Search

Medieval paper, a handmade product, is made with a mould which leaves an indelible imprint on the sheet of paper. This imprint includes chain lines, laid lines and watermarks which are often visible on the sheet. Extracting these features allows the identification of the paper stock and gives information about the chronology, localisation and movement of manuscripts and people. Most computational work for feature extraction of paper analysis has so far focused on radiography or transmitted light images. While these imaging methods provide clear visualisation of the features of interest, they are expensive and time consuming in their acquisition and not feasible for smaller institutions. However, reflected light images of medieval paper manuscripts are abundant and possibly cheaper in their acquisition. In this paper, we propose algorithms to detect and extract the laid and chain lines from reflected light images. We tackle the main drawback of reflected light images, that is, the low contrast attenuation of chain and laid lines and intensity jumps due to noise and degradation, by employing the spectral total variation decomposition and develop methods for subsequent chain and laid line extraction. Our results clearly demonstrate the feasibility of using reflected light images in paper analysis. This work enables feature extraction for paper manuscripts that have otherwise not been analysed due to a lack of appropriate images. We also open the door for paper stock identification at scale. [ABSTRACT FROM AUTHOR]

During the past decades, several academic paper recommendation systems were introduced in literature aiming to assist users in finding relevant papers close to their needs. In particular, those models have used users' past preferences combined with papers' side information to personalize their recommendations. Unfortunately, the majority of those models fail to utilize the implicit relationship between the crucial elements along with the semantic relations of the nodes in a Heterogeneous Information Network (HIN), which can further improve the accuracy of the models. In this paper, we propose a research article recommendation model that aims to tackle those issues by exploiting both textual and graph representations, simultaneously. The model employs SPECTER document embedding to learn context-preserving research article representations. In particular, it learns the features from a HIN, which is a knowledge graph of the node entities and their relationships. Then, an attention module strengthens the semantic feature extraction process and learns enhanced representations. The combined semantic and structural features are then provided as input to a Deep Neural Network (DNN) in order to learn high-level representations of query and candidate papers. We have evaluated our model against state-of-the-art models over two popular datasets. The results indicate a significant improvement in terms of MAP, recall, and MRR. [ABSTRACT FROM AUTHOR]

Nondestructive evaluation of aging state of oil–paper insulation is an important means to prevent aging failure. A characteristic model based on terahertz-sensing hydrogen bond content is proposed in this study to evaluate the aging state of oil-impregnated pressboard nondestructively by analyzing its terahertz absorption spectrum. Oil–paper samples at different aging stages were obtained using accelerated thermal aging in the laboratory, and the terahertz spectrum collection range of oil-impregnated pressboard was 0.3–2.0 THz. Spectral characteristic peaks were extracted and the relationship between peak area and degree of polymerization of oil-impregnated pressboard was investigated on the basis of quantum chemical simulation analysis and Gaussian function fitting of experimental spectral features. Finally, the mechanism of characteristic peak was verified by combining infrared spectroscopy. Results showed that the 1 THz characteristic peak is closely related to hydroxyl and hydrogen bond contents in cellulose, its area presents satisfactory semilogarithmic linear function relationship with the degree of polymerization of insulating paper, and the maximum Pierce correlation coefficient is 0.95. The prediction results of aging samples with different oil–paper ratios further verify the correctness of this relationship, and the prediction error of the degree of polymerization value of insulating paper is within ± 20%. The results of this study can provide new insights into the nondestructive evaluation of the aging state of oil–paper insulation using terahertz technology. [ABSTRACT FROM AUTHOR]

Recommender systems have been increasingly popular in entertainment and consumption, and they are evident in academics, especially for applications that suggest submitting scientific articles to scientists. However, due to the various acceptance rates, impact factors, and rankings in different publishers, searching for a proper venue or journal to submit a scientific work usually takes a lot of time and effort. In this paper, we aim to present two new approaches extended from our paper (Huynh et al. 2021) presented at IAE/AIE 2021. In the first approach, we employ RNN structures besides using Conv1D. In the second approach, we introduce a new method, using DistillBert for two cases of uppercase and lowercase words. It can help vectorize features (such as Title, Abstract, and Keywords) and then use Conv1d to perform feature extraction. Furthermore, we propose a new calculation method for similarity score for Aim & Scope with other features (we name this approach is DistilBertAims). It can help keep the weights of similarity score calculation continuously updated and then continue to fit more data. The experimental results show that the second approach could obtain a better performance, which are 62.46%, 90.32%, 94.89%, 97.96% while the best performance in previous studies barely gained 50.02%, 78.89%, 86.27%, 93.23% in terms of Top K Accuracy (K = 1, 3, 5, 10). Interestingly, our best approach in this paper is higher than 12.44% the best of the previous study (Huynh et al. 2021) in terms of the Top 1 accuracy, which was presented in the conference IEA/AIE 2021. [ABSTRACT FROM AUTHOR]

Single-frame infrared small target detection is affected by the low image resolution and small target size, and is prone to the problems of small target feature loss and positional offset during continuous downsampling; at the same time, the sparse features of the small targets do not correlate well with the global-local linkage of the background features. To solve the above problems, this paper proposes an efficient infrared small target detection method. First, this paper incorporates BlurPool in the feature extraction part, which reduces the loss and positional offset of small target features in the process of convolution and pooling. Second, this paper designs an interactive attention deep feature fusion module, which acquires the correlation information between the target and the background from a global perspective, and designs a compression mechanism based on deep a priori knowledge, which reduces the computational difficulty of the self-attention mechanism. Then, this paper designs the context local feature enhancement and fusion module, which uses deep semantic features to dynamically guide shallow local features to realize enhancement and fusion. Finally, this paper proposes an edge feature extraction module for shallow features, which utilizes the complete texture and location information in the shallow features to assist the network to initially locate the target position and edge shape. Numerous experiments show that the method in this paper significantly improves nIoU, F1-Measure and AUC on IRSTD-1k Datasets and NUAA-SIRST Datasets. [ABSTRACT FROM AUTHOR]

Around a third of the total population of Europe suffers from mental disorders. The use of electroencephalography (EEG) together with Machine Learning (ML) algorithms to diagnose mental disorders has recently been shown to be a prominent research area, as exposed by several reviews focused on the field. Nevertheless, previous to the application of ML algorithms, EEG data should be correctly preprocessed and prepared via Feature Engineering (FE). In fact, the choice of FE techniques can make the difference between an unusable ML model and a simple, effective model. In other words, it can be said that FE is crucial, especially when using complex, non-stationary data such as EEG. To this aim, in this paper we present a Systematic Mapping Study (SMS) focused on FE from EEG data used to identify mental disorders. Our SMS covers more than 900 papers, making it one of the most comprehensive to date, to the best of our knowledge. We gathered the mental disorder addressed, all the FE techniques used, and the Artificial Intelligence (AI) algorithm applied for classification from each paper. Our main contributions are: (i) we offer a starting point for new researchers on these topics, (ii) we extract the most used FE techniques to classify mental disorders, (iii) we show several graphical distributions of all used techniques, and (iv) we provide critical conclusions for detecting mental disorders. To provide a better overview of existing techniques, the FE process is divided into three parts: (i) signal transformation, (ii) feature extraction, and (iii) feature selection. Moreover, we classify and analyze the distribution of existing papers according to the mental disorder they treat, the FE processes used, and the ML techniques applied. As a result, we provide a valuable reference for the scientific community to identify which techniques have been proven and tested and where the gaps are located in the current state of the art. [ABSTRACT FROM AUTHOR]

Due to the large intra-class differences between the same categories and the scale imbalance between different categories in the remote sensing image dataset, the semantic segmentation task presents the problem of small-scale object information loss, the imbalance between foreground and background, and simultaneously the background dominates, which seriously affects the performance of the network model. To solve the above problems, this paper proposes an efficient bilateral branch depth neural network model based on the U-Net depth neural network, named BBU-Net. Firstly, one branch of the network learns the distribution characteristics of the original data, and the other focuses on difficult samples. Then the two branches improve the representation and classification ability of the neural network by accumulating learning strategies. Finally, considering the geometric diversity of remote sensing images, this paper adopts test time augmentation and reflection padding strategies and proposes a balanced weighted loss function named CombineLoss to alleviate the imbalance in the training process. The depth neural network proposed in this paper was first tested on the Inria Aerial Image Labeling Dataset, and 87.53% of mean intersection over union and 97.4% of mean pixel accuracy were obtained, respectively. At the same time, to verify the model's complexity, the model proposed in this paper is compared with the neural network based on integrated learning. The comparison results show that the spatial complexity of the network proposed in this paper is much lower than the neural network obtained by integrated learning, and the parameters are also much smaller than the neural network based on integrated learning. Then use the satellite building dataset I in the WHU Building Dataset and mainstream semantic segmentation methods for multiple groups of comparative experiments. The experimental results show that the method proposed in this paper can effectively extract the semantic information of remote sensing images, significantly improve the imbalance of remote sensing image data, improve the performance of the network model, and achieve a good semantic segmentation effect, which fully proves the effectiveness of this method. [ABSTRACT FROM AUTHOR]

Logistics parcel detection technology is critical for unmanned sorting. YOLOv5, the state-of-the-art (SOTA) object detection model, is a classic network widely used in engineering. However, on the premise of fast and accurate detection of logistics parcels, it confronts certain challenges of high computing load and parameter quantity. To address these issues, this paper proposes a two-scale lightweight deep learning model named SFYOLOv5. In this article, a lightweight feature extraction module called Pruned-Shuffle-Block (PSB) is proposed. Meanwhile, a double-layer pyramid structure for the medium and large target detection is created in accordance with the image size distribution of logistics parcels. With this structure, the Floating Point of Operations (FLOPs) and parameters in feature extraction were significantly reduced. In addition, a down sampling module named Focus for Downsampling (FFD) is designed and attention modules are introduced to extract high-level semantic information in logistics parcels. These modules not only compensate for the loss caused by down sampling but also improve the mean Average Precision (mAP). Finally, the comparison experiment is performed by using the self-built logistics parcel dataset. The results show that the mAP of the model reaches 99.1%, the number of model parameters decreased by 92.14%, and the FLOPs decreased by 89.57% compared with the existing SOTA model. This model can be used in logistics parcel intelligent sorting. [ABSTRACT FROM AUTHOR]

Freehand sketching on a paper is commonly found to be used as it is a natural and intuitive communication channel for portraying ideas. Additional strokes are often drawn over the existing ones to make the sketch clearer. For its usefulness in subsequent steps, a sketch is transformed to be a 3D CAD model that is typically performed manually by a designer due to a lack of technology support. Geometrical reconstruction from an offline scanning of a sketch has been a challenge because enriched information obtained during sketching is reduced to a batch of data points. This paper presents an algorithm for identifying a single-line drawing from a paper-based overtraced sketch as it is the first important step towards geometrical reconstruction. Key activities in this approach are thick-line sketch creation, contour boundary extraction, segments identification, and line drawing creation. The algorithm has been successfully implemented on LabVIEW and tested with several samples. [ABSTRACT FROM AUTHOR]

Thermal infrared image colorization is very difficult, and colorized images suffer from poor texture detail recovery and low color matching. To solve the above problems, this paper proposes an Efficient and Effective Generative Adversarial Network (E2GAN). This paper proposes multi-level dense module, feature fusion module, and color-aware attention module in the improved generator. Adding multi-level dense module can enhance the feature extraction capability and the improve detail recovery capability Using the feature fusion module in the middle of the encoder–decoder reduces the information loss caused by encoder down-sampling and improves the prediction of fine color of the image. Using the color-aware attention module during up-sampling allows for capturing more semantic details, focusing on more key objects, and generating high-quality colorized images. And the proposed discriminator is the PatchGAN with color-aware attention module, which enhances its ability to discriminate between true and false colorized images. Meanwhile, this paper proposes a novel composite loss function that can improve the quality of colorized images, generate fine local details, and recover semantic and texture information. Extensive experiments demonstrate that the proposed E2GAN has significantly improved SSIM, PSNR, LPIPS, and NIQE on the KAIST dataset and the FLIR dataset compared to existing methods. [ABSTRACT FROM AUTHOR]

The digital repository of research articles is increasing at a rapid rate and hence searching the right paper becoming a tedious task for researchers. A research paper recommendation system is advocated to help researchers in this context. In the process of designing such a system, proper representation of articles, more specifically, feature identification and extraction are two essential tasks. The existing approaches mainly consider direct features which are readily available from research articles. However, there are certain features which are not readily available from a paper, but may greatly influence the performance of recommendation systems. This paper proposes four indirect features: keyword diversification, text complexity, citation analysis over time, and scientific quality measurement to represent a research article. The keyword diversification measures the uniqueness of the keywords of a paper which helps variation in recommendation. The text complexity measurement helps to provide a paper by matching the user's understandability level. The citation analysis over time decides the relevancy of a paper. The scientific quality measurement helps to measure the scientific values of papers. Formal definitions of the proposed indirect features, schemes to extract the feature values given a research article, and metrics to measure them quantitatively are discussed in this paper. To substantiate the efficacy of the proposed features, a number of experiments have been carried out. The experimental results reveal that the proposed indirect features uniquely define a research article than the direct features. Given a research paper, extraction of feature vector is computationally fast and thus feasible to filter a large corpus of papers in real time. More significantly, indirect features are matchable with user's profile features, thus satisfying an important criterion in collaborative filtering. [ABSTRACT FROM AUTHOR]

Automatic Speech Emotion Recognition (ASER) has recently garnered attention across various fields including artificial intelligence, pattern recognition, and human–computer interaction. However, ASER encounters numerous challenges such as a shortage of diverse datasets, appropriate feature selection, and suitable intelligent recognition techniques. To address these challenges, a systematic literature review (SLR) was conducted following established guidelines. A total of 60 primary research papers spanning from 2011 to 2023 were reviewed to investigate, interpret, and analyze the related literature by addressing five key research questions. Despite being an emerging area with applications in real-life scenarios, ASER still grapples with limitations in existing techniques. This SLR provides a comprehensive overview of existing techniques, datasets, and feature extraction tools in the ASER domain, shedding light on the weaknesses of current research studies. Additionally, it outlines a list of limitations for consideration in future work. [ABSTRACT FROM AUTHOR]

The complex underwater environment, such as foreign object occlusion and dim light, causes the feature of underwater objects to be seriously missing. And ripple causes deformation of objects, which greatly increases the difficulty of feature extraction. Existing object recognition models cannot accurately recognize obscured objects due to incomplete features of underwater objects. To solve the above problems, this paper proposes an underwater occlusion object recognition method based on two-stage image reconstruction strategy. Firstly, the salient feature extraction network and the relevant environment feature extraction network are constructed to extract the salient feature and the relevant environment feature respectively. Secondly, the two-stage image reconstruction model with gradient penalty constraints is constructed to obtain finely reconstructed images. Finally, the object recognition with feature adaptive boundary regression is constructed to realize the recognition of finely reconstructed images. To prove the effectiveness of the proposed algorithm, it is compared with the existing object recognition model in datasets with different levels of complexity. The average recognition accuracy of the proposed model is 78.36%, and the recognition rate is improved by 14.16% compared to the original image. Experiments show that the object recognition algorithm proposed in this paper is effective and superior to the existing algorithms. [ABSTRACT FROM AUTHOR]

In this paper, a method based on transfer learning is proposed to recover the three-dimensional shape of cultural relics faces from a single old photo. It can simultaneously reconstruct the three-dimensional facial structure and align the texture of the cultural relics with fewer training samples. The UV position map is used to represent the three-dimensional shape in space and act as the output of the network. A convolutional neural network is used to reconstruct the UV position map from a single 2D image. In the training process, the human face data is used for pre-training, and then a small amount of artifact data is used for fine-tuning. A deep learning model with strong generalization ability is trained with fewer artifact data, and a three-dimensional model of the cultural relic face can be reconstructed from a single old photograph. The methods can train more complex deep networks without a large amount of cultural relic data, and no over-fitting phenomenon occurs, which effectively solves the problem of fewer cultural relic samples. The method is verified by restoring a Chinese Terracotta Warrior with facial cracks and a Buddha with a broken nose. Other applications can be used in the fields such as texture recovery, facial feature extraction, and three-dimensional model estimation of the damaged cultural relics or sculptures in the photos. [ABSTRACT FROM AUTHOR]

Deep learning-based hyperspectral remote sensing image classification methods are currently a research hotspot. However, they suffer from issues such as large feature network parameter size, complex calculations, and the need for a large number of training data to achieve good classification results. Moreover, hyperspectral remote sensing images face challenges such as difficulty in obtaining the ground truth of land cover, limited availability of effective datasets for training, and endmember spectral variability, making it difficult for existing algorithm models to be widely adopted. To address these issues, this paper proposes a multi-branch classification model with multi-dimensional feature fusion, constructing lightweight deep network models for one-dimensional spectral, two-dimensional spatial, and three-dimensional depth feature extraction, respectively. This enriches feature information while reducing the parameters of each branch's deep model, effectively improving the land cover classification accuracy using hyperspectral remote sensing images under limited training sample conditions. Experimental verification with open-source hyperspectral remote sensing datasets shows that the proposed classification method can obtain over 90% classification accuracy when the training set account for only 5% of the total dataset, which is significantly better than current mainstream deep network classification models. [ABSTRACT FROM AUTHOR]

With the development of image feature matching technology, feature matching algorithms based on deep learning have achieved excellent results, but in scenarios with low texture or extreme perspective changes, the matching accuracy is still difficult to guarantee. In this paper, a superresolution reconstruction method based on a Residual-ESPCN (efficient subpixel convolutional neural network) approach is proposed based on LoFTR (local feature matching with transformers). The superresolution method is used to improve the interpolation method used in ASFF (adaptive spatial feature fusion) to increase the image resolution, enhance the detailed information of the image, and make the extracted features richer. Then, ASFF is introduced into the local feature extraction module of LoFTR, which can alleviate the inconsistency problem of information transmission between different scale features of the feature pyramid and lessen the amount of information lost during transmission from low- to high-resolution levels. Moreover, to improve the adaptability of the algorithm to different scenarios, OTSU is introduced to adaptively calculate the threshold of feature matching. The experimental results show that in different indoor or outdoor scenarios, our proposed algorithm for matching features can effectively improve the adaptability of feature matching and can achieve good results in terms of the area under the curve (AUC), accuracy and recall. [ABSTRACT FROM AUTHOR]

Brain–computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEP) have received widespread attention due to their high information transmission rate, high accuracy, and rich instruction set. However, the performance of its identification methods strongly depends on the amount of calibration data for within-subject classification. Some studies use deep learning (DL) algorithms for inter-subject classification, which can reduce the calculation process, but there is still much room for improvement in performance compared with intra-subject classification. To solve these problems, an efficient SSVEP signal recognition deep learning network model e-SSVEPNet based on the soft saturation nonlinear module is proposed in this paper. The soft saturation nonlinear module uses a similar exponential calculation method for output when it is less than zero, improving robustness to noise. Under the conditions of the SSVEP data set, two sliding time window lengths (1 s and 0.5 s), and three training data sizes, this paper evaluates the proposed network model and compares it with other traditional and deep learning model baseline methods. The experimental results of the nonlinear module were classified and compared. A large number of experimental results show that the proposed network has the highest average accuracy of intra-subject classification on the SSVEP data set, improves the performance of SSVEP signal classification and recognition, and has higher decoding accuracy under short signals, so it has huge potential ability to realize high-speed SSVEP-based for BCI. [ABSTRACT FROM AUTHOR]

Existing saliency algorithms based on deep learning are not sufficient to extract features of images. And the features are fused only during decoding. As a result, the edge of saliency detection result is not clear and the internal structure display is not uniform. To solve the above problems, this paper proposes a saliency detection method of dual-stream encoding fusion based on RGB and grayscale image. Firstly, an interactive dual-stream encoder is constructed to extract the feature information of gray stream and RGB stream. Secondly, a multi-level fusion strategy is used to obtain more effective multi-scale features. These features are extended and optimized in the decoding stage by linear transformation with hybrid attention. Finally, We propose a hybrid weighted loss function. So that the prediction results of the model can keep a high level accuracy at pixel level and region level. The experimental results of the model proposed to this paper on 6 public datasets illustrate that: The prediction results of the proposed method are clearer about the edge of salient targets and more uniform within salient targets. And has a more lightweight model size. [ABSTRACT FROM AUTHOR]

Danmakus are user-generated comments that overlay on videos, enabling real-time interactions between viewers and video content. The emotional orientation of danmakus can reflect the attitudes and opinions of viewers on video segments, which can help video platforms optimize video content recommendation and evaluate users' abnormal emotion levels. Aiming at the problems of low transferability of traditional sentiment analysis methods in the danmaku domain, low accuracy of danmaku text segmentation, poor consistency of sentiment annotation, and insufficient semantic feature extraction, this paper proposes a video danmaku sentiment analysis method based on MIBE-RoBERTa-FF-BiLSTM. This paper constructs a "Bilibili Must-Watch List and Top Video Danmaku Sentiment Dataset" by ourselves, covering 10,000 positive and negative sentiment danmaku texts of 18 themes. A new word recognition algorithm based on mutual information (MI) and branch entropy (BE) is used to discover 2610 irregular network popular new words from trigrams to heptagrams in the dataset, forming a domain lexicon. The Maslow's hierarchy of needs theory is applied to guide the consistent sentiment annotation. The domain lexicon is integrated into the feature fusion layer of the RoBERTa-FF-BiLSTM model to fully learn the semantic features of word information, character information, and context information of danmaku texts and perform sentiment classification. Comparative experiments on the dataset show that the model proposed in this paper has the best comprehensive performance among the mainstream models for video danmaku text sentiment classification, with an F1 value of 94.06%, and its accuracy and robustness are also better than other models. The limitations of this paper are that the construction of the domain lexicon still requires manual participation and review, the semantic information of danmaku video content and the positive case preference are ignored. [ABSTRACT FROM AUTHOR]

In industry, the task of defect classification and defect localization is an important part of defect detection system. However, existing studies only focus on one task and it is difficult to ensure the accuracy of both tasks. This paper proposes a defect detection system based on improved Yolo_v4, which greatly improves the detection ability of minor defects. For K_Means algorithm clustering prianchors question with strong subjectivity, the paper proposes the Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to determine the number of Anchors. To solve the problem of low detection rate of small targets caused by insufficient reuse rate of low-level features in CSPDarknet53 feature extraction network, this paper proposes an ECA-DenseNet-BC-121 feature extraction network to improve it. And the Dual Channel Feature Enhancement (DCFE) module is proposed to improve the local information loss and gradient propagation obstruction caused by quad chain convolution in PANet networks to improve the robustness of the model. The experimental results on the fabric surface defect detection datasets show that the mAP of the improved Yolo_v4 is 98.97%, which is 7.67% higher than SSD, 3.75% higher than Faster_RCNN, 10.82% higher than Yolo_v4 tiny, and 5.35% higher than Yolo_v4, and the detection speed reaches 39.4 fps. It can meet the real-time monitoring needs of industrial sites. [ABSTRACT FROM AUTHOR]

The detection of pulmonary nodules has a low accuracy due to the various shapes and sizes of pulmonary nodules. In this paper, a multi-scale detection network for pulmonary nodules based on the attention mechanism is proposed to accurately predict pulmonary nodules. During data processing, the pseudo-color processing strategy is designed to enhance the gray image and introduce more contextual semantic information. In the feature extraction network section, this paper designs a basic module of ResSCBlock integrating attention mechanism for feature extraction. At the same time, the feature pyramid structure is used for feature fusion in the network, and the problem of the detection of small-size nodules which are easily lost is solved by multi-scale prediction method. The proposed method is tested on the LUNA16 data set, with an 83% mAP value. Compared with other detection networks, the proposed method achieves an improvement in detecting pulmonary nodules. [ABSTRACT FROM AUTHOR]

An efficient malicious node detection system in CR networks is proposed in this paper. This proposed system contains features extraction process and optimization algorithm with soft computing framework. This proposed methodology stated in this paper initially abstracts the features of each individual node in CR network and these individual features are now getting optimized using feed forward radial neural network algorithm, which differentiates each individual node in CR network into either normal or malicious/faulty. This paper analyzes the performance of this proposed work with respect to malicious node detection rate, throughput and latency. [ABSTRACT FROM AUTHOR]

In recent years, highway accidents occur frequently, the main reason is that there is always foreign body invasion on the highway, which makes people unable to respond to emergencies in time. In order to reduce the occurrence of highway incidents, an object detection algorithm for highway intrusion was proposed in this paper. Firstly, a new feature extraction module was proposed to better preserve the main information. Secondly, a new feature fusion method was proposed to improve the accuracy of object detection. Finally, a lightweight method was proposed to reduce the computational complexity. We compare the algorithm in this paper with existing algorithms, the experimental results showed that: On the Visdrone dataset (small size targets), (a) the CS-YOLO was 3.6% more accurate than the YOLO v8. (b) The CS-YOLO was 1.2% more accurate than the YOLO v8 on the Tinypersons dataset (minimal size targets). (c) CS-YOLO was 1.4% more accurate than YOLO v8 on VOC2007 data set (normal size). [ABSTRACT FROM AUTHOR]

Ancient murals have suffered from continuous damage over time, and especially paint loss disease. Therefore, disease labeling, as the basis for ancient mural restoration, plays an important role in the protection of cultural relics. The predominant method of disease labeling is currently manual labeling, which is highly dependent on expert experience, time consuming, inefficient and results in inconsistent accuracy of the marking effect. In this paper, we propose a labeling framework for paint loss disease of ancient murals based on hyperspectral image classification and segmentation. The proposed framework involves first the extraction of features from the hyperspectral image, and then image segmentation is performed based on the spatial features to obtain more accurate region boundaries. Then, the hyperspectral image's regions are classified based on their spatial-spectral characteristics, and the candidate areas of paint loss disease are obtained. Finally, by leveraging the true color image segmentation results, the proposed disease labeling strategy combines the results of classification and segmentation to propose the final paint loss disease labeling areas. The experimental results show that the proposed method can not only combine the hyperspectral space and spectral information effectively to obtain accurate labeling of paint loss disease, but can also mark the paint loss disease not easily observed using ordinary digital cameras. Compared with the state-of-the-art methods, the proposed framework could be promising for accurate and effective paint loss disease labeling for ancient murals. [ABSTRACT FROM AUTHOR]

Since the dawn of civilization, handwriting has been one of the most important forms of communication. However, as handwriting differs from person to person, writer identification has become a promising application of pattern recognition to identify the actual writer of a handwritten document. Handwriting can be either online or offline, depending on how it was obtained. Users can write directly on tablets, smartphones, touch screens, PDAs, and other devices using input devices with online handwriting, whereas offline handwriting is done with a pen and paper. With the advent of artificial intelligence, and most importantly deep learning techniques, the development of writer identification systems based on offline handwritten documents has gained a lot of attention. Deep learning models have the capability of automatic feature extraction, which results in increased performance. From the literature survey, it was revealed that least attention has been paid towards the development of deep learning-based writer identification systems for offline Urdu handwritten documents, unlike the English and Arabic scripts. Therefore, in this paper, we proposed an offline Urdu handwritten writer identification system using a deep learning model inspired by the VGG-16 model of CNN. The model was trained and tested on a novel Urdu handwritten dataset contributed by 318 distinct Urdu writers, resulting in an overall training accuracy of 98.71% and a testing accuracy of 99.11%. The results achieved showed that the proposed model outperformed the already existing writer identification techniques. [ABSTRACT FROM AUTHOR]

With a significant number of mechanical products approaching the retirement phase, the batch recycling of discarded mechanical parts necessitates a preliminary assessment of their surface condition. However, the presence of surface rust poses a challenge to defect identification. Therefore, this paper proposes a method for detecting heavily rusted surface defects based on an improved YOLOv8n network. In the Backbone, the C2f-DBB module of re-parameterized deep feature extraction was introduced, and the attention module was designed to improve the accuracy of information extraction. In the Neck part, a Bi-Afpn multiscale feature fusion strategy is designed to facilitate information exchange between features at different scales. Finally, Focal-CIoU is employed as the bounding box loss function to enhance the network's localization performance and accuracy for defects. Experimentally, it is proved that the improved network in this paper improves the Recall, Precision, and mAP0.5 by 1.2%, 2.1%, and 1.9%, respectively, on the original basis, which is better than other network models. [ABSTRACT FROM AUTHOR]

Video recognition is the fundamental way human eyes and brains perceive the world. The human brain's real recognition, thinking, and memory mechanism has nonlinear dynamical system models, and the interaction between neurons exists in chaotic iterative patterns characterized by nonlinear dynamics. This paper proposes a new method for extracting video face features based on a nonlinear dynamical system, which is constructed by using video images as ternary functions and Discrete Cosine Transform (DCT) basis functions as auxiliary functions. Then, chaotic iterations are performed on the system, and the iteration trajectories are used as video features. Finally, the Pearson correlation coefficient between features is calculated to determine whether the faces in different videos are similar to complete the recognition work. Regarding database preprocessing, we propose Adaptive Threshold Region Detection (ATRD) algorithm for video region segmentation to optimize the database and improve recognition performance. Overall, this paper's method is simple, flexible, and has high recognition accuracy, with good recognition results on the YouTube and VidTIMIT databases. Of course, as a new method, there is still room for improvement and refinement. [ABSTRACT FROM AUTHOR]

Today, with the rapid growth of communication technology and the development of social networks and smartphones, the amount of data stored by users in the form of images has significantly increased. Therefore, several researchers have focused on the field of image retrieval in the past decade. Image retrieval means retrieving the most similar images to the query sample from a vast image database in terms of content and semantic. So far, various image retrieval methods based on feature engineering or deep features have been proposed. Most methods that use handcrafted features, despite their lower runtime, usually provide lower performance than deep learning-based approaches. In this paper, to increase the efficiency of handcrafted features in image retrieval, their combination with the output of residual blocks in deep neural networks is used in the feature fusion phase. In this paper, the efficiency of feature generation layers in common deep networks such as residual, classical sequential convolution and bottleneck has been analyzed in order to increase the efficiency of image retrieval systems based on handcrafted features. For this purpose, in most popular deep networks, the layers that perform the classification task are removed, and the output of feature generation layers at different depths, with the help of a flattened layer, is converted to feature vectors that can be connected to handcrafted made numerical features in retrieval systems. A variety of popular handcrafted features in the spatial domain, such as color and texture and wavelet in frequency, are employed to generate handcrafted features. This paper presents a novel hybrid feature set based on a combination of classical feature engineering techniques and deep convolutional neural networks suitable for image retrieval. The efficiency of the proposed method is evaluated in terms of precision and recall metrics on the benchmark datasets, such as Corel-1 and -5k. The proposed method provided an accuracy of 96.68 and 94.56% on the Corel-1k and Corel-5k, respectively. The comparative results showed that combining residual block feature maps with handcrafted features compared to bottleneck and sequential classical convolution layers increases the final performance of the image retrieval system. The comparison of the proposed method with state-of-the-art machine learning and deep learning-based techniques shows that the proposed method provides promising results in terms of precision and recall rates. [ABSTRACT FROM AUTHOR]

Existing salient object detection networks are large, have many parameters, are bulky and take up a lot of computational resources. Seriously hinder its application and promotion in boning robot. To solve this problem, this paper proposes a lightweight saliency detection algorithm for real-time localization of livestock meat bones. First, a lightweight feature extraction network based on multi-scale attention is constructed in the encoding stage. To ensure that more adequate salient object features are extracted with fewer parameters. Second, the fusion of jump connections is introduced in the decoding phase. Used to capture fine-grained semantics and coarse-grained semantics at full scale. Finally, we added a residual refinement module at the end of the backbone network. For optimizing salient target regions and boundaries. Experimental results on both publicly available datasets and self-made Pig leg X-ray (PLX) datasets show that. The proposed method is capable of ensuring first-class detection accuracy with 40 times less parameters than the conventional model. In the most challenging SOD dataset. The proposed algorithm in this paper achieves a value of Fωβ of 0.699. And the segmentation of livestock bones can be effectively performed on the homemade PLX dataset. Our model has a detection speed of 5fps on industrial control equipment. [ABSTRACT FROM AUTHOR]

With the advancement of society, ensuring the safety of personnel involved in municipal construction projects, particularly in the context of pandemic control measures, has become a matter of utmost importance. This paper introduces a security measure for municipal engineering, combining deep learning with object detection technology. It proposes a lightweight artificial intelligence (AI) detection method capable of simultaneously identifying individuals wearing masks and safety helmets. The method primarily incorporates the ShuffleNetv2 feature extraction mechanism within the framework of the YOLOv5s network to reduce computational overhead. Additionally, it employs the ECA attention mechanism and optimized loss functions to generate feature maps with more comprehensive information, thereby enhancing the precision of target detection. Experimental results indicate that this algorithm improves the mean average precision (mAP) value by 4.3%. Furthermore, it reduces parameter and computational loads by 54.8% and 53.8%, respectively, effectively striking a balance between lightweight operation and precision. This study serves as a valuable reference for research pertaining to lightweight target detection in the realm of municipal construction safety measures. [ABSTRACT FROM AUTHOR]

With the increasing number and types of global power loads and the development and popularization of smart grid technology, a large number of researches on load-level non-intrusive load monitoring technology have emerged. However, the unique power characteristics of the load make NILM face the difficult problem of low robustness of feature extraction and low accuracy of classification and identification in the recognition stage. This paper proposes a structured V-I mapping method to address the inherent limitations of traditional V-I trajectory mapping methods from a new perspective. In addition, for the verification of the V-I trajectory mapping method proposed in this paper, the complexity of load characteristics is comprehensively considered, and a lightweight convolutional neural network is designed based on AlexNet. The experimental results on the NILM dataset show that the proposed method significantly improves recognition accuracy compared to existing VI trajectory mapping methods. [ABSTRACT FROM AUTHOR]

The segmentation and extraction on color features can provide useful information for many different application domains. However, most of the existing image processing algorithms on feature extraction are gray image-based and consider only one-dimensional parameters. In order to carry out a fast and accurate color feature extraction, this paper proposes a color extraction algorithm by segmentation that is called a color extraction algorithm This algorithm is compared under different color distribution situations, and the extraction effect on color is also shown by the combination of the segmentation and feature extraction algorithms. Experimental results show that such segmentation algorithm has some advantages for color segmentation. In the fuzzy color image preprocessing, this paper gives the location method of region of interest. Moreover, compared with other existing extraction algorithms, the presented segmentation extraction algorithm in this paper not only has higher accuracy, shorter extraction time and stronger anti-interference ability, but also has better effect on more divergent color edge. Experimental evaluation of the proposed color extraction algorithm demonstrates dominance over existing algorithms for feature extraction. These researches in this paper provide a new way of thinking for color feature extraction by segmentation, which has an important theoretical references and practical significance. [ABSTRACT FROM AUTHOR]

The bird's eye view, multi-scale and dense classes in remote sensing images challenge the object detection of remote sensing images. It is not satisfactory to directly apply the object detection method designed for natural scene images to the object detection of remote sensing images. In this paper, we propose a detector with enhanced feature extraction ability to solve the above challenges, namely TWDFPN. TWDFPN has designed the structure of a two-way feature pyramid network (TWFPN) by combining feature maps with different generation directions and different spatial resolutions, which not only improves the utilization of the underlying feature information, but also strengthens the repeated utilization of the feature information of the backbone network, and ultimately improves the feature extraction ability of the network. Meanwhile, the dense-connected module is used in TWFPN to enhance the feature representation ability through limited additional computation cost, which extends the network and deepens the network. To evaluate the effectiveness of the proposed algorithm, this paper carried out experiments on NWPUVHR-10 and RSOD public remote sensing datasets, and the average accuracy (mAP) of 92.98% and 96.16%, respectively, which achieves advanced performance. [ABSTRACT FROM AUTHOR]

In this paper, we aim to implement augmented reality (AR) on distant text documents or books. For this purpose, we propose a new paper-based AR system that can detect text documents in real scenes, markerize and identify them, estimate their relative 3D poses to the camera, and augment them with virtual contents. Unlike the previous paper-based AR systems (applicable to only close documents), the proposed system not only requires no detection of words or characters, but allows partial occlusions like the previous systems. In our experiments, the proposed system worked at 24 fps and could consistently achieve high identification rates for both occluded and unoccluded pages. [ABSTRACT FROM AUTHOR]

YOLOv5 is one of the most popular object detection algorithms, which is divided into multiple series according to the control of network depth and width. To realize the deployment of mobile devices or embedded devices, the paper proposes a lightweight aerial image object detection algorithm (LAI-YOLOv5s) based on the improvement of YOLOv5s with a relatively small amount of calculation and parameter and relatively fast reasoning speed. Firstly, to better detect small objects, the paper replaces the minimum detection head with the maximum detection head and proposes a new feature fusion method, DFM-CPFN(Deep Feature Map Cross Path Fusion Network), to enrich the semantic information of deep features. Secondly, the paper designs a new module based on VoVNet to improve the feature extraction ability of the backbone network. Finally, based on the idea of ShuffleNetV2, the paper makes the network more lightweight without affecting detection accuracy. Based on the VisDrone2019 dataset, the detection accuracy of LAI-YOLOv5s on the mAP@0.5 index is 8.3% higher than that of the original algorithm. Compared with other series of YOLOv5 and YOLOv3 algorithms, LAI-YOLOv5s has the advantages of low computational cost and high detection accuracy. [ABSTRACT FROM AUTHOR]

Scene classification and recognition have always been one of the most challenging tasks of scene understanding due to the inherent ambiguity in visual scenes. The core of scene classification and recognition tasks is scene representation. Deep learning advances in computer vision, especially deep CNNs, have significantly improved scene representation in the last decade. Deep convolutional features extracted from deep CNNs provide discriminative representations of the images and are widely used in various computer vision tasks, such as scene classification. Deep convolutional features capture the appearance characteristics of the image and the spatial information about different image regions. Meanwhile, the semantic and context information obtained from high-level concepts about scene images, such as objects and their relationships, can significantly contribute to identifying scene images. Therefore, in this paper, we divide visual scenes into two categories, object-based and layout-based. Object-based scenes are scenes that have scene-specific objects and, based on those objects, can be described and identified. In contrast, the layout-based scenes do not have scene-specific objects and are described and identified based on the appearance and layout of the image. This paper proposes a new neural network model for representing and classifying visual scenes, which we call G-CNN (GNN-CNN). The proposed model includes two modules, feature extraction and feature fusion, and the feature extraction module composes of visual and semantic branches. The visual branch is responsible for extracting deep CNN features from the image, and the semantic branch is responsible for extracting semantic GNN features from the scene graph corresponding to the image. The feature fusion module is a novel two-stream neural network that fuses the CNN and GNN feature vectors to produce a comprehensive representation of the scene image. Finally, a fully-connected classifier classified the obtained comprehensive feature vector into one of the pre-defined categories. The proposed model has been evaluated on three benchmark scene datasets, UIUC Sports, MIT67, and SUN397, and obtained classification accuracy of 99.91%, 96.01%, and 85.32%, respectively. In addition, a new dataset named Scene40, which has been introduced in our previous paper, is also used for further evaluation of the proposed method. The comparison results based on classification accuracy criteria show that the proposed model can outperform the best previous methods on three benchmark scene datasets. [ABSTRACT FROM AUTHOR]

The detection of steel surface defects is of great significance to steel production. In order to better meet the requirements of accuracy, real-time, and lightweight model, this paper proposes a highly efficient and lightweight steel surface defect detection method based on YOLOv5n. Firstly, ODMobileNetV2 composed of MobileNetV2 and ODConv is used as the backbone to improve the defect feature extraction capability. Secondly, GSConv is utilized in the neck to achieve deep information fusion through channel concatenation and shuffling, enhancing the ability of feature fusion. Finally, this paper proposes a spatial-channel reconstruction block (SCRB) designed to suppress redundant features and improve the representation ability of defect features through feature separation and reconstruction. Experimental results show that this method achieves 84.1% mAP and 109 FPS on the NEU-DET dataset, and 72.9% mAP and 110.1 FPS on the GC10-DET dataset, enabling accurate and efficient detection. Furthermore, the number of parameters is only 5.04M, which has a significant lightweight advantage. [ABSTRACT FROM AUTHOR]

Classification algorithm is the crucial technique for machine vision system. The Transfer Component Analysis has been introduced for workpiece classification system. The dataset with five classes of workpieces is constructed for experiments, and the data augmentation strategies have been employed to prevent class imbalance problems. The Histogram of the Oriented Gradient features are extracted over workpiece images, and they are transformed by the Transfer Component Analysis and the Principal Component Analysis separately. The Maximum Mean Discrepancy distances have been employed to measure the distribution distances between the training set and the testing set. And the results indicate that the Transfer Component Analysis has dramatically reduced the distribution differences in a dimensionality reduction subspace, better than the Principal Component Analysis. The Error Correction Output Code is combined with Support Vector Machine and K-Nearest Neighbor respectively as the classifiers. The macro-averaged F 1 measure is adopted to evaluate the classifiers performance. Experiment results show that, the Transfer Component Analysis is particularly beneficial for improving the classification performance, and the Support Vector Machine is more favorable than the K-Nearest Neighbor for workpiece classification. The method of the Transfer Component Analysis combined with the Support Vector Machine possesses the best classification performance among these methods in this paper. [ABSTRACT FROM AUTHOR]

For the limitations of global feature-based deep learning point cloud registration algorithms in partial point cloud registration, this paper proposes a partial point cloud registration algorithm NcPE-PNLK combining global features and correspondence. The NcPE-PNLK algorithm introduces a feature interaction module to complete the information interaction between two point clouds in the feature extraction stage, which can improve the credibility of the feature. Moreover, the algorithm predicts the correspondence through the non-corresponding point estimation module, which reduces the influence of non-overlapping regions on the global feature and effectively solves the problem of performance degradation of the global feature registration algorithms in partial point cloud registration. We test the registration performance of NcPE-PNLK on synthetic scene dataset and real dataset in this paper. The experimental results show that NcPE-PNLK can effectively reduce the impact of non-overlapping regions in the registration process and achieve better performance compared with the global feature-based registration algorithms. In addition, compared with the correspondence-based registration algorithms, the NcPE-PNLK algorithm does not need to calculate correspondence precisely, which can achieve high-precision partial point cloud registration with guaranteed efficiency. [ABSTRACT FROM AUTHOR]

Emergent news is characterized by few labels, and news detection methods that rely on a large number of labels are difficult to apply to learned features for emerging events and are ineffective in coping with less labeled emergent news detection. To address the challenge of limited labeled data, this study first establishes a scenario for detecting breaking news, ensuring that the domain of detecting events is distinct from the domain of historical events. Secondly, we propose the Clip-GCN multimodal fake news detection model. The model utilizes the Clip pre-training model to perform joint semantic feature extraction of image-text information, with text information as the supervisory signal, which solves the problem of semantic interaction between modalities. Meanwhile, considering the domain attributes of news, the model is trained to extract inter-domain invariant features through Adversarial Neural Network ideation, and intra-domain knowledge information is utilized through graph convolutional networks (GCN) to detect emergent news. Through an extensive number of experiments on Chinese and English datasets from two major social media platforms, Weibo and Twitter, it is demonstrated that the model proposed in this paper can accurately screen multimodal emergent news on social media with an average accuracy of 88.7%. The contribution of this study lies not only in the improvement of model performance but also in the proposal of a solution for the challenges posed by limited labels and multimodal breaking news. This provides robust support for research in related fields. [ABSTRACT FROM AUTHOR]

Epilepsy is one of the most well-known neurological disorders globally, leading to individuals experiencing sudden seizures and significantly impacting their quality of life. Hence, there is an urgent necessity for an efficient method to detect and predict seizures in order to mitigate the risks faced by epilepsy patients. In this paper, a new method for seizure detection and prediction is proposed, which is based on multi-class feature fusion and the convolutional neural network-gated recurrent unit-attention mechanism (CNN-GRU-AM) model. Initially, the Electroencephalography (EEG) signal undergoes wavelet decomposition through the Discrete Wavelet Transform (DWT), resulting in six subbands. Subsequently, time–frequency domain and nonlinear features are extracted from each subband. Finally, the CNN-GRU-AM further extracts features and performs classification. The CHB-MIT dataset is used to validate the proposed approach. The results of tenfold cross validation show that our method achieved a sensitivity of 99.24% and 95.47%, specificity of 99.51% and 94.93%, accuracy of 99.35% and 95.16%, and an AUC of 99.34% and 95.15% in seizure detection and prediction tasks, respectively. The results show that the method proposed in this paper can effectively achieve high-precision detection and prediction of seizures, so as to remind patients and doctors to take timely protective measures. [ABSTRACT FROM AUTHOR]

In the realm of prognosticating the remaining useful life (RUL) of pivotal components, such as aircraft engines, a prevalent challenge persists where the available historical life data often proves insufficient. This insufficiency engenders obstacles such as impediments in performance degradation feature extraction, inadequacies in capturing temporal relationships comprehensively, and diminished predictive accuracy. To address this issue, a 1D CNN-GRU prediction model for few-shot conditions is proposed in this paper. In pursuit of more comprehensive data feature extraction and enhanced RUL prognostication precision, the Convolutional Neural Network (CNN) is selected for its capacity to discern high-dimensional features amid the intricate dynamics of the data. Concurrently, the Gated Recurrent Unit (GRU) network is leveraged for its robust capability in extracting temporal features inherent within the data. We combine the two to construct a CNN-GRU hybrid network. Moreover, the integration of data distribution alongside correlation and monotonicity indices is employed to winnow the input of multi-sensor monitoring parameters into the CNN-GRU network. Finally, the engine RULs are predicted by the trained model. In this paper, experiments are conducted on a sub-dataset of the National Aeronautics and Space Administration (NASA) C-MAPSS multi-constraint dataset to validate the effectiveness of the method. Experimental results have demonstrated that this method has high accuracy in RUL prediction tasks, which can powerfully demonstrate its effectiveness. [ABSTRACT FROM AUTHOR]

The intelligent appearance quality classification method for Auricularia auricula is of great significance to promote this industry. This paper proposes an appearance quality classification method for Auricularia auricula based on the improved Faster Region-based Convolutional Neural Networks (improved Faster RCNN) framework. The original Faster RCNN is improved by establishing a multiscale feature fusion detection model to improve the accuracy and real-time performance of the model. The multiscale feature fusion detection model makes full use of shallow feature information to complete target detection. It fuses shallow features with rich detailed information with deep features rich in strong semantic information. Since the fusion algorithm directly uses the existing information of the feature extraction network, there is no additional calculation. The fused features contain more original detailed feature information. Therefore, the improved Faster RCNN can improve the final detection rate without sacrificing speed. By comparing with the original Faster RCNN model, the mean average precision (mAP) of the improved Faster RCNN is increased by 2.13%. The average precision (AP) of the first-level Auricularia auricula is almost unchanged at a high level. The AP of the second-level Auricularia auricula is increased by nearly 5%. And the third-level Auricularia auricula AP is increased by 1%. The improved Faster RCNN improves the frames per second from 6.81 of the original Faster RCNN to 13.5. Meanwhile, the influence of complex environment and image resolution on the Auricularia auricula detection is explored. [ABSTRACT FROM AUTHOR]

Automatic and accurate recognition of abnormal working conditions of fused magnesia furnace is of great significance to the safe and reliable production of fused magnesia. Aiming at the defects of manual judgment of abnormal working conditions in the production process of fused magnesium furnace and the existing recognition method of abnormal working conditions based on machine learning, this paper proposes a working condition recognition model for fused magnesium furnace based on stochastic configuration networks and reinforcement learning. Firstly, a hybrid data augmentation method of generative and non-generative is used to obtain high-quality sample data with salient features. Secondly, based on ResNeXt, multi-scale local features are extracted under the condition of limited parameter quantity by controlling the cardinality. Combining a mixed model that enjoys the benefit of both self-attention and convolution (ACmix) and bidirectional feature pyramid network (BiFPN), the extracted local feature maps of different scales are cross-scale fused and focused, and more differentiated detailed feature information of the region of interest is retained. Thirdly, based on Transformer, the working condition recognition network of fused magnesium furnace is constructed to improve the global correlation between adjacent local features in the spatial dimension. The fused features are sent to stochastic configuration networks to establish a classification criterion for working condition recognition of fused magnesium furnace with generalization ability. Finally, reinforcement learning is used to evaluate the credibility of uncertain recognition results of samples in real time, and a self-optimizing adjustment action strategy at the Transformer encoding layer is defined. Build a library of Transformer models with different encoding layers, which is adapt to the different feature extraction requirements of multi-modal working samples. The experimental results show that the method in this paper has better recognition performance and generalization ability than other algorithms. [ABSTRACT FROM AUTHOR]

Marine mammals are an important part of marine ecosystems, and human intervention seriously threatens their living environments. Few studies exist on the marine mammal call recognition task, and the accuracy of current research needs to improve. In this paper, a novel MG-ResFormer two-channel fusion network architecture is proposed, which can extract local features and global timing information from sound signals almost perfectly. Second, in the input stage of the model, we propose an improved acoustic feature energy fingerprint, which is different from the traditional single feature approach. This feature also contains frequency, energy, time sequence and other speech information and has a strong identity. Additionally, to achieve more reliable accuracy in the multiclass call recognition task, we propose a multigranular joint layer to capture the family and genus relationships between classes. In the experimental section, the proposed method is compared with the existing feature extraction methods and recognition methods. In addition, this paper also compares with the latest research, and the proposed method is the most advanced algorithm thus far. Ultimately, our proposed method achieves an accuracy of 99.39% in the marine mammal call recognition task. [ABSTRACT FROM AUTHOR]

The traditional local binary pattern (LBP) is susceptible to the influence of the centre pixel and noise and cannot accurately identify wire rope surface defects. To solve this problem, an image segmentation-based central multiscale local binary pattern (ISCM-LBP) and grey level cooccurrence matrix (GLCM) feature fusion method is proposed in this paper for defect recognition. Image segmentation and multiple scales are introduced into the local binary pattern algorithm to improve the image detail description and suppress noise sensitivity. Second, the centre pixel is connected with the neighbourhood pixel to enhance the robustness of the centre pixel. To further improve the image integrity description, PCA dimensionality reduction and GLCM feature fusion are performed on the features extracted by the ISCM-LBP algorithm, and the steel wire rope surface defects are identified by a support vector machine classifier. Experimental results show that the overall recognition rate reaches 97.5%, which is at least 5% higher than that of other algorithms and can effectively identify various defects on the surface of wire rope. [ABSTRACT FROM AUTHOR]

Knowledge graph embedding converts knowledge graphs based on symbolic representations into low-dimensional vectors. Effective knowledge graph embedding methods are key to ensuring downstream tasks. Some studies have shown significant performance differences among various knowledge graph embedding models on different datasets. They attribute this issue to the insufficient representation ability of the models. However, what representation ability knowledge graph embedding models possess is still unknown. Therefore, this paper first selects three representative models for analysis: translation and rotation models in distance models, and the Bert model in neural network models. Based on the analysis results, it can be concluded that the translation model focuses on clustering features, the rotation model focuses on hierarchy features, and the Bert model focuses on word co-occurrence features. This paper categorize clustering and hierarchy as structure features, and word co-occurrence as semantic features. Furthermore, a model that solely focuses on a single feature will lead to a lack of accuracy and generality, making it challenging for the model to be applicable to modern large-scale knowledge graphs. Therefore, this paper proposes an ensemble model with structure and semantic features for knowledge graph embedding. Specifically, the ensemble model includes a structure part and a semantic part. The structure part consists of three models: translation, rotation and cross. Translation and rotation models serve as basic feature extraction, while the cross model enhances the interaction between them. The semantic part is built based on Bert and integrated with the structure part after fine-tuning. In addition, this paper also introduces a frequency model to mitigate the training imbalance caused by differences in entity frequencies. Finally, we verify the effectiveness of the model through link prediction. Experiments show that the ensemble model has achieved improvement on FB15k-237 and YAGO3-10, and also has good performance on WN18RR, proving the effectiveness of the model. [ABSTRACT FROM AUTHOR]

Industrial defect detection is a hot topic in the field of computer vision. It is a challenging task due to complex features and many categories of industrial defects. In this paper, a deep learning model based on the multiscale feature extraction module is introduced for steel surface defect detection. The main focus on the feature extraction capability of the model and feature fusion capability to improve the accuracy of the model for steel surface defect detection. First, to improve the feature extraction ability of the model, a multiscale feature extraction (MSFE) module is introduced. The MSFE module can effectively extract multiscale features through three branches that have different convolution kernel sizes. Second, an efficient feature fusion (EFF) module is proposed to optimize feature fusion by adding features from the backbone network to the neck network. Third, this paper puts forward a new Bottleneck module by reducing the normalization layer and activation function in the original Bottleneck module. Finally, the backbone network is deepened to further enhance the feature extraction ability of the model. Extensive experiments are conducted on the public NEU-DET dataset. The experimental results validate the effectiveness of the designed modules and the proposed model. Compared with other state-of-the-art methods, the proposed model achieves optimal accuracy(73.08% mAP@0.5) while maintaining a small number of parameters. [ABSTRACT FROM AUTHOR]

Micro-expressions (MEs) are natural facial mechanisms with short duration and subtle changes. It has attracted much attention in the real world due to its accuracy and uncontrollability of mental expression. With the development of computer vision, micro-expression Recognition (MER) methods have been continuously proposed and improved by scholars. However, the existing MER methods still have some deficiencies in processing Spatio-temporal redundant information and feature extraction. This paper proposes an MER network based on Differential Feature Fusion (DFF) method to solve this problem. First, inputs the onset frame and apex frame of the face, divide each image into small blocks, and uses part of the SE-ResNet50 model for feature extraction. Second, the Spatio-Temporal information of the features is extracted by using a DFF module composed of a differential feature module, CapsuleNet, and a Fully Connected (FC) layer. Finally, inputs the feature vector to the FC module for classification. This study is based on the Leave One Subject Out (LOSO) cross-validation protocol and uses the CASMEII dataset. Experiments and comparisons show the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

Around 50 million individuals worldwide suffer from epilepsy, a chronic, non-communicable brain disorder. Several screening methods, including electroencephalography, have been proposed to identify epileptic episodes. EEG data, which are frequently utilised to enhance epilepsy analysis, offer essential information on the electrical processes of the brain. Prior to the emergence of deep learning (DL), feature extraction was accomplished by standard machine learning techniques. As a result, they were only as good as the people who made the features by hand. But with DL, both feature extraction and classification are fully automated. These methods have significantly advanced several fields of medicine, including the diagnosis of epilepsy. In this paper, the works focused on automated epileptic seizure detection using ML and DL techniques are presented as well as their comparative analysis is done. The UCI-Epileptic Seizure Recognition dataset is used for training and validation. Some of the conventional ML and DL algorithms are used with a proposed model which uses long short-term memory (LSTM) to find the best approach. Post that comparative analysis is performed on these algorithms to find the best approach for epileptic seizure detection. As a result, the proposed model LSTM gives a validation accuracy of 97% giving the most appropriate and precise result as compared to other mentioned algorithms used in this study. [ABSTRACT FROM AUTHOR]