Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery

In rice detection tasks, accurate identification of leaf streaks, pest and disease distribution, and spikelet hierarchies relies on high-quality images to distinguish between texture and hierarchy. However, existing images often suffer from texture blurring and contour shifting due to equipment and...

Full description

Saved in:
Bibliographic Details
Main Authors: Zexiao Zhang, Jie Zhang, Jinyang Du, Xiangdong Chen, Wenjing Zhang, Changmeng Peng
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/15/7/1729
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850067514502938624
author Zexiao Zhang
Jie Zhang
Jinyang Du
Xiangdong Chen
Wenjing Zhang
Changmeng Peng
author_facet Zexiao Zhang
Jie Zhang
Jinyang Du
Xiangdong Chen
Wenjing Zhang
Changmeng Peng
author_sort Zexiao Zhang
collection DOAJ
description In rice detection tasks, accurate identification of leaf streaks, pest and disease distribution, and spikelet hierarchies relies on high-quality images to distinguish between texture and hierarchy. However, existing images often suffer from texture blurring and contour shifting due to equipment and environment limitations, which affects the detection performance. In view of the fact that pests and diseases affect the whole situation and tiny details are mostly localized, we propose a rice image reconstruction method based on an adaptive two-branch heterogeneous structure. The method consists of a low-frequency branch (LFB) that recovers global features using orientation-aware extended receptive fields to capture streaky global features, such as pests and diseases, and a high-frequency branch (HFB) that enhances detail edges through an adaptive enhancement mechanism to boost the clarity of local detail regions. By introducing the dynamic weight fusion mechanism (CSDW) and lightweight gating network (LFFN), the problem of the unbalanced fusion of frequency information for rice images in traditional methods is solved. Experiments on the 4× downsampled rice test set demonstrate that the proposed method achieves a 62% reduction in parameters compared to EDSR, 41% lower computational cost (30 G) than MambaIR-light, and an average PSNR improvement of 0.68% over other methods in the study while balancing memory usage (227 M) and inference speed. In downstream task validation, rice panicle maturity detection achieves a 61.5% increase in mAP50 (0.480 → 0.775) compared to interpolation methods, and leaf pest detection shows a 2.7% improvement in average mAP50 (0.949 → 0.975). This research provides an effective solution for lightweight rice image enhancement, with its dual-branch collaborative mechanism and dynamic fusion strategy establishing a new paradigm in agricultural rice image processing.
format Article
id doaj-art-3341206d02ad44cbae0e2b392b1708ea
institution DOAJ
issn 2073-4395
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-3341206d02ad44cbae0e2b392b1708ea2025-08-20T02:48:17ZengMDPI AGAgronomy2073-43952025-07-01157172910.3390/agronomy15071729Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice ImageryZexiao Zhang0Jie Zhang1Jinyang Du2Xiangdong Chen3Wenjing Zhang4Changmeng Peng5School of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaSchool of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaSchool of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaSchool of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaSchool of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaSchool of Information Engineering, Sichuan Agricultural University, No. 46, Xinkang Road, Ya’an 625014, ChinaIn rice detection tasks, accurate identification of leaf streaks, pest and disease distribution, and spikelet hierarchies relies on high-quality images to distinguish between texture and hierarchy. However, existing images often suffer from texture blurring and contour shifting due to equipment and environment limitations, which affects the detection performance. In view of the fact that pests and diseases affect the whole situation and tiny details are mostly localized, we propose a rice image reconstruction method based on an adaptive two-branch heterogeneous structure. The method consists of a low-frequency branch (LFB) that recovers global features using orientation-aware extended receptive fields to capture streaky global features, such as pests and diseases, and a high-frequency branch (HFB) that enhances detail edges through an adaptive enhancement mechanism to boost the clarity of local detail regions. By introducing the dynamic weight fusion mechanism (CSDW) and lightweight gating network (LFFN), the problem of the unbalanced fusion of frequency information for rice images in traditional methods is solved. Experiments on the 4× downsampled rice test set demonstrate that the proposed method achieves a 62% reduction in parameters compared to EDSR, 41% lower computational cost (30 G) than MambaIR-light, and an average PSNR improvement of 0.68% over other methods in the study while balancing memory usage (227 M) and inference speed. In downstream task validation, rice panicle maturity detection achieves a 61.5% increase in mAP50 (0.480 → 0.775) compared to interpolation methods, and leaf pest detection shows a 2.7% improvement in average mAP50 (0.949 → 0.975). This research provides an effective solution for lightweight rice image enhancement, with its dual-branch collaborative mechanism and dynamic fusion strategy establishing a new paradigm in agricultural rice image processing.https://www.mdpi.com/2073-4395/15/7/1729agricultural image enhancementneural networksdual-branch structuredynamic fusioncrop detectiondisease identification
spellingShingle Zexiao Zhang
Jie Zhang
Jinyang Du
Xiangdong Chen
Wenjing Zhang
Changmeng Peng
Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
Agronomy
agricultural image enhancement
neural networks
dual-branch structure
dynamic fusion
crop detection
disease identification
title Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
title_full Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
title_fullStr Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
title_full_unstemmed Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
title_short Frequency-Domain Collaborative Lightweight Super-Resolution for Fine Texture Enhancement in Rice Imagery
title_sort frequency domain collaborative lightweight super resolution for fine texture enhancement in rice imagery
topic agricultural image enhancement
neural networks
dual-branch structure
dynamic fusion
crop detection
disease identification
url https://www.mdpi.com/2073-4395/15/7/1729
work_keys_str_mv AT zexiaozhang frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery
AT jiezhang frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery
AT jinyangdu frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery
AT xiangdongchen frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery
AT wenjingzhang frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery
AT changmengpeng frequencydomaincollaborativelightweightsuperresolutionforfinetextureenhancementinriceimagery