Prof. Mingliang Xu Profile

Prof. Mingliang Xu

SPIE Involvement:

Author

Publications (3)

This will count as one of your downloads.

You will have access to both the presentation and article (if available).

DOWNLOAD NOW

This content is available for download via your institution's subscription. To access this item, please sign in to your personal account.

Email or Username Forgot your username?

Password Forgot your password?

Show

Keep me signed in

No SPIE account? Create an account

Proceedings Article | 13 December 2024 Paper

MFBF-Net: dehazing network based on multifrequency branch fusion

Fengchang Miao, Yan Zhang, Fuchen Li, Haopeng Zhao, Ziqin Xu, Pengyun Chen, Shaohui Jin, Mingliang Xu, Hao Liu

Proceedings Volume 13501, 1350103 (2024) https://doi.org/10.1117/12.3045874

KEYWORDS: Image restoration, Feature fusion, Transformers, Image processing, Convolutional neural networks, Atmospheric modeling, Image transmission

Read Abstract +

Image dehazing technology is a hot topic in the fields of image processing and computer vision, aiming to obtain details and texture features of the original scene from foggy images, and then obtain clear and fog free images. Most of the existing research methods are suitable for tasks in low fog scenarios. As the fog concentration increases, the image reconstruction quality of the algorithm significantly decreases, accompanied by detail loss and distortion. In addition, most existing algorithms require a large amount of foggy datasets during model training, and model training takes a long time, which reduces the practicality of the model. In response to the above issues, this paper proposes an image dehazing model based on a small sample multi attention mechanism and multi frequency branch fusion (MFBF-Net). This model can effectively extract high-frequency and low-frequency detail information in the image, and reconstruct the real image to the greatest extent possible. The experimental results show that the dehazing model proposed in this paper exhibits good dehazing performance on small sample datasets, and has good performance in different concentrations of foggy scenes.

Proceedings Article | 30 April 2024 Paper

Active non-line-of-sight imaging through dense fog via hybrid intelligent enhancement perception system

Ziqin Xu, Yanxin Zhang, Haopeng Zhao, Fengchang Miao, Hao Liu, Shaohui Jin, Mingliang Xu

Proceedings Volume 13155, 131550G (2024) https://doi.org/10.1117/12.3015924

Read Abstract +

Non-line-of-sight(NLOS) imaging through fog has been extensively researched in the fields of optics and computer vision. However, due to the influence of strong backscattering and diffuse reflection generated by the dense fog on the temporal-spatial correlations of photons returning from the target object, the reconstruction quality of most existing methods is significantly reduced under dense fog conditions. In this study, we define the optical imaging process in a foggy environment and propose a hybrid intelligent enhancement perception(HIEP) system based on Time-of-Flight(ToF) methods and physics-driven Swin transformer(ToFormer) to eliminate scattering effects and reconstruct targets under heterogeneous fog with varying optical thickness. Furthermore, we assembled a prototype of the HIEP system and established the Active Non-Line-of-Sight Imaging Through Dense Fog(NLOSTDF) dataset to train the reconstruction network. The experimental results demonstrate that even in dense fog short-distance scenarios with an optical thickness of up to 2.5 and imaging distances less than 6 meters, our approach achieves clear imaging of the target scene, surpassing existing optical and computer vision methods.

Proceedings Article | 25 March 2024 Paper

A deep encoder-decoder based primal-dual proximal network for image restoration

Siqi Wang, Mingyuan Jiu, Li Chen, Shupan Li, Mingliang Xu

Proceedings Volume 13089, 130890X (2024) https://doi.org/10.1117/12.3021256

Read Abstract +

Image restoration is a popular and challenge task, which is regarded as a classical inverse problem. Condat-V ũ primal-dual algorithm based on proximal operator is one of successful optimization methods. It is further reformulated as a primal-dual proximal network, where one iteration in the original algorithm corresponds to one layer in the network. The drawback of primal-dual network is that blur kernels should be given as prior information, however, it is usually very hard to be known in the real situation. In this work, we propose a deep encoder-decoder primal-dual proximal network, named ED-PDPNet. In each layer, the blur kernels and the projections between the primal and dual variables are designed as encoder-decoder modules, in this way, the network can be learned in an end-to-end way and all the parameters in the primal-dual algorithm are learned. The proposed method is applied on the MNIST and BSD68 datasets for image restoration. The preliminary results show that the proposed method by combining simple encoder-decoder modules obtained very promising and competitive performance compared to the state-of-the-art methods. In addition, the proposed network is shown to be a lightweight network with fewer learning parameters in comparison to the recent popular transformer-based method.

My Library

You currently do not have any folders to save your paper to! Create a new folder below.

Folder Name

Folder Description

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks. You are receiving this notice because your organization may not have SPIE eBooks access.*

*Shibboleth/Open Athens users─please sign in to access your institution's subscriptions.

To obtain this item, you may purchase the complete book in print or electronic format on SPIE.org.

ORGANIZATIONAL
Sign in with credentials provided by your organization.

Organizational Username

Organizational Password

Show/Hide Password

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members:

Non-members: ADD TO CART

Keywords/Phrases

Search In:

Publication Years