learning modules enables a more thorough restoration of missing information in mosaicked images from a new perspective. Experimental results demonstrate that our network can accurately demosaic the mosaicked output of the proposed array, effectively recovering both the color and the polarization data. Our compact 3×3 array outperforms the classical 4×4 polar-bayer array on benchmark color-polarization image datasets.Additionally, the reconstruction algorithm preserves the fidelity of color and polarization after demosaicing.
- 18 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 36407-36422
A Dual-Channel Deep Learning Method for Demosaicing Based on an Optimized Color-Polarization Array
Polarization and color are two essential attributes of color-polarization imaging In recent years, the application of both polarization and color information has gained widespread attention across various image processing research domains. This paper proposes a novel color-polarization array design method by employing the non-dominated sorting genetic algorithm (NSGA-II) to generate mosaicked images with color and polarization information, followed by a dual-channel guided deep-learning reconstruction network to recover these attributes. This optimization method produces a compact 3×3 array with an innovative filter layout. This minimizes polarization angles and unit array size while maintaining the integrity of color-polarization information in the reconstructed images. In addition, the use of deep learning modules enables a more thorough restoration of missing information in mosaicked images from a new perspective. Experimental results demonstrate that our network can accurately demosaic the mosaicked output of the proposed array, effectively recovering both the color and the polarization data. Our compact 3×3 array outperforms the classical 4×4 polar-bayer array on benchmark color-polarization image datasets.Additionally, the reconstruction algorithm preserves the fidelity of color and polarization after demosaicing.
- 4 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 30 (3) , 911-924
An Adaptive Dual-Domain Prediction Strategy Based on Second-Order Derivatives for Dynamic Multiobjective Optimization
This paper tackles dynamic multi-objective optimization problems (DMOPs) by proposing novel change prediction strategies within an evolutionary algorithm framework. This framework combines an adaptive dual-domain change-response strategy with a second-order derivative prediction mechanism. In real-world scenarios, some Pareto Sets (PS) and Pareto Fronts (PF) evolve dynamically with environmental changes, while others remain stationary. Our algorithm employs an adaptive dual-domain approach that simultaneously monitors changes in both the PS and PF, and dynamically adjusts the allocation of prediction efforts between the decision and objective spaces according to environmental characteristics, thereby ensuring efficient sampling when objectives change. Furthermore, we incorporate a second-order derivative prediction scheme to actively reinitialize the population, enhancing the algorithm’s responsiveness to sudden or nonlinear changes. We evaluate the proposed method on 28 standard benchmark DMOPs and compare it with six state-of-the-art algorithms. The experimental results indicate that the proposed method achieves significant advantages in convergence and diversity on most test problems.
- 4 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 30 (3) , 911-924
An Adaptive Dual-Domain Prediction Strategy Based on Second-Order Derivatives for Dynamic Multiobjective Optimization
This paper tackles dynamic multi-objective optimization problems (DMOPs) by proposing novel change prediction strategies within an evolutionary algorithm framework. This framework combines an adaptive dual-domain change-response strategy with a second-order derivative prediction mechanism. In real-world scenarios, some Pareto Sets (PS) and Pareto Fronts (PF) evolve dynamically with environmental changes, while others remain stationary. Our algorithm employs an adaptive dual-domain approach that simultaneously monitors changes in both the PS and PF, and dynamically adjusts the allocation of prediction efforts between the decision and objective spaces according to environmental characteristics, thereby ensuring efficient sampling when objectives change. Furthermore, we incorporate a second-order derivative prediction scheme to actively reinitialize the population, enhancing the algorithm’s responsiveness to sudden or nonlinear changes. We evaluate the proposed method on 28 standard benchmark DMOPs and compare it with six state-of-the-art algorithms. The experimental results indicate that the proposed method achieves significant advantages in convergence and diversity on most test problems.
- Open Access
- 1 July 2024
- Published by KEAI Publishing in Defence Technology
- Vol. 37, 50-61
Color-polarization synergistic target detection method considering shadow interference
- 1 July 2024
- Published by KEAI Publishing in Defence Technology
- Vol. 37, 50-61
Open AccessColor-polarization synergistic target detection method considering shadow interference
- 18 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Miniaturization for Air and Space Systems
- Vol. 5 (2) , 79-84
Infrared Image Super-Resolution Quality Prediction Driven by Frequency-Domain Features
In order to predict infrared image super-resolution (SR) quality under conditions of no SR image generation and no high-resolution reference image, this article proposes a fully connected neural network model for infrared image SR quality prediction driven by wavelet domain energy features. Utilizing the multiresolution and scale-invariant properties of undecimated wavelets, our model separates low-frequency information component and three high-frequency information components from a low-resolution image. Our model achieves decorrelation of image pixel information and generates wavelet domain energy normalization features. Utilizing a fully connected neural network, we construct a wavelet-based image SR quality prediction network. This neural network combines four subnetworks to enhance the network representation and learning capabilities. The network is trained using wavelet domain energy normalization features, while autonomously learning the mapping relationship between input data and evaluation metric. This model is validated on infrared image data sets. The experimental results indicate that the proposed model can accurately predict SR image quality metrics by utilizing frequency-domain energy features of a low-resolution image.
- 18 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Miniaturization for Air and Space Systems
- Vol. 5 (2) , 79-84
Infrared Image Super-Resolution Quality Prediction Driven by Frequency-Domain Features
In order to predict infrared image super-resolution (SR) quality under conditions of no SR image generation and no high-resolution reference image, this article proposes a fully connected neural network model for infrared image SR quality prediction driven by wavelet domain energy features. Utilizing the multiresolution and scale-invariant properties of undecimated wavelets, our model separates low-frequency information component and three high-frequency information components from a low-resolution image. Our model achieves decorrelation of image pixel information and generates wavelet domain energy normalization features. Utilizing a fully connected neural network, we construct a wavelet-based image SR quality prediction network. This neural network combines four subnetworks to enhance the network representation and learning capabilities. The network is trained using wavelet domain energy normalization features, while autonomously learning the mapping relationship between input data and evaluation metric. This model is validated on infrared image data sets. The experimental results indicate that the proposed model can accurately predict SR image quality metrics by utilizing frequency-domain energy features of a low-resolution image.
- 9 September 2023
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 171
Accurate phase aberration compensation with convolutional neural network PACUnet3+ in digital holographic microscopy
- 9 September 2023
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 171
Accurate phase aberration compensation with convolutional neural network PACUnet3+ in digital holographic microscopy