Morphological Processing Research Articles

Paper Defect Detection Algorithm Based on the Mathematical Morphology and Computer Vision

A method for detecting paper’s surface defects is proposed, which is based on mathematical morphology algorithm and computer vision technique. Firstly, the mean filtering and wavelet denoising in image pre-processing are used to eliminate the image noise, and wavelet analysis can achieve image enhancement and increase the contrast between the target area and the background. And then, an improved threshold Canny operator is used to segment the image to extract the complete scratch line and other defect edges. Finally, the target region features are coloured by mathematical morphology processing. The method’s stimulating result is verified in the Matlab software, from which we can conclude that the method’s detecting effect is very effective.

GNSS Urban Positioning with Vision-Aided NLOS Identification

The global navigation satellite system (GNSS) has played an important role in a broad range of consumer and industrial applications. In particular, cities have become GNSS major application scenarios; however, GNSS signals suffer from blocking, reflection and attenuation in harsh urban environments, resulting in diverse received signals, e.g., non-line-of-sight (NLOS) and multipath signals. NLOS signals often cause severe deterioration in positioning, navigation, and timing (PNT) solutions, which should be identified and excluded. In this paper, we propose a vision-aided NLOS identification method to augment GNSS urban positioning. A skyward omnidirectional camera is installed on a GNSS antenna to collect omnidirectional images of the sky region. After being rectified, these images are processed for sky region segmentation, which is improved by leveraging gradient information and energy function optimization. Image morphology processing is further employed to smooth slender boundaries. After sky region segmentation, the satellites are projected onto the omnidirectional image, from which NLOS satellites are identified. Finally, the identified NLOS satellites are excluded from GNSS PNT estimation, promoting accuracy and stability. Practical test results show that the proposed sky region segmentation module achieves over 96% accuracy, and that completely accurate NLOS identification is achieved for the experimental images. We validate the performance of our method on public datasets. Compared with the raw measurements without screening, the vision-aided NLOS identification method enables improvements of 60.3%, 12.4% and 63.3% in the E, N, and U directions, respectively, as well as an improvement of 58.5% in 3D accuracy.

Detection of skin defects on loquat using hyperspectral imaging combining both band radio and improved three-phase level set segmentation method

AbstractBackground and objectivesSkin defects are one of the primary problems that occur in post-harvest grading and processing of loquats. Skin defects lead to the loquat being easily destroyed during transportation and storage, which causes the risk of other loquats being infected, affecting the selling price.Materials and MethodsIn this paper, a method combining band radio image with an improved three-phase level set segmentation algorithm (ITPLSSM) is proposed to achieve high accuracy, rapid, and non-destructive detection of skin defects of loquats. Principal component analysis (PCA) was used to find the characteristic wavelength and PC images to distinguish four types of skin defects. The best band ratio image based on characteristic wavelength was determined.ResultsThe band ratio image (Q782/944) based on PC2 image is the best segmented image. Based on pseudo-color image enhancement, morphological processing, and local clustering criteria, the band ratio image (Q782/944) has better contrast between defective and normal areas in loquat. Finally, the ITPLSSM was used to segment the processing band ratio image (Q782/944), with an accuracy of 95.28%.ConclusionsThe proposed ITPLSSM method is effective in distinguishing four types of skin defects. Meanwhile, it also effectively segments images with intensity inhomogeneities.

The Effects of Precursors on the Morphology and Chemical Mechanical Polishing Performance of Ceria-Based Abrasives.

Ceria-based abrasives are widely used in precision chemical mechanical polishing (CMP) fields, such as thin film transistor liquid crystal display (TFT-LCD) glass substrates and integrated circuits, because of their excellent physicochemical properties. Rare earth carbonates, as precursors of ceria-based abrasives, directly affect the morphology of ceria-based abrasives, which, in turn, affects the material removal rate (MRR) and the surface roughness (Ra) after polishing. Herein, rare earth carbonates with different morphologies were obtained by adjusting reaction parameters during precipitation, including flake, spindle, and spheroid. Moreover, the phase of precursors was analyzed, and the evolution process of morphology from precursors to ceria-based abrasives was investigated. Furthermore, the effect of precursors on the polishing performance of ceria-based abrasives was explored. The results show that the primary particles of ceria-based abrasives are near-spherical, but the morphology and dispersion of the secondary particles are obviously inherited from precursors. Among them, near-spherical ceria-based abrasives prepared by nearly monodisperse near-spherical precursors show better uniformity and higher dispersion, and they not only achieve the lowest Ra but also obtain a higher MRR of 555 nm/min (9 wt.%) for polishing TFT-LCD glass substrates. The result is significant for the further optimization and application of high-performance ceria-based abrasives.

Pseudo Overlap Functions, Fuzzy Implications and Pseudo Grouping Functions with Applications

Overlap and grouping functions are important aggregation operators, especially in information fusion, classification and decision-making problems. However, when we do more in-depth application research (for example, non-commutative fuzzy reasoning, complex multi-attribute decision making and image processing), we find overlap functions as well as grouping functions are required to be commutative (or symmetric), which limit their wide applications. For the above reasons, this paper expands the original notions of overlap functions and grouping functions, and the new concepts of pseudo overlap functions and pseudo grouping functions are proposed on the basis of removing the commutativity of the original functions. Some examples and construction methods of pseudo overlap functions and pseudo grouping functions are presented, and the residuated implication (co-implication) operators derived from them are investigated. Not only that, some applications of pseudo overlap (grouping) functions in multi-attribute (group) decision-making, fuzzy mathematical morphology and image processing are discussed. Experimental results show that, in many application fields, pseudo overlap functions and pseudo grouping functions have greater flexibility and practicability.

Mesozooplankton biodiversity, vertical assemblages, and diel migration in the western tropical Pacific Ocean revealed by eDNA metabarcoding and morphological methods

Zooplankton plays a crucial role in marine ecosystems. However, due to issues with morphological identification and sampling methods, knowledge of their biodiversity and community structure in the vast pelagic zone of the western tropical Pacific Ocean is still limited compared with other areas in the North Pacific Ocean. In this study, we used environmental DNA (eDNA) metabarcoding and morphological identification approaches to investigate mesozooplankton’s biodiversity and vertical assemblages in the western tropical Pacific Ocean from the surface to 1000 m deep. Two different barcodes were used in metabarcoding: mitochondrial cytochrome oxidase I (COI) and small subunit ribosomal RNA genes (18S). Our results showed that eDNA metabarcoding revealed 2-4 times more medusae taxa than morphological processing. Both methods detected that small copepod species were dominant in the community and six of the top-10 most abundant copepod genera overlapped. Morphological results showed that mesozooplankton communities were classified into three groups: epipelagic (0–200 m), upper mesopelagic (200-500 m), and lower mesopelagic (500–1000 m). And eDNA results showed distinct vertical variations of mesozooplankton communities within the epipelagic. Both methods showed that the dominant genera Clausocalanus, Paracalanus, and Calocalanus exhibited diel patterns in relative abundance, demonstrating that the eDNA method can capture diel vertical migration (DVM) signals of copepods. Overall, our results indicate that eDNA metabarcoding is an effective and efficient tool for the study and monitoring of mesozooplankton diversity and vertical assemblages in the open sea.

Automated Analysis of Treatment Log Files and Videos for MR-Guided Stereotactic Body Radiation Therapy Treatments of Thoracic and Abdominal Tumors

<h3>Purpose/Objective(s)</h3> MR-guided stereotactic body radiation therapy (MRgSBRT) is a novel method of treating mobile tumors with robust real-time gating and on-table adaptive planning. Delivery log files and cine videos of the treatment generated by the software system can be a rich source of information for assessing MRgSBRT treatment sessions. We hypothesized that developing scripts for automated review of MRgSBRT treatments using delivery logs and cine videos would be feasible. We have included a summary of our findings from our initial experience using this automated analysis. <h3>Materials/Methods</h3> We conducted a retrospective review of patients undergoing MRgSBRT for treatment of liver, lung, and pancreas tumors. Custom Python scripts were written to automatically extract treatment delivery information from log files and export data neatly into to a spreadsheet. A separate Python script was written to extract information from delivery cine videos using optical character recognition (OCR), color thresholding, and morphological processing as well. <h3>Results</h3> Using our automated system we were able to extract the following data points from the log files: treatment site, date and time of treatment, tracking algorithm configuration, average treatment duration, total dose, fraction (fx) number, MU per fx, couch shifts, and need for plan re-optimization. Additionally, through automated analysis of treatment cine video files we were able to extract frame by frame information for tracking structure centroid displacement, beam on/off status, frame-frame similarity, and tracking confidence (TC) over the course of treatment. Table 1 is an example of the data output. <h3>Conclusion</h3> Automated analysis of MRgSBRT delivery log files and log videos is feasible and can serve as a rich resource for QA/QI to improve treatment efficiency on an MR-LINAC by quantifying a multitude of measurable parameters and formulating strategies to address any unrealized bottlenecks in the process. Future plans include expanding our automation to examine total patient time on table, monitor units per beam segment, and predicting internal tumor volumes (ITV) based on tracking target displacement obtained from the delivery videos.

Automatic calibration of crack and flaking diseases in ancient temple murals

Many precious ancient murals are seriously deteriorated due to long-term environmental influences and man-made destructions. How to effectively protect ancient murals and restore these murals’ original appearance has become an urgent problem for field experts. Modern computer technology makes it possible to virtually restore the deteriorated areas in ancient murals. However, most existing mural restoration approaches require manual calibration of the deteriorated areas, which is very difficult and time-consuming. It has been noticed that the earth layer flaking and cracks are the most common problems of ancient temple murals. This paper proposes an automatic calibration method for the earth layer flaking and cracking deterioration of murals by taking temple murals from the Ming Dynasty in Zhilin Temple as the study object. First, we extract the texture and line features of the deteriorated murals by using multi-dimensional gradient detection in the HSV space. Then, a guided filter operation is employed to highlight the disease (deteriorated) areas and meanwhile suppress other unwanted areas, which helps to extract the flaked areas or cracked lines from the digital murals. The filtered images are segmented by using an automatic threshold to obtain the initial masks of the mural disease areas. Next, we use a two-dimensional tensor voting technique to connect the discontinuous edge curves of the extracted disease areas. Afterwards, the masks of flaking and cracking areas can be generated after morphological processing. Finally, we obtained the calibration results by adding the masks to the original digital murals. Experimental results show that our method can rapidly and accurately calibrate the cracks and the earth layer flaking diseases in the ancient murals. As compared to existing calibration approaches, our method can achieve better performance in subjective visual quality and objective evaluation metrics. Moreover, the method does not need human-computer interaction. This research work provides a solid foundation for the following virtual and practical restoration of ancient murals.

U-Net for Taiwan Shoreline Detection from SAR Images

Climate change and global warming lead to changes in the sea level and shoreline, which pose a huge threat to island regions. Therefore, it is important to effectively detect the shoreline changes. Taiwan is a typical island, located at the junction of the East China Sea and the South China Sea in the Pacific Northwest, and is deeply affected by shoreline changes. In this research, Taiwan was selected as the study area. In this research, an efficient shoreline detection method was proposed based on the semantic segmentation U-Net model using the Sentinel-1 synthetic aperture radar (SAR) data of Taiwan island. In addition, the batch normalization (BN) module was added to the convolution layers in the U-Net architecture to further improve the generalization ability of U-Net and accelerate the training process. A self-built shoreline dataset was introduced to train the U-Net model and test its detection efficiency. The dataset consists of a total of 4029 SAR images covering all coastal areas of Taiwan. The training samples of the dataset were annotated by morphological processing and manual inspection. The segmentation results of U-Net were then processed by edge detection and morphological postprocessing to extract the shoreline. The experimental results showed that the proposed method could achieve a satisfactory detection performance compared with the related methods using the data provided by the Ministry of the Interior of Taiwan from 2016 to 2019 for different coastal landforms in Taiwan. Within a 5-pixel difference between the detected shoreline and the ground truth data, the F1-Meaure of the proposed method exceeded 80%. In addition, the potential of this method in shoreline change detection was validated with a sandbar located on the southwestern coast of Taiwan. Finally, the entire shoreline of Taiwan has been described by the proposed approach and the detected shoreline length was close to the actual length.

Deep Neural Network for Accurate Age Group Prediction through Pupil Using the Optimized UNet Model

Predicting age automatically from the image is a difficult task and shortening the challenge to be more concise is also a challenging task. Nevertheless, the existing implementations using manually designed features using a wide variety of input features using benchmark datasets are unsatisfactory as they suffer from unknown subject information. It is challenging to judge CNN’s performance using such approaches. The proposed system performs the segmentation through UNet without using a dense layer to perform the segmentation and classification. The proposed system uses the skip connection to hold the loss at the max-pooling layer. Also, the morphological processing and probabilistic classification served as the proposed system’s novelty. The proposed method used three benchmark datasets, MMU, CASIA, and UBIRIS, to experiment with building a training model and tested using various optimization techniques to perform an accurate segmentation. To further test and improve the quality of the proposed method, we experimented with random images. The proposed system’s accuracy is 96% when experimented on random images of subjects collected purely for experimentation. Three optimizers, namely, Stochastic Gradient Descent, RMS Prop, and Adaptive Moment Optimizer, were experimented with in the proposed system to fit the system. The average accuracy we received using optimizers is 71.9, 84.3, and 96.0 for the loss value of 2.36, 2.30, and 1.82, respectively.

Design and Implementation of Artificial Intelligence of Things for Tea (Camellia sinensis L.) Grown in a Plant Factory

An innovative plant growth monitoring and environmental control platform is designed and implemented in this study. In addition to using multi-band artificial light sources for plant growth and development, an artificial intelligence of things (AIoT) approach is also utilised for environmental parameter monitoring, control, and the recording of plant growth traits and diseases. The five LED bands are white (5000 K), cool white (5500 K), blue (peak: 450 nm), red (660 nm), and light red (630 nm). The tea plant (Camellia sinensis f. formosana) is irradiated using lighting-emitting diodes (LED) composed of bands of different wavelengths. In addition, the number of leaves, contour area of the leaves, and leaf colour during the growth period of two varieties of tea plants (Taicha No. 18 and Taicha No. 8) under different irradiation intensities are analysed. Morphological image processing and deep learning models are simultaneously used to obtain plant growth characterization traits and diseases. The effect of the spectral distribution of the light source on the growth response of tea leaves and the effect of disease suppression are not fully understood. This study depicts how light quality affects the lighting formula changes in tea plants under controlled environments. The experimental results show that in three wavelength ranges (360–500 nm, 500–600 nm, and 600–760 nm), the light intensity ratio was 2.5:2.0:5.5 when the illuminance intensity was about 150 µmol∙m−2∙s−1 with a photoperiod of 20:4 (dark); this enabled more leaves, a smaller contour area of the leaves, and a light green colour of the leaves of the tea plant (Taicha No. 18). In addition, during the lighting treatment, when the ratio of the band with an irradiation intensity of 360–500 nm to that with an irradiation intensity of 500–600 nm was 2:1.5, it resulted in a better leaf disease inhibition effect. When the light intensity was increased to more than 400 µmol∙m−2∙s−1, it had little effect on the growth and development of the tea plants and the inhibition of diseases. The results of the study also found that there was a significant difference between the colour of the leaves and the relative chlorophyll content of the tea trees. Finally, the tea plant growth response data obtained from manual records and automatic records are compared and discussed. The accuracy rates of leaf number and disease were 94% and 87%, respectively. Compared with the results of manual measurement and recording, the errors were about 3–15%, which verified the effectiveness and practicability of the proposed solution. The innovative platform provides a data-driven crop modeling application for plant factories.

Ovarian morphology and oocyte development in the seabob shrimp Xiphopenaeus kroyeri: a histological and histochemical approach for the study of penaeid reproduction

Histological and histochemical changes during oocyte development were evaluated in order to establish a model for better understanding nutrient dynamics throughout vitellogenesis and overall penaeid reproduction. Samples of Xiphopenaeus kroyeri females (n = 587) at different stages of gonadal maturation were subjected to histological and histochemical analysis for the identification of neutral polysaccharides, total proteins, and tissue differentiation. We observed that the ovarian structure was composed of collagen fibers and oviducts were lined by three layers of epithelium separated by muscular tissue. Oogenesis was classified into six distinct phases: chromatin nucleolar oocytes, chromatin perinucleolar oocytes, oocytes in primary vitellogenesis, oocytes in secondary vitellogenesis, mature oocytes, and atretic oocytes. Histochemical observations suggested that during the previtellogenesis phase, proteins and neutral polysaccharides were not present. However, a progressive increase in these substances occurred from vitellogenic oocytes until their final maturation stage. Follicular cells composed the follicular complex forming an external epithelium in vitellogenic oocytes, and are probably involved in protein transfer during maturation process. Collagen and glycoprotein vesicles were identified in the cytoplasm of mature oocytes of X. kroyeri. Taken together, the present findings contribute to the current knowledge on the ovarian morphology and maturation process of penaeids, supporting future studies on reproductive biology and breeding performance in captivity.

Compound Adjectives in General and in French Language in Specific

Compounding is an effectiv process of morphology to produce new items in dictionary. In this research we focus on compound adjectives in general and its’ properties in in French language. In this research we easily identify that compound adjectives are very rarely studied compared to compound nouns. This research starts by an introduction about compound adjectives and then the basic items that are used to create compound adjectives are presented, then we foucus on the cases that are no considerd as compound in French language, after that we specify the types of compounding and the pattern that are used to form compound adjectives in French language. At the end we talk about endocentric and exocentric compound adjectives. Finaly, we come to the conclution that this subject is ill-posed and very far from being clearly resolved and thesre are many different approaches to define this subject. For this reason, we tried to present add the different approaches.

Automated Optical Inspection for Defect Identification and Classification in Actual Woven Fabric Production Lines.

This paper presents a turnkey integrated system that can be operated in real time for real textile manufacturers. Eight types of defects in woven fabric, including stain, broken end, broken weft, hole, nep, double pick, kinky weft and float can be recognized and classified. First, an image is captured by a CMOS industrial camera with a pixel size of 4600 × 600 above the batcher at 20 m/min. After that, the four-stage image processing procedure is applied to detect defects and for classification. Stage 1 is image pre-processing; the filtration of the image noise is carried out by a Gaussian filter. The light source is corrected to reduce the uneven brightness resulting from halo formation. The improved mask dodging algorithm is used to reduce the standard deviation of the corrected original image. Afterwards, the background texture is filtered by an averaging filter, and the mean value is corrected for histogram shifting, so that this system is robust to the texture and color changes of woven fabric. The binary segmentation threshold is determined using the mean value and standard deviation of an image with a normal sample. Stage 2 uses adaptive binarization for separation of the background and defects and to filter the noise. In Stage 3, the morphological processing is used before the defect contour is circled, i.e., four features of each block, including the defect area, the aspect ratio of the defect, the average gray level of the defect and the defect orientation, which are calculated according to the range of contour. The image defect recognition dataset consists of 2246 images. The results show that the detection success rate is 96.44%, and the false alarm rate is 3.21%. In Stage 4, the defect classification is implemented. The support vector machine (SVM) is used for classification, 230 defect images are used as training samples, and 206 are used as test samples. The experimental results show that the overall defect recognition rate is 96.60%, providing that the software and hardware equipment designed in this study can implement defect detection and classification for woven fabric effectively.

Scene Text detection and Recognition by Using Multi-Level Features Extractions Based on You Only Once Version Five (YOLOv5) and Maximally Stable Extremal Regions (MSERs) with Optical Character Recognition (OCR)

Textual information within scene images is very important in computer vision applications such asimage retrieval based on its content, Tourist translator, and Navigation systems assistant. This paper presented ascene text recognition system based on YOLO. The main steps of the suggested methodology are: text detectionand localization (using YOLOv5), text segmentation (using Morphological processing as a new method), featuresextraction (using MSERs), character segmentation and word segmentation (using bounding boxes with graph), andfinally character recognition (using OCR). In this work we create a new dataset model that includes most of thetext challenges such as Font type, Font size, Font color, and Font. The proposed system gives higher performancefor detection, localization, and recognition when using dataset containing many challenges, the results were 80%,96%, and 87.6 for precision, recall, and F-score respectively. Comparing with other similar works it was better.The accuracy of (OCR) is more than 99%.

Combined effects of microstructural characteristics on anisotropic transport properties of gas diffusion layers for PEMFCs

Gas diffusion layer (GDL) plays a key role in proton exchange membrane fuel cells, which provides multi-functions for gas transport, thermal-electrical conduction and mechanical support. Coupling manipulation of different microstructural characteristics could potentially improve transport properties of GDLs. This work proposes an approach to reconstruct heterogenous GDLs and conduct pore-scale modeling to evaluate the anisotropic transport properties. The models are reconstructed using X-ray computed tomography, stochastically reconstruction methods and morphological processing techniques, which consider different fiber diameter, GDL thickness and local porosity distribution type. Combined effects of microstructure characteristics on tortuosity, diffusivity, thermal-electrical conductivity and anisotropic ratios are investigated comprehensively. The results show that the diffusivity with fiber diameter of 7 μm is approximately 7% lower, and the conductivity is 8% higher than that of 9 μm. The anisotropic ratios of diffusivity, thermal conductivity and electrical conductivity range from 1.25 to 1.65, 5 to 20, and 20 to 55, respectively. Local porosity distribution of uniform-fluctuated type, fiber diameter of 7 μm and GDL thickness of 126 μm are suggested to balance diffusivity and thermal-electrical conductivity simultaneously. The methods and results can guide microstructure design of other porous electrodes with higher performance.

Identification of the BolA Protein Reveals a Novel Virulence Factor in K. pneumoniae That Contributes to Survival in Host.

ABSTRACTBolA has been characterized as an important transcriptional regulator, which is induced in the stationary phase of growth and is often associated with bacterial virulence. This study was initiated to elucidate the role of the BolA in the virulence of K. pneumoniae. Using a mouse infection model, we revealed bolA mutant strain yielded significantly decreased bacterial loads in the liver, spleen, lung, and kidney, and failed to form liver abscesses. Gene deletion demonstrated that the bolA was required for siderophore production, biofilm formation, and adhesion to human colon cancer epithelial cells HCT116. Quantitative reverse transcriptase PCR (RT-qPCR) indicated that BolA could impact the expression of pulK, pulF, pulE, clpV, vgrG, entE, relA, and spoT genes on a genome-wide scale, which are related to type II secretion system (T2SS), type VI secretion system (T6SS), guanosine tetraphosphate (ppGpp), and siderophore synthesis and contribute to fitness in the host. Furthermore, the metabolome analysis showed that the deletion of the bolA gene led to decreased pools of five metabolites: biotin, spermine, cadaverine, guanosine, and flavin adenine dinucleotide, all of which are involved in pathways related to virulence and stress resistance. Taken together, we provided evidence that BolA was a significant virulence factor in the ability of K. pneumoniae to survive, and this was an important step in progress to an understanding of the pathways underlying bacterial virulence.IMPORTANCE BolA has been characterized as an important transcriptional regulator, which is induced in the stationary phase of growth and affects different pathways directly associated with bacterial virulence. Here, we unraveled the role of BolA in several phenotypes associated with the process of cell morphology, siderophore production, biofilm formation, cell adhesion, tissue colonization, and liver abscess. We also uncovered the importance of BolA for the success of K. pneumoniae infection and provided new clues to the pathogenesis strategies of this organism. This work constitutes a relevant step toward an understanding of the role of BolA protein as a master regulator and virulence factor. Therefore, this study is of great importance for understanding the pathways underlying K. pneumoniae virulence and may contribute to public health care applications.

Self-crystallization behavior of paraffin and the mechanism study of SiO2 nanoparticles affecting paraffin crystallization

The crystallization behavior of paraffin severely inhibits the extraction and transportation of waxy crude oil. Therefore, in this study, the self-crystallization behavior of paraffin and the mechanism of SiO2 nanoparticles (NPs) affecting paraffin crystallization were investigated microscopically through molecular dynamics (MD) simulations. By analyzing the crystallization morphology and crystallization process of paraffin at different simulated temperatures, we observed the difference in the crystallization of long-chain paraffin molecules and short-chain paraffin molecules from a molecular point of view. Three processes of nucleation, aggregation and crystallization of paraffin crystals were clearly observed. Meanwhile, compared with the light oil system, in waxy heavy oil, a small amount of asphaltene can slightly inhibit the molecular crystallization of paraffin. However, the existence of a large amount of asphaltene changes the structure of paraffin crystals, which makes the paraffin crystals with asphaltene as heterogeneous core intertwined, showing a more complex network aggregation structure. Most importantly, the addition of NPs can significantly improve the diffusion coefficient of paraffin. From the microscopic point of view, the co-crystallization behavior of the grafted carbon chains on the surface of the grafted nanoparticles and the paraffin was observed, which further enhanced the nucleation of the nanoparticles and reduced the crystal size.

Automatic morpheme identification across development: Magnetoencephalography (MEG) evidence from fast periodic visual stimulation.

The present study combined magnetoencephalography (MEG) recordings with fast periodic visual stimulation (FPVS) to investigate automatic neural responses to morphemes in developing and skilled readers. Native English-speaking children (N = 17, grade 5–6) and adults (N = 28) were presented with rapid streams of base stimuli (6 Hz) interleaved periodically with oddballs (i.e., every fifth item, oddball stimulation frequency: 1.2 Hz). In a manipulation-check condition, tapping into word recognition, oddballs featured familiar words (e.g., roll) embedded in a stream of consonant strings (e.g., ktlq). In the experimental conditions, the contrast between oddball and base stimuli was manipulated in order to probe selective stem and suffix identification in morphologically structured pseudowords (e.g., stem + suffix pseudowords such as softity embedded in nonstem + suffix pseudowords such as trumess). Neural responses at the oddball frequency and harmonics were analyzed at the sensor level using non-parametric cluster-based permutation tests. As expected, results in the manipulation-check condition revealed a word-selective response reflected by a predominantly left-lateralized cluster that emerged over temporal, parietal, and occipital sensors in both children and adults. However, across the experimental conditions, results yielded a differential pattern of oddball responses in developing and skilled readers. Children displayed a significant response that emerged in a mostly central occipital cluster for the condition tracking stem identification in the presence of suffixes (e.g., softity vs. trumess). In contrast, adult participants showed a significant response that emerged in a cluster located in central and left occipital sensors for the condition tracking suffix identification in the presence of stems (e.g., softity vs. stopust). The present results suggest that while the morpheme identification system in Grade 5–6 children is not yet adult-like, it is sufficiently mature to automatically analyze the morphemic structure of novel letter strings. These findings are discussed in the context of theoretical accounts of morphological processing across reading development.

MEA-Net: A Lightweight SAR Ship Detection Model for Imbalanced Datasets

The existing synthetic aperture radar (SAR) ship datasets have an imbalanced number of inshore and offshore ship targets, and the number of small, medium and large ship targets differs greatly. At the same time, the existing SAR ship detection models in the application have a huge structure and require high computing resources. To solve these problems, we propose a SAR ship detection model named mask efficient adaptive network (MEA-Net), which is lightweight and high-accuracy for imbalanced datasets. Specifically, we propose the following three innovative modules. Firstly, we propose a mask data balance augmentation (MDBA) method, which solves the imbalance of sample data between inshore and offshore ship targets by combining mathematical morphological processing and ship label data to greatly improve the ability of the model to detect inshore ship targets. Secondly, we propose an efficient attention mechanism (EAM), which effectively integrates channel features and spatial features through one-dimensional convolution and two-dimensional convolution, to improve the feature extraction ability of the model for SAR ship targets. Thirdly, we propose an adaptive receptive field block (ARFB), which can achieve more effective multi-scale detection by establishing the mapping relationship between the size of the convolution kernel and the channel of feature map, to improve the detection ability of the model for ship targets of different sizes. Finally, MEA-Net is deployed on the Jeston Nano edge computing device of the 2 GB version. We conducted experimental validation on the SSDD and HRSID datasets. Compared with the baseline, the AP of MEA-Net increased by 2.18% on the SSDD dataset and 3.64% on the HRSID dataset. The FLOPs and model parameters of MEA-Net were only 2.80 G and 0.96 M, respectively. In addition, the FPS reached 6.31 on the Jeston Nano, which has broad application prospects.