Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MvKSR: Multi-view Knowledge-guided Scene Recovery for Hazy and Rainy Degradation

Dong Yang, Wenyu Xu, Yuan Gao, Yuxu Lu, Jingming Zhang, Yu Guo

TL;DR

A novel multiview knowledge-guided scene recovery network (termed MvKSR) is proposed that outperforms other state-of-the-art methods in terms of efficiency and stability for restoring degraded scenarios, and can better serve the needs of advanced vision tasks in VMS.

Abstract

High-quality imaging is crucial for ensuring safety supervision and intelligent deployment in fields like transportation and industry. It enables precise and detailed monitoring of operations, facilitating timely detection of potential hazards and efficient management. However, adverse weather conditions, such as atmospheric haziness and precipitation, can have a significant impact on image quality. When the atmosphere contains dense haze or water droplets, the incident light scatters, leading to degraded captured images. This degradation is evident in the form of image blur and reduced contrast, increasing the likelihood of incorrect assessments and interpretations by intelligent imaging systems (IIS). To address the challenge of restoring degraded images in hazy and rainy conditions, this paper proposes a novel multi-view knowledge-guided scene recovery network (termed MvKSR). Specifically, guided filtering is performed on the degraded image to separate high/low-frequency components. Subsequently, an en-decoder-based multi-view feature coarse extraction module (MCE) is used to coarsely extract features from different views of the degraded image. The multi-view feature fine fusion module (MFF) will learn and infer the restoration of degraded images through mixed supervision under different views. Additionally, we suggest an atrous residual block to handle global restoration and local repair in hazy/rainy/mixed scenes. Extensive experimental results demonstrate that MvKSR outperforms other state-of-the-art methods in terms of efficiency and stability for restoring degraded scenarios in IIS.

MvKSR: Multi-view Knowledge-guided Scene Recovery for Hazy and Rainy Degradation

TL;DR

A novel multiview knowledge-guided scene recovery network (termed MvKSR) is proposed that outperforms other state-of-the-art methods in terms of efficiency and stability for restoring degraded scenarios, and can better serve the needs of advanced vision tasks in VMS.

Abstract

High-quality imaging is crucial for ensuring safety supervision and intelligent deployment in fields like transportation and industry. It enables precise and detailed monitoring of operations, facilitating timely detection of potential hazards and efficient management. However, adverse weather conditions, such as atmospheric haziness and precipitation, can have a significant impact on image quality. When the atmosphere contains dense haze or water droplets, the incident light scatters, leading to degraded captured images. This degradation is evident in the form of image blur and reduced contrast, increasing the likelihood of incorrect assessments and interpretations by intelligent imaging systems (IIS). To address the challenge of restoring degraded images in hazy and rainy conditions, this paper proposes a novel multi-view knowledge-guided scene recovery network (termed MvKSR). Specifically, guided filtering is performed on the degraded image to separate high/low-frequency components. Subsequently, an en-decoder-based multi-view feature coarse extraction module (MCE) is used to coarsely extract features from different views of the degraded image. The multi-view feature fine fusion module (MFF) will learn and infer the restoration of degraded images through mixed supervision under different views. Additionally, we suggest an atrous residual block to handle global restoration and local repair in hazy/rainy/mixed scenes. Extensive experimental results demonstrate that MvKSR outperforms other state-of-the-art methods in terms of efficiency and stability for restoring degraded scenarios in IIS.
Paper Structure (23 sections, 12 equations, 7 figures, 6 tables)

This paper contains 23 sections, 12 equations, 7 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Physical Imaging Models
  3. Proposed Method
  4. Guided Filter Decomposition
  5. Atrous Residual Block
  6. Multi-view Feature Coarse Extraction Module
  7. Multi-view Feature Fine Fusion Module
  8. Ultimate Loss Function
  9. Multi-Scale Structural Similarity Loss
  10. Contrastive Regularization Loss
  11. Experiments and Discussions
  12. Dataset and Implementation Details
  13. Competitive Methods and Evaluation Metrics
  14. Synthetic Quantitative and Visual Analysis
  15. Dehazing
  16. ...and 8 more sections

Figures (7)

  • Figure 1: The flowchart of the multi-view knowledge-guided scene recovery network (termed MvKSR). Multi-view feature coarse extraction module (MCE) will perform en-decoder based learning inference on the degraded image and the corresponding high/low-frequency components. Multi-view feature fine fusion module (MFF) (including front and back fusion) will guide the restoration of degraded images through mixed supervision.
  • Figure 2: Clear and synthetic different types of degraded images and corresponding high- and low-frequency components.
  • Figure 3: The pipeline of proposed standard and atrous mixed residual block (MRB) with atrous rate $r=3$.
  • Figure 4: Cross supervision in multi-view feature fine fusion module (MFF).
  • Figure 5: Example of the scene recovery of MFF and MvKSR under different views. The lower triangles in (b)-(d) are degraded patterns, and the corresponding restored patterns by our method are shown in the upper triangles.
  • ...and 2 more figures