CS-SegFormer: a dual-path attention enhanced segmentation network for metallographic grain size evaluation

As a detection tool to identify metal or alloy, metallographic quantitative analysis has received increasing attention for its ability to evaluate quality control and reveal mechanical properties. The detection procedure is mainly operated manually to locate and characterize the constitution in metallographic images. The automatic detection is still a challenge even with the emergence of several excellent models. Benefiting from the development of deep learning, with regard to two different metallurgical structural steel image datasets, we propose two attention-aware deep neural networks, Modified Attention U-Net (MAUNet) and Self-adaptive Attention-aware Soft Anchor-Point Detector (SASAPD), to identify structures and evaluate their performance. Specifically, in the case of analyzing single-phase metallographic image, MAUNet investigates the difference between low-frequency and high-frequency and prevents duplication of low-resolution information in skip connection used in an U-Net like structure, and incorporates spatial-channel attention module with the decoder to enhance interpretability of features. In the case of analyzing multi-phase metallographic image, SASAPD explores and ranks the importance of anchor points, forming soft-weighted samples in subsequent loss design, and self-adaptively evaluates the contributions of attention-aware pyramid features to assist in detecting elements in different sizes. Extensive experiments on the above two datasets demonstrate the superiority and effectiveness of our two deep neural networks compared to state-of-the-art models on different metrics.

The refined segmentation of nuclei and the cytoplasm is the most challenging task in the automation of cervical cell screening. The U-Shape network structure has demonstrated great superiority in the field of biomedical imaging. However, the classical U-Net network cannot effectively utilize mixed domain information and contextual information, and fails to achieve satisfactory results in this task. To address the above problems, a module based on global dependency and local attention (GDLA) for contextual information modeling and features refinement, is proposed in this study. It consists of three components computed in parallel, which are the global dependency module, the spatial attention module, and the channel attention module. The global dependency module models global contextual information to capture a priori knowledge of cervical cells, such as the positional dependence of the nuclei and cytoplasm, and the closure and uniqueness of the nuclei. The spatial attention module combines contextual information to extract cell boundary information and refine target boundaries. The channel and spatial attention modules are used to provide adaption of the input information, and make it easy to identify subtle but dominant differences of similar objects. Comparative and ablation experiments are conducted on the Herlev dataset, and the experimental results demonstrate the effectiveness of the proposed method, which surpasses the most popular existing channel attention, hybrid attention, and context networks in terms of the nuclei and cytoplasm segmentation metrics, achieving better segmentation performance than most previous advanced methods.

Marine raft aquaculture (MFA) plays an important role in the marine economy and ecosystem. With the characteristics of covering a large area and being sparsely distributed in sea area, MFA monitoring suffers from the low efficiency of field survey and poor data of optical satellite imagery. Synthetic aperture radar (SAR) satellite imagery is currently considered to be an effective data source, while the state-of-the-art methods require manual parameter tuning under the guidance of professional experience. To preclude the limitation, this paper proposes a segmentation network combined with nonsubsampled contourlet transform (NSCT) to extract MFA areas using Sentinel-1 images. The proposed method is highlighted by several improvements based on the feature analysis of MFA. First, the NSCT was applied to enhance the contour and orientation features. Second, multiscale and asymmetric convolutions were introduced to fit the multisize and strip-like features more effectively. Third, both channel and spatial attention modules were adopted in the network architecture to overcome the problems of boundary fuzziness and area incompleteness. Experiments showed that the method can effectively extract marine raft culture areas. Although further research is needed to overcome the problem of interference caused by excessive waves, this paper provides a promising approach for periodical monitoring MFA in a large area with high efficiency and acceptable accuracy.

Breast UltraSound (BUS) imaging is a commonly used diagnostic tool in the field of counter fighting breast diseases, especially for early detection and diagnosis of breast cancer. Due to the inherent characteristics of ultrasound images such as blurry boundaries and diverse tumor morphologies, it is challenging for doctors to manually segment breast tumors. In recent years, the Convolutional Neural Network (CNN) technology has been widely applied to automatically segment BUS images. However, due to the inherent limitations of CNNs in capturing global contextual information, it is difficult to capture the full context. To address this issue, the paper proposes a novel BGRD-TransUNet model for breast lesion segmentation, based on TransUNet. The proposed model, first, replaces the original ResNet50 backbone network of TransUNet with DenseNet121 for initial feature extraction. Next, newly designed Residual Multi-Scale Feature Modules (RMSFMs) are employed to extract features from various layers of DenseNet121, thus capturing richer features within specific layers. Thirdly, a Boundary Guidance (BG) network is added to enhance the contour information of BUS images. Additionally, newly designed Boundary Attentional Feature Fusion Modules (BAFFMs) are used to integrate edge information and features extracted through RMSFMs. Finally, newly designed Parallel Channel and Spatial Attention Modules (PCSAMs) are used to refine feature extraction using channel and spatial attention. An extensive experimental testing performed on two public datasets demonstrates that the proposed BGRD-TransUNet model outperforms all state-of-the-art medical image segmentation models, participating in the experiments, according to all evaluation metrics used (except for few separate cases), including the two most important and widely used metrics in the field of medical image segmentation, namely the Intersection over Union (IoU) and Dice Similarity Coefficient (DSC). More specifically, on the BUSI dataset and dataset B, BGRD-TransUNet achieves IoU values of 76.77% and 86.61%, and DSC values of 85.08% and 92.47%, respectively, which are higher by 7.27 and 3.64, and 5.81 and 2.54 percentage points, than the corresponding values achieved by the baseline (TransUNet).

Rain, snow, fog, and dust can degrade road-scene images, blur fine details, and consequently reduce the reliability of perception systems for autonomous driving. To address this problem, this paper proposes PF-ConvNeXt, an adverse weather recognition model built upon the ConvNeXt architecture. First, a lightweight pyramid split attention (PSA) module is introduced to enable multi-scale feature fusion, so that both global degradation patterns and local texture details can be captured simultaneously. Second, a feature enhancement channel and spatial attention module (FECS) is designed. It adaptively recalibrates features along the channel and spatial paths, thereby suppressing interference from complex backgrounds and noise. Third, during training, Focal Loss is adopted to strengthen learning for hard samples and minority weather categories, alleviating recognition bias caused by class imbalance. Experiments are conducted on a dataset of 5000 images constructed by integrating RTTS, DAWN, and a self-collected rainy-weather dataset. The results show that PF-ConvNeXt achieves 90.16% accuracy, 95.24% mean average precision, and a 92.18% F1-score. It outperforms the ConvNeXt baseline by 4.74%, 5.46%, and 5.95%, respectively, and surpasses multiple mainstream classification models. This study provides an effective recognition framework for robust environmental perception under challenging weather conditions and demonstrates promising potential for practical deployment.

As sensor technology advances, the variety and number of sensors increase, leading to the capture of more signals. Existing multi-source fusion methods often face issues such as increased model complexity or the failure to fully utilize the potential correlations among multi-sensor data, thereby affecting the accuracy and reliability of fault diagnosis. To address this issue, this paper proposes a multi-source fusion convolutional neural network (MFCNN) that diagnoses bearing faults by integrating features from multi-source signals. Firstly, multiple convolution blocks with gradually increasing one-dimensional kernel sizes are utilized to extract features from the integrated multi-source data. This approach enhances feature extraction efficiency and simplifies the network architecture. Secondly, a feature fusion based on the convolutional block attention module attention mechanism is proposed, which refines feature representation through channel and spatial attention modules. This makes the model more focused on important information, thereby improving recognition accuracy. The diagnostic capabilities of the proposed MFCNN are evaluated utilizing two datasets.

Moderate resolution imaging spectrometer (MODIS) images are widely used in land, ocean, and atmospheric monitoring, due to their wide spectral coverage, high temporal resolution, and convenient data acquisition. Accurate cloud detection is critical to the fine processing and application of MODIS images. Owing to spatial resolution limitations and the influence of mixed pixels, most MODIS cloud detection algorithms struggle to effectively recognize of clouds and ground objects. Here, we propose a novel cloud detection method based on land surface reflectance and a multiscale feature convolutional neural network to achieve high-precision cloud detection, particularly for thin clouds and clouds over bright surface. A monthly surface reflectance dataset was constructed by MODIS products (MOD09A1) and employed to provide background information for cloud detection. Difference-based samples were obtained using surface reflectance as well MODIS images of different phases based on difference operations. The multiscale feature network (MFCD-Net) using an atrous spatial pyramid pooling and a channel and spatial attention module integrated low-level spatial features and high-level semantic information to capture multiscale features and generate a high-precision cloud mask. For cloud detection experiments and quantitative analysis, 61 MODIS images acquired at different times on various underlying surface types were used. Cloud detection results were compared to those of UNet, Deeplabv3+, UNet++, PSPNet, and top of atmosphere-based (MFCD-TOA) methods. The proposed method performed well, with the highest overall accuracy (96.55%), precision (92.13%), and recall (88.90%). It improved cloud detection accuracy in various scenarios, reducing thin cloud omission and bright surface misidentification.

How to properly involve text characteristics like multi-scale, arbitrary direction, length aspect ratio, into detection network design has become a hot topic in computer vision. Feature Pyramid Network (FPN) is a typical method to achieve robust text detection, where its low-level and high-level feature map retains spatial structure and global semantic information, respectively. However, its strict hierarchical structure fails to fuse low-level and high-level information to improve the distinguish ability of feature map. To address this problem, we propose a novel feature fusion pyramid network for end-to-end scene text detection by fusing multi-modal information. By dividing pyramid structure into high-level and low-level layers, channel and spatial attention modules are adopted to enhance high-level and low-level feature representation by encoding channel and spatial-wise context information, respectively. In order to reduce information loss by layer transmission, a special residual network is designed to achieve short-cut between high-level and low-level features, so as to realize multi-modal feature fusion. Experiments show the precision and recall of the proposed method on ICDAR2015, ICDAR2017-MLT, and MSRA-TD500 datasets reach 88.7%/82.1%, 77.0%/60.3%, and 85.3%/74.8%, respectively.

Proceedings of the international conference on advanced infrared detectors and systems: (Institution of Electrical Engineers, London, 1981) ISBN 0 85296250 9, ISSN 0537-9989

Accurate segmentation of key anatomical structures in pediatric echocardiography is essential for the diagnosis and treatment of congenital heart disease. However, most of the existing segmentation methods for echocardiography have the problem of loss of detailed information, which has a certain impact on the accuracy of segmentation. Based on this, we propose a multi-directional attention (MDA) network for echocardiographic segmentation. This method uses U-Net as the backbone network to extract the initial features of different layers, and then sends the initial features to our proposed MDA module for feature enhancement. Among them, MDA includes two parts: First, considering the different contribution rates of spatial information in different directions to features, we construct a multi-directional spatial attention (MDSA) module to extract spatial information in different directions. Then to avoid the loss of channel information, we construct a channel weight constraint module (CWC) to constrain the weight of the spatial features extracted by MDSA. Finally, the group fusion feature output by MDA is used as the input of the decoder, and the final segmentation prediction result is obtained by setting the layered feature fusion (LFF) module. We conduct an extensive evaluation of 4,485 two-dimensional (2D) pediatric echocardiograms from 127 echocardiographic videos. Experiments show that the proposed algorithm can achieve the results of pediatric echocardiographic anatomical structures (left ventricle (LV), left atrium (LA)) with the average dice, precision, and recall were 0.9346, 0.9370, and 0.9406.

Point cloud segmentation is essential for scene understanding, which provides advanced information for many applications, such as autonomous driving, robots, and virtual reality. To improve the accuracy and robustness of point cloud segmentation, many researchers have attempted to fuze camera images to complement the color and texture information. The common fusion strategy is the combination of convolutional operations with concatenation, element-wise addition or element-wise multiplication. However, conventional convolutional operators tend to confine the fusion of modal features within their receptive fields, which can be incomplete and limited. In addition, the inability of encoder–decoder segmentation networks to explicitly perceive segmentation boundary information results in semantic ambiguity and classification errors at object edges. These errors are further amplified in point cloud segmentation tasks, significantly affecting the accuracy of point cloud segmentation. To address the above issues, we propose a novel self-attention multi-modal fusion semantic segmentation network for point cloud semantic segmentation. Firstly, to effectively fuze different modal features, we propose a self-cross fusion module (SCF), which models long-range modality dependencies and transfers complementary image information to the point cloud to fully leverage the modality-specific advantages. Secondly, we design the salience refinement module (SR), which calculates the importance of channels in the feature maps and global descriptors to enhance the representation capability of salient modal features. Finally, we propose the local-aware anisotropy loss measure the element-level importance in the data and explicitly provide boundary information for the model, which alleviates the inherent semantic ambiguity problem in segmentation networks. Extensive experiments on two benchmark datasets demonstrate that our proposed method surpasses current state-of-the-art methods.

The durability study of thermochromic vanadium dioxide films with the addition of barrier coatings

Even if many steels and alloys have been welded on the last years, nowadays there are some other stainless steel alloys that need a further comprehension when they have to be welded. Typically these alloys are martensitic and precipitation hardening ones that still present some problems to be weld, i.e. hot cracks, fragile beads, an excessive grain size and other surface defects. In this work some martensitic stainless steels of which a AISI 420B, a AISI 440C and a AISI 630 have been studied. The last one is always with a martensitic structure but, in particular, some interesting mechanical properties are reached by a precipitation hardening process. This research has experimented and studied the mechanical and technological properties of the welds obtained on the above cited AISI 420B, AISI 440C and AISI 630, welded by 1.5 kW CO2 laser. The results have also been compared with the ones obtained on ferritic stainless steels AISI 430 and 430F. A technological characterization of the welds has followed as metallographic tests and evaluations, microhardness, tensile and fatigue tests.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Chapter 8 - Miscellaneous industrial applications and products

The microstructure of arc sprayed stainless steel 316L coatings appears mainly in bright white matrix, deteriorated layers (grey), and black pores under optical microscopy. The black pores and the chromium-depleted areas in the deteriorated layers are known as the factors for decreasing the ability of protecting substrate under corrosive environments. Results of experiments in this paper suggests, in the condition of this study besides the factors mentioned above, Fe-Cr oxides should be another factor of dominating the corrosion resistance in the coatings. It also describes that the quantity and the distributions of such oxides are great influence on the corrosion behaviors. In this study, two kinds of coatings were used, one with thick deteriorated layers and another with thin deteriorated layers, which were sprayed on mild steel substrate by air atomization and nitrogen atomization respectively. Salt spray test and salt-water dip test were carried out to investigate corrosion behavior in macro and micro view. An effect of sealing treatment on the performance of the coatings was also examined. Results of metallographic examination and image processing analysis are well supported by a detailed investigation of corrosion behaviors of individual phases.