SRNet-YOLO: A model for detecting tiny and very tiny pests in cotton fields based on super-resolution reconstruction.

Introduction: Effective pest management is important during the natural growth phases of cotton in the wild. As cotton fields are infested with "tiny pests" (smaller than 32×32 pixels) and "very tiny pests" (smaller than 16×16 pixels) during growth, making it difficult for common object detection models to accurately detect and fail to make sound agricultural decisions. Methods: In this study, we proposed a framework for detecting "tiny pests" and "very tiny pests" in wild cotton fields, named SRNet-YOLO. SRNet-YOLO includes a YOLOv8 feature extraction module, a feature map super-resolution reconstruction module (FM-SR), and a fusion mechanism based on BiFormer attention (BiFormerAF). Specially, the FM-SR module is designed for the feature map level to recover the important feature in detail, in other words, this module reconstructs the P5 layer feature map into the size of the P3 layer. And then we designed the BiFormerAF module to fuse this reconstruct layer with the P3 layer, which greatly improves the detection performance. The purpose of the BiFormerAF module is to solve the problem of possible loss of feature after reconstruction. Additionally, to validate the performance of our method for "tiny pests" and "very tiny pests" detection in cotton fields, we have developed a large dataset, named Cotton-Yellow-Sticky-2023, which collected pests by yellow sticky traps. Results: Through comprehensive experimental verification, we demonstrate that our proposed framework achieves exceptional performance. Our method achieved 78.2% mAP on the "tiny pests" test result, it surpasses the performance of leading detection models such as YOLOv3, YOLOv5, YOLOv7 and YOLOv8 by 6.9%, 7.2%, 5.7% and 4.1%, respectively. Meanwhile, our results on "very tiny pests" reached 57% mAP, which are 32.2% higher than YOLOv8. To verify the generalizability of the model, our experiments on Yellow Sticky Traps (low-resolution) dataset still maintained the highest 92.8% mAP. Discussion: The above experimental results indicate that our model not only provides help in solving the problem of tiny pests in cotton fields, but also has good generalizability and can be used for the detection of tiny pests in other crops.

Graphitic carbon @ silver nanoparticle @ porous silicon Bragg mirror composite SERS substrate for gallic acid detection.

Graphite carbon (G) @ silver (Ag) @ porous silicon Bragg mirror (PSB) composite SERS substrate was successfully synthesized using electrochemical etching (ec) and hydrothermal carbonization (HTC) techniques with silver nitrate as the source of silver and glucose as the source of carbon. The PSB was used as a functional scaffold for the synthesis of graphite-carbon and silver composite nanoparticles (G@AgNPs) on its surface, thereby combining SERS activity and antioxidant properties. To our knowledge, this is the first time that G@AgNPs has been synthesized on the PSB using glucose as a carbon source. The synthesized G@Ag@PSB was utilized as a SERS platform for the detection of gallic acid (GA). Test results demonstrated that the substrate exhibited a remarkable SERS enhancement capability for GA, with the enhancement factor (EF) reaching 2 × 105. The reproducibility of the SERS spectral signal was excellent, with a relative standard deviation (RSD) of 7.5 %. The sensitivity test results showed that the linear range of GA detection based on G@Ag@PSB composite SERS substrate was 2 × 10-3-2 × 10-12M. The relationship between GA concentration and SERS signal intensity exhibited a strong linear correlation, with a linear correlation coefficient (R2) of 0.97634. Moreover, even with an extended storage period, only a marginal decline in the signal intensity of GA on the substrate was observed. The results of this study demonstrate that the prepared G@Ag@PSB composite SERS substrate had good potential application performance as a low-cost SERS detection platform suitable for commercial use. In addition, this advance facilitates the further exploration of more nanomaterials with ultra-high sensitivity in SERS technology.

Research on the perception method of tiny objects in low-light and wide-field video.

At present, many trackers exhibit commendable performance in well-illuminated scenarios but overlook target tracking in low-light environments. As night falls, the tracker's accuracy drops dramatically. Challenges such as high image resolution, intricate backgrounds, uneven illumination, and the resemblance between targets and backgrounds in Hawk-Eye surveillance videos make tracking small objects in low-light and wide-field scenarios exceedingly difficult for previous trackers. To address these challenges, this paper introduces an innovative approach by integrating the difference constraint method into the CF (correlation filters) tracker, which generates a change-aware mask using inter-frame difference information. In addition, a dual regression model and inter-frame difference constraint term are introduced to restrict each other for dual filter learning. In this paper, we construct a new benchmark comprising 41 night surveillance sequences captured by Hawk-Eye cameras. Exhaustive experiments are conducted on this benchmark. The results show that the proposed method maintains superior accuracy, surpasses state-of-the-art trackers in this dataset, and achieves a real-time performance of 27 fps on a single CPU, substantially advancing tiny object tracking on Hawk-Eye surveillance videos in low light and in night scenes.

Microplastics and their interaction with microorganisms in Bosten Lake sediment.

Microplastics (MPs), discovered in oceans, lakes, and rivers, can infiltrate the food chain through ingestion by organisms, potentially posing health risks. Our research is the first to study the composition and distribution of MPs in Bosten Lake's sediment. In May, the average abundance of MPs was 0.95±0.72 particles per 10 gs, and in October, it was 0.90±0.61 particles per 10 gs. Bohu Town had the highest MP abundance, with 1.75±0.35 particles per 10 gs in spring and 2 ± 0 particles per 10 gs in autumn. In May, 53 % of the MPs were transparent, while in October, black MPs constituted 58 %. The predominant morphology was fibrous, accounting for 61 % of the total. MPs in the size range of 0.2-1 mm made up 91 % and 66 % of the total in May and October, respectively. The most common types of MPs in May were polyethylene terephthalate (PET) at 40 % and polyethylene (PE) at 26 %. In October, PET was the most prevalent at 71 %, followed by poly(ether-ether-ketone)(PEEK) at 11 %. Certain microbial taxa, such as Actinobacteriota, Pseudomonas, and Vicinamibacteraceae, associated with MP degradation or complex carbon chain breakdown, were notably enriched in sediment areas with high MP concentrations. A significant positive correlation was observed between the abundance of MPs in sediments and Actinobacteriota. Additionally, the abundance of Thiobacillus, Ca.competibacter, and other bacteria involved in soil element cycling showed a significant positive correlation with the organic matter content in the sediments. Anaerobic bacteria like Thermoanaerobacterium displayed a significant positive correlation with water depth. Our study reveals the presence, composition, and distribution of MPs in Bosten Lake's sediments, shedding light on their potential ecological impact.

Non-invasive detection of systemic lupus erythematosus using SERS serum detection technology and deep learning algorithms.

Systemic lupus erythematosus (SLE) is an autoimmune disease with multiple symptoms, and its rapid screening is the research focus of surface-enhanced Raman scattering (SERS) technology. In this study, gold@silver-porous silicon (Au@Ag-PSi) composite substrates were synthesized by electrochemical etching and in-situ reduction methods, which showed excellent sensitivity and accuracy in the detection of rhodamine 6G (R6G) and serum from SLE patients. SERS technology was combined with deep learning algorithms to model serum features using selected CNN, AlexNet, and RF models. 92 % accuracy was achieved in classifying SLE patients by CNN models, and the reliability of these models in accurately identifying sera was verified by ROC curve analysis. This study highlights the great potential of Au@Ag-PSi substrate in SERS detection and introduces a novel deep learning approach for SERS for accurate screening of SLE. The proposed method and composite substrate provide significant value for rapid, accurate, and noninvasive SLE screening and provide insights into SERS-based diagnostic techniques.

Sandwich-like CuNPs@AgNPs@PSB SERS substrates for sensitive detection of R6G and Forchlorfenuron.

The development of a highly sensitive, synthetically simple and economical SERS substrate is technically very important. A fast, economical, sensitive and reproducible CuNPs@AgNPs@ Porous silicon Bragg reflector (PSB) SERS substrate was prepared by electrochemical etching and in situ reduction method. The developed CuNPs@AgNPs@PSB has a large specific surface area and abundant "hot spot" region, which makes the SERS performance excellent. Meanwhile, the successful synthesis of CuNPs@AgNPs can not only modulate the plasmon resonance properties of nanoparticles, but also effectively prolong the time stability of Cu nanoparticles. The basic performance of the substrate was evaluated using rhodamine 6G (R6G). (Detection limit reached 10-15 M, R2 = 0.9882, RSD = 5.3 %) The detection limit of Forchlorfenuron was 10 µg/L. The standard curve with a regression coefficient of 0.979 was established in the low concentration range of 10 µg/L -100 µg/L. This indicates that the prepared substrates can accomplish the detection of pesticide residues in the low concentration range. The prepared high-performance and high-sensitivity SERS substrate have a very promising application in detection technology.

Carboxyl-functionalized two-dimensional MXene-Au nanocomposites were prepared as SERS substrates for the detection of melamine in dairy products.

In the present study, we address the limitations of conventional surface-enhanced Raman scattering (SERS) techniques for sensitive and stable detection of melamine in food products, especially dairy. To overcome these challenges, we developed a novel SERS-active substrate by incorporating gold nanoparticles (AuNPs) onto carboxyl-functionalized two-dimensional (2D) MXene material doped with nitrides, specifically Au-Ti2N-COOH. Our strategy leverages the unique physicochemical properties of MXene, a class of atomically thin, 2D transition metal carbides/nitrides, with tunable surface functionalities. By modifying the MXene surface with AuNPs and introducing carboxyl groups (-COOH), we successfully enhanced the interaction between the substrate and melamine molecules. The carboxyl groups form hydrogen bonds with the amino groups on the melamine's triazine ring, facilitating the adsorption of melamine molecules within the 'hotspot' regions responsible for SERS signal amplification. A series of characterization methods were used to confirm the successful synthesis of Au-Ti2N-COOH composites.Using Au-Ti2N-COOH as the SERS substrate, we detected melamine in spiked dairy product samples with significantly enhanced sensitivity and stability compared to nitride-doped MXene alone. The detection limit in liquid milk stands at 3.7008 µg kg-1, with spike recovery rates ranging from 99.84% to 107.55% and an approximate RSD of 5%. This work demonstrates the effectiveness of our approach in designing a label-free, rapid, and robust SERS platform for the accurate quantitation of melamine contamination in food, thereby mitigating health risks associated with melamine adulteration.

Surface defect detection of industrial components based on vision.

Early and effective surface defect detection in industrial components can avoid the occurrence of serious safety hazards. Since most industrial component surfaces have tiny defects with high similarity to the detection background, there are often issues of missed or false detections when defects are detected, leading to low detection accuracy. To deal with the aforementioned issue, this essay suggests a high-precision detection model for surface defects in industrial components based on the YOLOv5 algorithm. First, the original spatial pyramid pooling (SPPF) is innovated by proposing the SPPFKCSPC module, which improves the network's capacity for feature extraction from targets at different scales and fuses multiscale features better. Then, C3 is combined with SPPFKCSPC and replaces the C3 module of the backbone network, which improves feature expression and enhances the receptive field of the network. Finally, the coordinate attention mechanism (CA) has been embedded into the YOLOv5 neck network, and the bounding box regression loss function of the algorithm is improved to EIOU, not only improving the precision of the target localization and recognition model but also enhancing the overall network performance. Based on the public datasets NEU-DET and PV-Multi-Defect, multiple sets of experiments were conducted using innovative algorithms. On the NEU-DET dataset, we got a mean average accuracy (mAP) of 88.3%, which is 7.2% greater than the original approach. On the PV-Multi-Defect dataset, the mAP value reached 97.5%, an improvement of 1.5%. As shown by the experimental data, the detection results significantly improved.

MS-Net: a novel lightweight and precise model for plant disease identification.

The rapid development of image processing technology and the improvement of computing power in recent years have made deep learning one of the main methods for plant disease identification. Currently, many neural network models have shown better performance in plant disease identification. Typically, the performance improvement of the model needs to be achieved by increasing the depth of the network. However, this also increases the computational complexity, memory requirements, and training time, which will be detrimental to the deployment of the model on mobile devices. To address this problem, a novel lightweight convolutional neural network has been proposed for plant disease detection. Skip connections are introduced into the conventional MobileNetV3 network to enrich the input features of the deep network, and the feature fusion weight parameters in the skip connections are optimized using an improved whale optimization algorithm to achieve higher classification accuracy. In addition, the bias loss substitutes the conventional cross-entropy loss to reduce the interference caused by redundant data during the learning process. The proposed model is pre-trained on the plant classification task dataset instead of using the classical ImageNet for pre-training, which further enhances the performance and robustness of the model. The constructed network achieved high performance with fewer parameters, reaching an accuracy of 99.8% on the PlantVillage dataset. Encouragingly, it also achieved a prediction accuracy of 97.8% on an apple leaf disease dataset with a complex outdoor background. The experimental results show that compared with existing advanced plant disease diagnosis models, the proposed model has fewer parameters, higher recognition accuracy, and lower complexity.

Salient Object Detection Based on Optimization of Feature Computation by Neutrosophic Set Theory.

In recent saliency detection research, too many or too few image features are used in the algorithm, and the processing of saliency map details is not satisfactory, resulting in significant degradation of the salient object detection result. To overcome the above deficiencies and achieve better object detection results, we propose a salient object detection method based on feature optimization by neutrosophic set (NS) theory in this paper. First, prior object knowledge is built using foreground and background models, which include pixel-wise and super-pixel cues. Simultaneously, the feature maps are selected and extracted for feature computation, allowing the object and background features of the image to be separated as much as possible. Second, the salient object is obtained by fusing the features decomposed by the low-rank matrix recovery model with the object prior knowledge. Finally, for salient object detection, we present a novel mathematical description of neutrosophic set theory. To reduce the uncertainty of the obtained saliency map and then obtain good saliency detection results, the new NS theory is proposed. Extensive experiments on five public datasets demonstrate that the results are competitive and superior to previous state-of-the-art methods.

Novel SERS biosensor for rapid detection of breast cancer based on Ag2O-Ag-PSi nanochips.

Ag2O-Ag-PSi (porous silicon) surface-enhanced Raman scattering (SERS) chip was successfully synthesized by electrochemical corrosion, in situ reduction and heat treatment technology. The influence of different heat treatment temperature on SERS performance of the chip is studied. The results show that the chip treated at 300 °C has the best SERS performance. The chip was composed of Ag2O-Ag nano core shell with a diameter of 40-60 nm and porous silicon substrate. Then, the optimized chip was used to perform SERS test on serum samples from 30 healthy volunteers and 30 early breast cancer patients, and the baseline was corrected by LabSpec6 software. Finally, the data were analyzed by principal component analysis combined with t-distributed Stochastic Neighbor Embedding (PCA-t-SNE). The results showed that the accuracy of the improved substrate combined with multivariate statistical method was 98%. The shelf life of the chips exceeded six months due to the presence of the Ag2O shell. This study provides a basis for developing a low-cost rapid and sensitive early screening technology for breast cancer.

Multi-Focus Image Fusion via Distance-Weighted Regional Energy and Structure Tensor in NSCT Domain.

In this paper, a multi-focus image fusion algorithm via the distance-weighted regional energy and structure tensor in non-subsampled contourlet transform domain is introduced. The distance-weighted regional energy-based fusion rule was used to deal with low-frequency components, and the structure tensor-based fusion rule was used to process high-frequency components; fused sub-bands were integrated with the inverse non-subsampled contourlet transform, and a fused multi-focus image was generated. We conducted a series of simulations and experiments on the multi-focus image public dataset Lytro; the experimental results of 20 sets of data show that our algorithm has significant advantages compared to advanced algorithms and that it can produce clearer and more informative multi-focus fusion images.

Functionalized magnetic nanobeads for SERS-based detection of Staphylococcus aureus.

In this study, we introduced a Raman detection technique based on a combination of functionalized magnetic beads and surface-enhanced Raman scattering (SERS) tags to develop a rapid and sensitive strategy for the detection of Staphylococcus aureus (S. aureus), a typical foodborne pathogen. Polyethylene glycol (PEG) and bovine serum albumin (BSA) dual-mediated teicoplanin functionalized magnetic beads (TEI-BPBs) were prepared for separation of target bacteria. SERS tags were used to immobilize antibodies on gold surfaces with bifunctional linker proteins to ensure specific recognition of S. aureus. Under optimal conditions, the combination of TEI-BPBs and SERS tags showed reliable performance, exhibiting good capture efficiency even in the presence of 106 CFU mL-1 of non-target bacteria. The SERS tag provided an effective hot spot for subsequent Raman detection, presenting good linearity in the range of 102-107 CFU mL-1. Good performance has also been shown in detecting target bacteria in milk samples, where it has a recovery of 95.5-101.3%. Thus, the highly sensitive Raman detection technique combined with TEI-BPBs capture probes and SERS tags is a promising method for the detection of foodborne pathogens in food or clinical samples.

A Lightweight Feature Distillation and Enhancement Network for Super-Resolution Remote Sensing Images.

Super-resolution (SR) images based on deep networks have achieved great accomplishments in recent years, but the large number of parameters that come with them are not conducive to use in equipment with limited capabilities in real life. Therefore, we propose a lightweight feature distillation and enhancement network (FDENet). Specifically, we propose a feature distillation and enhancement block (FDEB), which contains two parts: a feature-distillation part and a feature-enhancement part. Firstly, the feature-distillation part uses the stepwise distillation operation to extract the layered feature, and here we use the proposed stepwise fusion mechanism (SFM) to fuse the retained features after stepwise distillation to promote information flow and use the shallow pixel attention block (SRAB) to extract information. Secondly, we use the feature-enhancement part to enhance the extracted features. The feature-enhancement part is composed of well-designed bilateral bands. The upper sideband is used to enhance the features, and the lower sideband is used to extract the complex background information of remote sensing images. Finally, we fuse the features of the upper and lower sidebands to enhance the expression ability of the features. A large number of experiments show that the proposed FDENet both produces less parameters and performs better than most existing advanced models.

Coupling annealed silver nanoparticles with a porous silicon Bragg mirror SERS substrate and machine learning for rapid non-invasive disease diagnosis.

Ag2O-Ag-porous silicon Bragg mirror (PSB) composite SERS substrates were successfully synthesized by using a combination of electrochemical and thermochemical methods. Test results showed that the SERS signal increased and decreased as the annealing temperature used for the substrate increased, where the most intense SERS signal was obtained using a substrate annealed at 300 °C. Stability test results showed substantial enhancement of the SERS signal intensity of the Ag2O-Ag-PSB composite one month after preparation compared with that of conventional Ag-PSB. We conclude that Ag2O nanoshells play an essential role in SERS signal enhancement. Ag2O prevents natural oxidation of Ag nanoparticles (AgNPs) and has a solid localized surface plasmon resonance (LSPR). SERS signal enhancement was tested using this substrate for serum from patients with Sjögren's syndrome (SS) and Diabetic nephropathy (DN), as well as from healthy controls (HC). SERS feature extraction was performed using principal component analysis (PCA). The extracted features were analyzed by a support vector machine (SVM) algorithm. Finally, a rapid screening model for SS and HC, as well as DN and HC, was developed and used to perform controlled experiments. The results showed that the diagnostic accuracy, sensitivity and selectivity for SERS technology combined with machine learning algorithms reached 90.7%, 93.4% and 86.7% for SS/HC and 89.3%, 95.6% and 80% for DN/HC, respectively. The results of this study show that the composite substrate has excellent potential to be developed into a commercially available SERS chip for medical testing.

Sparse Representation-Based Multi-Focus Image Fusion Method via Local Energy in Shearlet Domain.

Multi-focus image fusion plays an important role in the application of computer vision. In the process of image fusion, there may be blurring and information loss, so it is our goal to obtain high-definition and information-rich fusion images. In this paper, a novel multi-focus image fusion method via local energy and sparse representation in the shearlet domain is proposed. The source images are decomposed into low- and high-frequency sub-bands according to the shearlet transform. The low-frequency sub-bands are fused by sparse representation, and the high-frequency sub-bands are fused by local energy. The inverse shearlet transform is used to reconstruct the fused image. The Lytro dataset with 20 pairs of images is used to verify the proposed method, and 8 state-of-the-art fusion methods and 8 metrics are used for comparison. According to the experimental results, our method can generate good performance for multi-focus image fusion.

An efficient sharding consensus algorithm for consortium chains.

The consensus algorithm is very critical in any blockchain system, because it directly affects the performance and security of the blockchain system. At present, the classic Practical Byzantine Fault Tolerance Algorithm (PBFT), which is mainly used in the consortium chain, will lead to system communication congestion and reduced throughput when the number of nodes increases, so the PBFT algorithm is not suitable for large-scale consortium chains. In response to the above problems, this paper proposes a new clustering-based sharding consensus algorithm (KBFT), which aims to ensure that the consortium chain takes into account decentralization, security and scalability. The KBFT algorithm first uses the K-prototype clustering algorithm to shard the nodes in the network according to mixed attributes, and second, disjoint transactions are used to reach consensus in parallel in different shards. Concurrently, the KBFT algorithm introduces a supervision mechanism and a node credit mechanism, which is used to supervise and score the behavior of the nodes and select the proxy nodes, which improves security. We discuss the choice of shard size with the help of the binomial probability distribution and analyze the probability that the system can successfully form a global block under different node failure probabilities. Finally, the proposed algorithm is evaluated through theoretical analysis and simulation experiments. Results show that the proposed algorithm achieves a marked improvement in scalability and throughput along with a marked reduction in communication complexity compared with the classic baseline algorithm PBFT in this field of study, which improves the operating efficiency of the system and simultaneously guarantees the security and robustness of the system.

A high-performance fluorescent probe for detection of cysteine in plasma constructed by combining Cu(I) and 2,5-dimercapto-1,3,4-thiadiazole.

A high-performance fluorescent probe 2,5-dimercapto-1,3,4-thiadiazole copper nanoparticles (DMTD-CuNPs) was synthesized by hydrothermal method based on monovalent copper (Cu(I)) and 2,5-dimercapto-1,3,4-thiadiazole (DMTD), and it can effectively detect cysteine (Cys) in plasma. Experiments show that DMTD can reduces band gap of Cu(I) in DMTD-CuNPs, promote charge transfer transition from DMTD to Cu(I) and significantly enhance fluorescence intensity of DMTD-CuNPs at 515 nm. The large Stokes shift of DMTD-CuNPs is 315 nm, which can reduce the self-quenching of probe fluorescence and improves detection accuracy of the probe. In the presence of Cys, fluorescence of DMTD-CuNPs at 515 nm is significantly quenched because Cys reacts with Cu(I) in DMTD-CuNPs through Cu-S bond to form reduced charge transfer, which can be successfully used for the detection of Cys. Linear range and detection limit for Cys detection are 25-65 µM and 50 nM, respectively. Furthermore, feasibility of detecting Cys in plasma using DMTD-CuNPs probe was evaluated by standard addition method, and the absolute recovery is 96-99%. Such a DMTD-CuNPs probe shows high sensitivity, good selectivity and low detection limit for Cys, which is expected to be used for the practical analysis of Cys in plasma.

Spectral-Free Double Light Detection of DNA Based on a Porous Silicon Bragg Mirror.

To improve the detection sensitivity of a porous silicon optical biosensor in the real-time detection of biomolecules, a non-spectral porous silicon optical biosensor technology, based on dual-signal light detection, is proposed. Double-light detection is a combination of refractive index change detection and fluorescence change detection. It uses quantum dots to label probe molecules to detect target molecules. In the double-signal-light detection method, the first detection-signal light is the detection light that is reflected from the surface of the porous silicon Bragg mirror. The wavelength of the detection light is the same as the wavelength of the photonic band gap edge of the porous silicon Bragg mirror. CdSe/ZnS quantum dots are used to label the probe DNA and hybridize it with the target DNA molecules in the pores of porous silicon to improve its effective refractive index and enhance the detection-reflection light. The second detection-signal light is fluorescence, which is generated by the quantum dots in the reactant that are excited by light of a certain wavelength. The Bragg mirror structure further enhances the fluorescence signal. A digital microscope is used to simultaneously receive the digital image of two kinds of signal light superimposed on the surface of porous silicon, and the corresponding algorithm is used to calculate the change in the average grey value before and after the hybridization reaction to calculate the concentration of the DNA molecules. The detection limit of the DNA molecules was 0.42 pM. This method can not only detect target DNA by hybridization, but also detect antigen by immune reaction or parallel biochip detection for a porous silicon biosensor.

Image Reconstruction Based on Progressive Multistage Distillation Convolution Neural Network.

To address the problem that some current algorithms suffer from the loss of some important features due to rough feature distillation and the loss of key information in some channels due to compressed channel attention in the network, we propose a progressive multistage distillation network that gradually refines the features in stages to obtain the maximum amount of key feature information in them. In addition, to maximize the network performance, we propose a weight-sharing information lossless attention block to enhance the channel characteristics through a weight-sharing auxiliary path and, at the same time, use convolution layers to model the interchannel dependencies without compression, effectively avoiding the previous problem of information loss in channel attention. Extensive experiments on several benchmark data sets show that the algorithm in this paper achieves a good balance between network performance, the number of parameters, and computational complexity and achieves highly competitive performance in both objective metrics and subjective vision, which indicates the advantages of this paper's algorithm for image reconstruction. It can be seen that this gradual feature distillation from coarse to fine is effective in improving network performance. Our code is available at the following link: https://github.com/Cai631/PMDN.