Multiscale Processing Research Articles

Image-Based Quantitative Analysis of Epidermal Morphology in Wild Potato Leaves

The epidermal leaf patterns of plants exhibit remarkable diversity in cell shapes, sizes, and arrangements, driven by environmental interactions that lead to significant adaptive changes even among closely related species. The Solanaceae family, known for its high diversity of adaptive epidermal structures, has traditionally been studied using qualitative phenotypic descriptions. To advance this, we developed a workflow combining multi-scale computer vision, image processing, and data analysis to extract digital descriptors for leaf epidermal cell morphology. Applied to nine wild potato species, this workflow quantified key morphological parameters, identifying descriptors for trichomes, stomata, and pavement cells, and revealing interdependencies among these traits. Principal component analysis (PCA) highlighted two main axes, accounting for 45% and 21% of variance, corresponding to features such as guard cell shape, trichome length, stomatal density, and trichome density. These axes aligned well with the historical and geographical origins of the species, separating southern from Central American species, and forming distinct clusters for monophyletic groups. This workflow thus establishes a quantitative foundation for investigating leaf epidermal cell morphology within phylogenetic and geographic contexts.

Self-Powered Dye-Sensitized Solar-Cell-Based Synaptic Devices for Multi-Scale Time-Series Data Processing in Physical Reservoir Computing.

Physical reservoir computing (PRC) using synaptic devices has attracted attention as a promising edge artificial intelligence device. To handle time-series data on various time scales, it is necessary to fabricate devices with the desired time scale. In this study, we fabricated a dye-sensitized solar-cell-based synaptic device with controllable time constants by changing the light intensity. This device showed synaptic features, such as paired-pulse facilitation and paired-pulse depression, in response to light intensity. Moreover, we found that the high computational performance of the time-series data processing task was achieved by changing the light intensity, even when the input pulse width was varied. In addition, the fabricated device can be used for motion recognition tasks. This study paves the way for realizing multiple time-scale PRC.

Attenuative Unified U-Net with Conditional PatchMatch Algorithm for Edge Preserving Salient Object Detection

The rise of Deep Neural Networks (DNNs) has significantly boosted salient object detection in computer vision tasks. However, downsampling methods like striding and pooling often introduce bluish artifacts near edges, hindering detection accuracy. To address this, the "Attenuative Unified U-Net with Conditional PatchMatch Algorithm" is proposed. The method employs Gaussian smoothing for noise reduction and depth refinement in preprocessing. In existing methods, the Edge-preserving salient object detection struggles with cluttered backgrounds and connected objects, like a bird on a rock, making it difficult to accurately distinguish edges and salient regions due to shared boundaries. Hence, an Attenuative Unified Backpropagated Entropy UNet is proposed, which integrates an Attention Mechanism that enhance feature map spatial weight assignment, emphasizing features or regions that are judged most relevant for edge and salient region detection. Then, the Cascaded Laplace pyramid is incorporated into the Unified U-Net’s design for multi-scale information processing, capturing contextual details effectively. The Backpropagated Entropy Loss Function is then created to ensure accurate and confident fused image output. The depth map, edge map, and salient area map are combined in the U-Net’s final output layer to create an overall fused image. Next, to addresses depth discontinuity problem a noteworthy novel idea the Conditional PatchMatch Random Field Algorithm (CPMRF), which combines PatchMatch efficiency and CRFs’ contextual modeling. PatchMatch iteratively propagates matches from neighboring patches to find similar patches in the remaining portion of the image and Conditional Random Field (CRF) is the next step, which improves the PatchMatch results by taking into account the connections between the matched patches and their neighboring areas. Hence, the experimental outcomes of the proposed model effectively show the improved detection of edge in salient object detection with better accuracy, precision, recall, MaxF, and F1 score with minimized MAE.

ECM-YOLO: a real-time detection method of steel surface defects based on multiscale convolution

Steel surface defects, characterized by multiple types, varied scales, and overlapping occurrences, directly impact the quality, performance, and reliability of industrial products. Proposing a high-precision and high-speed steel surface defect detection algorithm is crucial for ensuring product quality. In this regard, this paper introduces ECM-YOLO, a detection network based on YOLOv8n. First, addressing the insufficient information capture of the C2f module, the C2f enhanced multiscale convolution processing (C2f_EMSCP) module is proposed, enhancing global and local feature capture capabilities through multiscale convolutions. Second, to further enhance the network’s robustness and focus on critical information, the channel prior convolutional attention (CPCA) mechanism is integrated between the backbone and neck networks to facilitate more efficient information transmission. Last, a novel, to the best of our knowledge, detection head, i.e., multiscale simple and efficient anchor matching head (MultiSEAMHead), is proposed to mitigate accuracy issues arising from overlaps between different types of defects. Experimental results demonstrate that ECM-YOLO achieves mAPs of 78.9% and 68.2% on the NEU-DET and GC 10-DET data sets, respectively, outperforming YOLOv8n by 2.5% and 4.4%. Moreover, ECM-YOLO excels in model parameters, computational efficiency, and inference speed compared with other models. These findings highlight the applicability of ECM-YOLO for real-time defect detection in industrial settings.

Study on the detection of power line operation anomalies in the Hexi Corridor based on multi-scale filtering of BeiDou satellite signals

In the artificial intelligence detection of power line anomalies, the identification effect is determined by the positioning process of two-dimensional map, which is also a difficult point in this field Therefore, this paper proposes a method to detect the abnormal operation of power lines in Hexi Corridor based on multi-scale filtering of Beidou satellite signals. Multi-scale processing is carried out by wavelet transform to obtain multiple wavelet basis functions of different scales. On this basis, a two-dimensional time–frequency diagram is drawn according to the results of continuous wavelet transform. The adaptive multi-time frequency entropy method is selected to divide the two-dimensional time–frequency map into multiple layers and scales, and the adaptive multi-scale time–frequency entropy of the time–frequency plane is calculated. After smoothing the obtained power line position signal by S-G filtering method, the power line position signal of Hexi Corridor is obtained by GPS/ Beidou dual-mode satellite positioning system, and the positioning signal is optimized by using the positioning mechanism of multiple receiving stations. Realize the abnormal operation detection of power lines in Hexi Corridor. The experimental results show that this method can obtain more accurate positioning signals, and the geometric accuracy factor GDOP fluctuates in the range of 0 ∼ 2. The unsmooth signal can be effectively removed; The maximum error result of abnormal position detection is 1.22kn；; At the same time, the abnormal displacement section of the line can be determined.

High-quality femtosecond laser cutting of battery electrodes with enhanced electrochemical performance by regulating the taper angle: Promoting green manufacturing and chemistry

Laser cutting electrode technology is an environmentally friendly and sustainable manufacturing method that offers many benefits, such as low energy consumption and greenhouse gases, no waste production, no tool wear, flexible manipulation, and multi-scale processing. However, achieving high cutting quality with adjustable notch shape and controllable dimension precision is still a cutting-edge challenge. In addition, the influence of laser cutting on the microscopic structure and electrochemical performance of battery electrodes has not been fully illustrated. Here, the Li4Ti5O12 (LTO) electrode is cut using a femtosecond laser technology. The processing parameters are systematically optimized, and the influence of laser cutting taper structure on the structure and performance of LTO electrodes is comprehensively investigated. It is found that laser photoionization and the generated plasma could create oxygen vacancies and thus more Ti3+/Ti4+ defects in the LTO electrodes, resulting in increased electronic conductivity. Moreover, proper taper structure could help improve electrolyte infiltration as well as Li+ diffusion kinetics. As a result, the laser cutting LTO electrode with the optimized taper angle of T = 0.004 shows excellent electrochemical performance. The capacity retention is more than 99.2 % after 400 cycles at 1C, and the specific capacity reaches ∼60 mAh g−1 at a high rate of 60C, which is about 1.7 times higher than that of mechanically cutting LTO electrodes. The outstanding electrochemical performance is also demonstrated by LiFePO4||Li4Ti5O12 full cells. This work could pave the way for the development and application of laser technology in cutting battery electrodes.

A Deformable and Multi-Scale Network with Self-Attentive Feature Fusion for SAR Ship Classification

The identification of ships in Synthetic Aperture Radar (SAR) imagery is critical for effective maritime surveillance. The advent of deep learning has significantly improved the accuracy of SAR ship classification and recognition. However, distinguishing features between different ship categories in SAR images remains a challenge, particularly as the number of categories increases. The key to achieving high recognition accuracy lies in effectively extracting and utilizing discriminative features. To address this, we propose DCN-MSFF-TR, a novel recognition model inspired by the Transformer encoder–decoder architecture. Our approach integrates a deformable convolutional module (DCN) within the backbone network to enhance feature extraction. Additionally, we introduce multi-scale self-attention processing from the Transformer into the feature hierarchy and fuse these representations at appropriate levels using a feature pyramid strategy. This enables each layer to leverage both its own information and synthesized features from other layers, enhancing feature representation. Extensive evaluations on the OpenSARShip-3-Complex and OpenSARShip-6-Complex datasets demonstrate the effectiveness of our method. DCN-MSFF-TR achieves average recognition accuracies of 78.1% and 66.7% on the three-class and six-class datasets, respectively, outperforming existing recognition models and showcasing its superior capability in accurately identifying ship categories in SAR images.

A differential network with multiple gated reverse attention for medical image segmentation

UNet architecture has achieved great success in medical image segmentation applications. However, these models still encounter several challenges. One is the loss of pixel-level information caused by multiple down-sampling steps. Additionally, the addition or concatenation method used in the decoder can generate redundant information. These limitations affect the localization ability, weaken the complementarity of features at different levels and can lead to blurred boundaries. However, differential features can effectively compensate for these shortcomings and significantly enhance the performance of image segmentation. Therefore, we propose MGRAD-UNet (multi-gated reverse attention multi-scale differential UNet) based on UNet. We utilize the multi-scale differential decoder to generate abundant differential features at both the pixel level and structure level. These features which serve as gate signals, are transmitted to the gate controller and forwarded to the other differential decoder. In order to enhance the focus on important regions, another differential decoder is equipped with reverse attention. The features obtained by two differential decoders are differentiated for the second time. The resulting differential feature obtained is sent back to the controller as a control signal, then transmitted to the encoder for learning the differential feature by two differential decoders. The core design of MGRAD-UNet lies in extracting comprehensive and accurate features through caching overall differential features and multi-scale differential processing, enabling iterative learning from diverse information. We evaluate MGRAD-UNet against state-of-theart (SOTA) methods on two public datasets. Our method surpasses competitors and provides a new approach for the design of UNet.

Research on sports image classification method based on SE-RES-CNN model

As computer image processing and digital technologies advance, creating an efficient method for classifying sports images is crucial for the rapid retrieval and management of large image datasets. Traditional manual methods for classifying sports images are impractical for large-scale data and often inaccurate when distinguishing similar images. This paper introduces an SE module that adaptively adjusts the weights of input feature mapping channels, and a Res module that excels in deep feature extraction, preventing gradient vanishing, multi-scale processing, and enhancing generalization in image recognition. Through extensive experimentation on network structure adjustments, the SE-RES-CNN neural network model is applied to sports image classification. The model is trained on a sports image classification dataset from Kaggle, alongside VGG-16 and ResNet50 models. Training results show that the proposed SE-RES-CNN model improves classification accuracy by approximately 5% compared to VGG-16 and ResNet50 models. Testing revealed that the SE-RES-CNN model classifies 100 out of 500 sports images in 6 s, achieving an accuracy rate of up to 98% and a single prediction time of 0.012 s. This validates the model's accuracy and effectiveness, significantly enhancing sports image retrieval and classification efficiency. This validates the model's accuracy and effectiveness, significantly enhancing sports image retrieval and classification efficiency.

ICPNet: Advanced Maize Leaf Disease Detection with Multidimensional Attention and Coordinate Depthwise Convolution.

Maize is an important crop, and the detection of maize diseases is critical for ensuring food security and improving agricultural production efficiency. To address the challenges of difficult feature extraction due to the high similarity among maize leaf disease species, the blurring of image edge features, and the susceptibility of maize leaf images to noise during acquisition and transmission, we propose a maize disease detection method based on ICPNet (Integrated multidimensional attention coordinate depthwise convolution PSO (Particle Swarm Optimization)-Integrated lion optimisation algorithm network). Firstly, we introduce a novel attention mechanism called Integrated Multidimensional Attention (IMA), which enhances the stability and responsiveness of the model in detecting small speckled disease features by combining cross-attention and spatial channel reconstruction methods. Secondly, we propose Coordinate Depthwise Convolution (CDC) to enhance the accuracy of feature maps through multi-scale convolutional processing, allowing for better differentiation of the fuzzy edges of maize leaf disease regions. To further optimize model performance, we introduce the PSO-Integrated Lion Optimisation Algorithm (PLOA), which leverages the exploratory stochasticity and annealing mechanism of the particle swarm algorithm to enhance the model's ability to handle mutation points while maintaining training stability and robustness. The experimental results demonstrate that ICPNet achieved an average accuracy of 88.4% and a precision of 87.3% on the self-constructed dataset. This method effectively extracts the tiny and fuzzy edge features of maize leaf diseases, providing a valuable reference for disease control in large-scale maize production.

An improved YOLOv8 safety helmet wearing detection network

In the field of industrial safety, wearing helmets plays a vital role in ensuring workers’ health. Aiming at addressing the complex background in the industrial environment, caused by differences in distance, the helmet small target wearing detection methods for misdetection and omission detection problems are needed. An improved YOLOv8 safety helmet wearing detection network is proposed to enhance the capture of details, improve multiscale feature processing and improve the accuracy of small target detection by introducing Dilation-wise residual attention module, atrous spatial pyramid pooling and normalized Wasserstein distance loss function. Experiments were conducted on the SHWD dataset, and the results showed that the mAP of the improved network improved to 92.0%, which exceeded that of the traditional target detection network in terms of accuracy, recall, and other key metrics. These findings further improved the detection of helmet wearing in complex environments and greatly enhanced the accuracy of detection.

A multiscale time-series decomposition learning for crude oil price forecasting

Crude oil price forecasting is important for market participants and policymakers. However, accurately tracking oil prices is quite a challenging task due to the complexity of temporal oil data and the nonlinear relationships involved in the forecasting task. In this study, a multiscale time-series decomposition learning framework is proposed to deal with this issue. First, a multiscale temporal processing module is designed to capture different frequency time-series patterns in historical data at various scales. Then, a multiscale decomposition technique is applied to decompose historical crude oil data into various temporal modes, involving global shared information across multiple scales, as well as local specific information that varies at each scale. Finally, a multiscale fusion mechanism is employed to combine these information, which are further used as inputs to construct nonlinear and complex predictive models for crude oil prices. A series of experiments conducted on Shanghai crude oil market demonstrate that the proposed approach outperforms several econometric and machine learning models.

GCFormer: Multi-scale feature plays a crucial role in medical images segmentation

Transformer-based networks are becoming indispensable in the field of medical image segmentation. However, most Transformer-based methods overlook the impact of different scale features on encoding efficiency and neglect the fusion and further processing of multi-scale features. Due to the lack of learning different scale features, the reduction of noise interference is not effectively achieved. Consequently, there is an inability to distinctly differentiate between the target segmented area and the surrounding tissues, leading to subpar segmentation results. This issue is particularly pronounced when dealing with multiple segmented areas, where the edges of various organs and tissues cannot be well identified. To enhance learning diversity, we propose the Global Context Transformer (GCFormer), a medical image segmentation network that combines Transformers with CNN. In our proposed network, a novel multi-scale feature processing mechanism is adopted to reasonably encode and decode features at different scales, thereby improving segmentation efficiency. We employ the Global Token Generator (GTG) module to filter and partition multi-scale features, extracting useful information. The encoder incorporates the Pass Down module to fuse multi-scale information, while the decoder efficiently concatenates different-scale features using the Cat module. Experimental results indicate that our proposed algorithm surpasses other mainstream methods in terms of segmentation capability.

Multiscale Modes of Functional Brain Connectivity.

Information processing in the brain spans from localised sensorimotor processes to higher-level cognition that integrates across multiple regions. Interactions between and within these subsystems enable multiscale information processing. Despite this multiscale characteristic, functional brain connectivity is often either estimated based on 10-30 distributed modes or parcellations with 100-1000 localised parcels, both missing across-scale functional interactions. We present Multiscale Probabilistic Functional Modes (mPFMs), a new mapping which comprises modes over various scales of granularity, thus enabling direct estimation of functional connectivity within- and across-scales. Crucially, mPFMs emerged from data-driven multilevel Bayesian modelling of large functional MRI (fMRI) populations. We demonstrate that mPFMs capture both distributed brain modes and their co-existing subcomponents. In addition to validating mPFMs using simulations and real data, we show that mPFMs can predict ~900 personalised traits from UK Biobank more accurately than current standard techniques. Therefore, mPFMs can offer a paradigm shift in functional connectivity modelling and yield enhanced fMRI biomarkers for traits and diseases.

Hyperspectral Anomaly Detection Using Reconstruction Fusion of Quaternion Frequency Domain Analysis.

Most existing techniques consider hyperspectral anomaly detection (HAD) as background modeling and anomaly search problems in the spatial domain. In this article, we model the background in the frequency domain and treat anomaly detection as a frequency-domain analysis problem. We illustrate that spikes in the amplitude spectrum correspond to the background, and a Gaussian low-pass filter performing on the amplitude spectrum is equivalent to an anomaly detector. The initial anomaly detection map is obtained by the reconstruction with the filtered amplitude and the raw phase spectrum. To further suppress the nonanomaly high-frequency detailed information, we illustrate that the phase spectrum is critical information to perceive the spatial saliency of anomalies. The saliency-aware map obtained by phase-only reconstruction (POR) is used to enhance the initial anomaly map, which realizes a significant improvement in background suppression. In addition to the standard Fourier transform (FT), we adopt the quaternion FT (QFT) for conducting multiscale and multifeature processing in a parallel way, to obtain the frequency domain representation of the hyperspectral images (HSIs). This helps with robust detection performance. Experimental results on four real HSIs validate the remarkable detection performance and excellent time efficiency of our proposed approach when compared to some state-of-the-art anomaly detection methods.

Detecting Road Intersections from Crowdsourced Trajectory Data Based on Improved YOLOv5 Model

In recent years, the rapid development of autonomous driving and intelligent driver assistance has brought about urgent demands on high-precision road maps. However, traditional road map production methods mainly rely on professional survey technologies, such as remote sensing and mobile mapping, which suffer from high costs, object occlusions, and long updating cycles. In the era of ubiquitous mapping, crowdsourced trajectory data offer a new and low-cost data resource for the production and updating of high-precision road maps. Meanwhile, as key nodes in the transportation network, maintaining the currency and integrity of road intersection data is the primary task in enhancing map updates. In this paper, we propose a novel approach for detecting road intersections based on crowdsourced trajectory data by introducing an attention mechanism and modifying the loss function in the YOLOv5 model. The proposed method encompasses two key steps of training data preparation and improved YOLOv5s model construction. Multi-scale training processing is first adopted to prepare a rich and diverse sample dataset, including various kinds and different sizes of road intersections. Particularly to enhance the model’s detection performance, we inserted convolutional attention mechanism modules into the original YOLOv5 and integrated other alternative confidence loss functions and localization loss functions. The experimental results demonstrate that the improved YOLOv5 model achieves detection accuracy, precision, and recall rates as high as 97.46%, 99.57%, and 97.87%, respectively, outperforming other object detection models.

Effects of Aspergillus flavus infection on multi-scale structures and physicochemical properties of maize starch during storage

This study systematically analyzed the effect of Aspergillus flavus infection on the maize starch multi-scale structure, physicochemical properties, processing characteristics, and synthesis regulation. A. flavus infection led to a decrease in the content of starch, an increase in the content of reactive oxygen species (ROS) and malondialdehyde (MDA), a significant decrease in the activities of peroxidase (POD) and superoxide dismutase (SOD). In addition, A. flavus infection had a significant destructive effect on the double helix structure, relative crystallinity and lamellar structure of starch, resulting in the reduction of starch viscosity, affecting the viscoelastic properties of starch, and complicating the gel formation process. However, the eugenol treatment group significantly inhibited the growth of A. flavus during maize storage, protecting the multi-scale structure and processing characteristics of maize starch from being damaged. Transcriptome analysis showed that genes involved in carbohydrate synthesis in maize were significantly downregulated and genes involved in energy synthesis were significantly upregulated, indicating that maize converted its energy storage into energy synthesis to fight the invasion of A. flavus. These results of this study enriched the mechanism of quality deterioration during maize storage, and provide theoretical and technical support for the prevention of A. flavus infection during maize storage.

Multivariate Multiscale Higuchi Fractal Dimension and Its Application to Mechanical Signals

Fractal dimension, as a common nonlinear dynamics metric, is extensively applied in biomedicine, fault diagnosis, underwater acoustics, etc. However, traditional fractal dimension can only analyze the complexity of the time series given a single channel at a particular scale. To characterize the complexity of multichannel time series, multichannel information processing was introduced, and multivariate Higuchi fractal dimension (MvHFD) was proposed. To further analyze the complexity at multiple scales, multivariate multiscale Higuchi fractal dimension (MvmHFD) was proposed by introducing multiscale processing algorithms as a technology that not only improved the use of fractal dimension in the analysis of multichannel information, but also characterized the complexity of the time series at multiple scales in the studied time series data. The effectiveness and feasibility of MvHFD and MvmHFD were verified by simulated signal experiments and real signal experiments, in which the simulation experiments tested the stability, computational efficiency, and signal separation performance of MvHFD and MvmHFD, and the real signal experiments tested the effect of MvmHFD on the recognition of multi-channel mechanical signals. The experimental results show that compared to other indicators, A achieves a recognition rate of 100% for signals in three features, which is at least 17.2% higher than for other metrics.

Deep Learning Model for Quality Assessment of Urinary Bladder Ultrasound Images using Multi-scale and Higher-order Processing.

Autonomous Ultrasound Image Quality Assessment (US-IQA) is a promising tool to aid the interpretation by practicing sonographers and to enable the future robotization of ultrasound procedures. However, autonomous US-IQA has several challenges. Ultrasound images contain many spurious artifacts, such as noise due to handheld probe positioning, errors in the selection of probe parameters and patient respiration during the procedure. Further, these images are highly variable in appearance with respect to the individual patient's physiology. We propose to use a deep Convolutional Neural Network (CNN), USQNet, which utilizes a Multi-scale and Local-to-Global Second-order Pooling (MS-L2GSoP) classifier to conduct the sonographer-like assessment of image quality. This classifier first extracts features at multiple scales to encode the inter-patient anatomical variations, similar to a sonographer's understanding of anatomy. Then, it uses second-order pooling in the intermediate layers (local) and at the end of the network (global) to exploit the second-order statistical dependency of multi-scale structural and multi-region textural features. The L2GSoP will capture the higher-order relationships between different spatial locations and provide the seed for correlating local patches, much like a sonographer prioritizes regions across the image. We experimentally validated the USQNet for a new dataset of the human urinary bladder ultrasound images. The validation involved first with the subjective assessment by experienced radiologists' annotation, and then with state-of-the-art CNN networks for US-IQA and its ablated counterparts. The results demonstrate that USQNet achieves a remarkable accuracy of 92.4% and outperforms the SOTA models by 3 - 14% while requiring comparable computation time.

Multiscale method for identifying and marking the multiform fractures from visible-light rock-mass images

Multiform fractures have a direct impact on the mechanical performance of rock masses. To accurately identify multiform fractures, the distribution patterns of grayscale and the differential features of fractures in their neighborhoods are summarized. Based on this, a multiscale processing algorithm is proposed. The multiscale process is as follows. On the neighborhood of pixels, a grayscale continuous function is constructed using bilinear interpolation, the smoothing of the grayscale function is realized by Gaussian local filtering, and the grayscale gradient and Hessian matrix are calculated with high accuracy. On small-scale blocks, the pixels are classified by adaptively setting the grayscale threshold to identify potential line segments and mini-fillings. On the global image, potential line segments and mini-fillings are spliced together by progressing the block frontier layer-by-layer to identify and mark multiform fractures. The accuracy of identifying multiform fractures is improved by constructing a grayscale continuous function and adaptively setting the grayscale thresholds on small-scale blocks. And the layer-by-layer splicing algorithm is performed only on the domain of the 2-layer small-scale blocks, reducing the complexity. By using rock mass images with different fracture types as examples, the identification results show that the proposed algorithm can accurately identify the multiform fractures, which lays the foundation for calculating the mechanical parameters of rock masses.