Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network

The complex underground environment of coal coalmines, influenced by lighting, coal dust, and water mist, often results in collected images with blurred details and missing textures, leading to decreased image resolution and posing significant limitations to the intelligent development of coal coalm...

Full description

Saved in:
Bibliographic Details
Main Authors: Jian CHENG, Lifei MI, Hao LI, Heping LI, Guangfu WANG, Yongzhuang MA
Format: Article
Language:zho
Published: Editorial Department of Coal Science and Technology 2024-11-01
Series:Meitan kexue jishu
Subjects:
Online Access:http://www.mtkxjs.com.cn/article/doi/10.12438/cst.2024-1055
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850124161086652416
author Jian CHENG
Lifei MI
Hao LI
Heping LI
Guangfu WANG
Yongzhuang MA
author_facet Jian CHENG
Lifei MI
Hao LI
Heping LI
Guangfu WANG
Yongzhuang MA
author_sort Jian CHENG
collection DOAJ
description The complex underground environment of coal coalmines, influenced by lighting, coal dust, and water mist, often results in collected images with blurred details and missing textures, leading to decreased image resolution and posing significant limitations to the intelligent development of coal coalmine safety monitoring. Image super-resolution reconstruction, an essential image processing technology, aims to recover clear high-resolution images from low-resolution coalmine images, thereby significantly enhancing the reliability of intelligent monitoring and safety management in coal coalmines. To address issues such as the loss of edge texture information and blurring of details in coalmine images, a coalmine image super-resolution reconstruction method integrating multi-dimensional features and residual attention networks is proposed. First, a multi-branch network is employed to parallelly integrate dynamic convolution and channel attention mechanisms, capturing different spatial statistical characteristics through “horizontal-channel” and “vertical-channel” interactions. Secondly, a recursive sparse self-attention mechanism is designed to aggregate representative feature maps under linear complexity, adaptively selecting weight distribution and reducing information redundancy during computation. Finally, the basic unit of deep feature extraction is constructed based on the standard multi-head self-attention mechanism and residual connection, with the obtained feature information and shallow features jointly input into the reconstruction module via skip connections to complete super-resolution reconstruction of coalmine images. Experimental results indicate that the proposed method significantly outperforms existing mainstream algorithms in both objective evaluation metrics and subjective visual analysis. In tests on the coalmine dataset, LPIPS (Learned Perceptual Image Patch Similarity) decreases by an average of 10.97% and 9.91%, while PSNR (Peak Signal-to-Noise Ratio) increases by an average of 4.10% and 2.30% for 2x and 4x scaling factors, respectively, demonstrating the method's effectiveness in restoring the structure and texture details of coalmine images.
format Article
id doaj-art-0a876c30d4324d3cba4516dff962c8bc
institution OA Journals
issn 0253-2336
language zho
publishDate 2024-11-01
publisher Editorial Department of Coal Science and Technology
record_format Article
series Meitan kexue jishu
spelling doaj-art-0a876c30d4324d3cba4516dff962c8bc2025-08-20T02:34:23ZzhoEditorial Department of Coal Science and TechnologyMeitan kexue jishu0253-23362024-11-01521111712810.12438/cst.2024-10552024-1055Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention networkJian CHENG0Lifei MI1Hao LI2Heping LI3Guangfu WANG4Yongzhuang MA5China Coal Research Institute, Beijing 100013, ChinaChina Coal Research Institute, Beijing 100013, ChinaResearch Institute of Mine Artificial Intelligence, Chinese Institute of Coal Science, Beijing 100013, ChinaChina Coal Research Institute, Beijing 100013, ChinaResearch Institute of Mine Artificial Intelligence, Chinese Institute of Coal Science, Beijing 100013, ChinaResearch Institute of Mine Artificial Intelligence, Chinese Institute of Coal Science, Beijing 100013, ChinaThe complex underground environment of coal coalmines, influenced by lighting, coal dust, and water mist, often results in collected images with blurred details and missing textures, leading to decreased image resolution and posing significant limitations to the intelligent development of coal coalmine safety monitoring. Image super-resolution reconstruction, an essential image processing technology, aims to recover clear high-resolution images from low-resolution coalmine images, thereby significantly enhancing the reliability of intelligent monitoring and safety management in coal coalmines. To address issues such as the loss of edge texture information and blurring of details in coalmine images, a coalmine image super-resolution reconstruction method integrating multi-dimensional features and residual attention networks is proposed. First, a multi-branch network is employed to parallelly integrate dynamic convolution and channel attention mechanisms, capturing different spatial statistical characteristics through “horizontal-channel” and “vertical-channel” interactions. Secondly, a recursive sparse self-attention mechanism is designed to aggregate representative feature maps under linear complexity, adaptively selecting weight distribution and reducing information redundancy during computation. Finally, the basic unit of deep feature extraction is constructed based on the standard multi-head self-attention mechanism and residual connection, with the obtained feature information and shallow features jointly input into the reconstruction module via skip connections to complete super-resolution reconstruction of coalmine images. Experimental results indicate that the proposed method significantly outperforms existing mainstream algorithms in both objective evaluation metrics and subjective visual analysis. In tests on the coalmine dataset, LPIPS (Learned Perceptual Image Patch Similarity) decreases by an average of 10.97% and 9.91%, while PSNR (Peak Signal-to-Noise Ratio) increases by an average of 4.10% and 2.30% for 2x and 4x scaling factors, respectively, demonstrating the method's effectiveness in restoring the structure and texture details of coalmine images.http://www.mtkxjs.com.cn/article/doi/10.12438/cst.2024-1055coalmine underground imagessuper-resolution reconstructionattention mechanismresidual networkrecursive algorithm
spellingShingle Jian CHENG
Lifei MI
Hao LI
Heping LI
Guangfu WANG
Yongzhuang MA
Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
Meitan kexue jishu
coalmine underground images
super-resolution reconstruction
attention mechanism
residual network
recursive algorithm
title Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
title_full Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
title_fullStr Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
title_full_unstemmed Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
title_short Coalmine image super-resolution reconstruction via fusing multi-dimensional feature and residual attention network
title_sort coalmine image super resolution reconstruction via fusing multi dimensional feature and residual attention network
topic coalmine underground images
super-resolution reconstruction
attention mechanism
residual network
recursive algorithm
url http://www.mtkxjs.com.cn/article/doi/10.12438/cst.2024-1055
work_keys_str_mv AT jiancheng coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork
AT lifeimi coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork
AT haoli coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork
AT hepingli coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork
AT guangfuwang coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork
AT yongzhuangma coalmineimagesuperresolutionreconstructionviafusingmultidimensionalfeatureandresidualattentionnetwork