A Random Fusion of Mix 3D and Polar Mix to Improve Semantic Segmentation Performance in 3D Lidar Point Cloud

This paper focuses on the effective utilization of data augmentation techniques for 3Dlidar point clouds to enhance the performance of neural network models.These point clouds,which represent spatial information through a collection of 3D coordinates,have found wide-ranging applications.Data augmentation has emerged as a potent solution to the challenges posed by limited labeled data and the need to enhance model generalization capabilities.Much of the existing research is devoted to crafting novel data augmentation methods specifically for 3D lidar point clouds.However,there has been a lack of focus on making the most of the numerous existing augmentation techniques.Addressing this deficiency,this research investigates the possibility of combining two fundamental data augmentation strategies.The paper introduces PolarMix andMix3D,two commonly employed augmentation techniques,and presents a new approach,named RandomFusion.Instead of using a fixed or predetermined combination of augmentation methods,RandomFusion randomly chooses one method from a pool of options for each instance or sample.This innovative data augmentation technique randomly augments each point in the point cloud with either PolarMix or Mix3D.The crux of this strategy is the random choice between PolarMix and Mix3Dfor the augmentation of each point within the point cloud data set.The results of the experiments conducted validate the efficacy of the RandomFusion strategy in enhancing the performance of neural network models for 3D lidar point cloud semantic segmentation tasks.This is achieved without compromising computational efficiency.By examining the potential of merging different augmentation techniques,the research contributes significantly to a more comprehensive understanding of how to utilize existing augmentation methods for 3D lidar point clouds.RandomFusion data augmentation technique offers a simple yet effective method to leverage the diversity of augmentation techniques and boost the robustness of models.The insights gained from this research can pave the way for future work aimed at developing more advanced and efficient data augmentation strategies for 3D lidar point cloud analysis.

Teaching Reform and Practice of Software Engineering Graduate Students by Implanting Scientific Research Literacy into Classroom Teaching Content

How to cultivate and improve graduate students’innovation and practical abilities in software engineering through the curriculum and teaching mode reform is an important issue.In this paper,a research literacy-driven teaching mode is proposed.It assists in the reform of the curriculum system.Then,a curriculum system construction framework is proposed,which involves the integration of research literacy into classroom teaching content.It assists in the cultivation of research abilities of graduate students in software engineering.The effectiveness of the curriculum reform is demonstrated through questionnaire surveys and research outcomes of the project team.The results show that the methods explored in this paper can serve as valuable references for future course design and teaching practice in computer-related courses for graduates.

A Privacy Preservation Method for Attributed Social Network Based on Negative Representation of Information

Due to the presence of a large amount of personal sensitive information in social networks,privacy preservation issues in social networks have attracted the attention of many scholars.Inspired by the self-nonself discrimination paradigmin the biological immune system,the negative representation of information indicates features such as simplicity and efficiency,which is very suitable for preserving social network privacy.Therefore,we suggest a method to preserve the topology privacy and node attribute privacy of attribute social networks,called AttNetNRI.Specifically,a negative survey-based method is developed to disturb the relationship between nodes in the social network so that the topology structure can be kept private.Moreover,a negative database-based method is proposed to hide node attributes,so that the privacy of node attributes can be preserved while supporting the similarity estimation between different node attributes,which is crucial to the analysis of social networks.To evaluate the performance of the AttNetNRI,empirical studies have been conducted on various attribute social networks and compared with several state-of-the-art methods tailored to preserve the privacy of social networks.The experimental results show the superiority of the developed method in preserving the privacy of attribute social networks and demonstrate the effectiveness of the topology disturbing and attribute hiding parts.The experimental results show the superiority of the developed methods in preserving the privacy of attribute social networks and demonstrate the effectiveness of the topological interference and attribute-hiding components.

Expression Recognition Method Based on Convolutional Neural Network and Capsule Neural Network

Convolutional neural networks struggle to accurately handle changes in angles and twists in the direction of images,which affects their ability to recognize patterns based on internal feature levels. In contrast, CapsNet overcomesthese limitations by vectorizing information through increased directionality and magnitude, ensuring that spatialinformation is not overlooked. Therefore, this study proposes a novel expression recognition technique calledCAPSULE-VGG, which combines the strengths of CapsNet and convolutional neural networks. By refining andintegrating features extracted by a convolutional neural network before introducing theminto CapsNet, ourmodelenhances facial recognition capabilities. Compared to traditional neural network models, our approach offersfaster training pace, improved convergence speed, and higher accuracy rates approaching stability. Experimentalresults demonstrate that our method achieves recognition rates of 74.14% for the FER2013 expression dataset and99.85% for the CK+ expression dataset. By contrasting these findings with those obtained using conventionalexpression recognition techniques and incorporating CapsNet’s advantages, we effectively address issues associatedwith convolutional neural networks while increasing expression identification accuracy.

Identifying Counterexamples Without Variability in Software Product Line Model Checking

Product detection based on state abstraction technologies in the software product line(SPL)is more complex when compared to a single system.This variability constitutes a new complexity,and the counterexample may be valid for some products but spurious for others.In this paper,we found that spurious products are primarily due to the failure states,which correspond to the spurious counterexamples.The violated products correspond to the real counterexamples.Hence,identifying counterexamples is a critical problem in detecting violated products.In our approach,we obtain the violated products through the genuine counterexamples,which have no failure state,to avoid the tedious computation of identifying spurious products dealt with by the existing algorithm.This can be executed in parallel to improve the efficiency further.Experimental results showthat our approach performswell,varying with the growth of the system scale.By analyzing counterexamples in the abstract model,we observed that spurious products occur in the failure state.The approach helps in identifying whether a counterexample is spurious or genuine.The approach also helps to check whether a failure state exists in the counterexample.The performance evaluation shows that the proposed approach helps significantly in improving the efficiency of abstraction-based SPL model checking.

Brain Functional Networks with Dynamic Hypergraph Manifold Regularization for Classification of End-Stage Renal Disease Associated with Mild Cognitive Impairment

The structure and function of brain networks have been altered in patients with end-stage renal disease(ESRD).Manifold regularization(MR)only considers the pairing relationship between two brain regions and cannot represent functional interactions or higher-order relationships between multiple brain regions.To solve this issue,we developed a method to construct a dynamic brain functional network(DBFN)based on dynamic hypergraph MR(DHMR)and applied it to the classification of ESRD associated with mild cognitive impairment(ESRDaMCI).The construction of DBFN with Pearson’s correlation(PC)was transformed into an optimization model.Node convolution and hyperedge convolution superposition were adopted to dynamically modify the hypergraph structure,and then got the dynamic hypergraph to form the manifold regular terms of the dynamic hypergraph.The DHMR and L_(1) norm regularization were introduced into the PC-based optimization model to obtain the final DHMR-based DBFN(DDBFN).Experiment results demonstrated the validity of the DDBFN method by comparing the classification results with several related brain functional network construction methods.Our work not only improves better classification performance but also reveals the discriminative regions of ESRDaMCI,providing a reference for clinical research and auxiliary diagnosis of concomitant cognitive impairments.

Heterogeneous Image Knowledge Driven Visual Perception

Dear Editor,This letter is concerned with visual perception closely related to heterogeneous images.Facing the huge challenge brought by different image modalities,we propose a visual perception framework based on heterogeneous image knowledge,i.e.,the domain knowledge associated with specific vision tasks,to better address the corresponding visual perception problems.

Part-Whole Relational Few-Shot 3D Point Cloud Semantic Segmentation

This paper focuses on the task of few-shot 3D point cloud semantic segmentation.Despite some progress,this task still encounters many issues due to the insufficient samples given,e.g.,incomplete object segmentation and inaccurate semantic discrimination.To tackle these issues,we first leverage part-whole relationships into the task of 3D point cloud semantic segmentation to capture semantic integrity,which is empowered by the dynamic capsule routing with the module of 3D Capsule Networks(CapsNets)in the embedding network.Concretely,the dynamic routing amalgamates geometric information of the 3D point cloud data to construct higher-level feature representations,which capture the relationships between object parts and their wholes.Secondly,we designed a multi-prototype enhancement module to enhance the prototype discriminability.Specifically,the single-prototype enhancement mechanism is expanded to the multi-prototype enhancement version for capturing rich semantics.Besides,the shot-correlation within the category is calculated via the interaction of different samples to enhance the intra-category similarity.Ablation studies prove that the involved part-whole relations and proposed multi-prototype enhancement module help to achieve complete object segmentation and improve semantic discrimination.Moreover,under the integration of these two modules,quantitative and qualitative experiments on two public benchmarks,including S3DIS and ScanNet,indicate the superior performance of the proposed framework on the task of 3D point cloud semantic segmentation,compared to some state-of-the-art methods.

Pre-training transformer with dual-branch context content module for table detection in document images

Background Document images such as statistical reports and scientific journals are widely used in information technology.Accurate detection of table areas in document images is an essential prerequisite for tasks such as information extraction.However,because of the diversity in the shapes and sizes of tables,existing table detection methods adapted from general object detection algorithms,have not yet achieved satisfactory results.Incorrect detection results might lead to the loss of critical information.Methods Therefore,we propose a novel end-to-end trainable deep network combined with a self-supervised pretraining transformer for feature extraction to minimize incorrect detections.To better deal with table areas of different shapes and sizes,we added a dualbranch context content attention module(DCCAM)to high-dimensional features to extract context content information,thereby enhancing the network's ability to learn shape features.For feature fusion at different scales,we replaced the original 3×3 convolution with a multilayer residual module,which contains enhanced gradient flow information to improve the feature representation and extraction capability.Results We evaluated our method on public document datasets and compared it with previous methods,which achieved state-of-the-art results in terms of evaluation metrics such as recall and F1-score.https://github.com/Yong Z-Lee/TD-DCCAM.

Energy and Bandwidth Based Link Stability Routing Algorithm for IoT

Internet of Things(IoT)is becoming popular nowadays for collecting and sharing the data from the nodes and among the nodes using internet links.Particularly,some of the nodes in IoT are mobile and dynamic in nature.Hence maintaining the link among the nodes,efficient bandwidth of the links among the mobile nodes with increased life time is a big challenge in IoT as it integrates mobile nodes with static nodes for data processing.In such networks,many routing-problems arise due to difficulties in energy and bandwidth based quality of service.Due to the mobility and finite nature of the nodes,transmission links between intermediary nodes may fail frequently,thus affecting the routing-performance of the network and the accessibility of the nodes.The existing protocols do not focus on the transmission links and energy,bandwidth and link stability of the nodes,but node links are significant factors for enhancing the quality of the routing.Link stability helps us to define whether the node is within or out of a coverage range.This paper proposed an Optimal Energy and bandwidth based Link Stability Routing(OEBLS)algorithm,to improve the link stable route with minimized error rate and throughput.In this paper,the optimal route from the source to the sink is determined based on the energy and bandwidth,link stability value.Among the existing routes,the sink node will choose the optimal route which is having less link stability value.Highly stable link is determined by evaluating link stability value using distance and velocity.Residual-energy of the node is estimated using the current energy and the consumed energy.Consumed energy is estimated using transmitted power and the received power.Available bandwidth in the link is estimated using the idle time and channel capacity with the consideration of probability of collision.

DAO to HANOI via DeSci: AI Paradigm Shifts from AlphaGo to ChatGPT

From AlphaGo to ChatGPT,the field of AI has launched a series of remarkable achievements in recent years.Analyzing,comparing,and summarizing these achievements at the paradigm level is important for future AI innovation,but has not received sufficient attention.In this paper,we give an overview and perspective on machine learning paradigms.First,we propose a paradigm taxonomy with three levels and seven dimensions from a knowledge perspective.Accordingly,we give an overview on three basic and twelve extended learning paradigms,such as Ensemble Learning,Transfer Learning,etc.,with figures in unified style.We further analyze three advanced paradigms,i.e.,AlphaGo,AlphaFold and ChatGPT.Second,to enable more efficient and effective scientific discovery,we propose to build a new ecosystem that drives AI paradigm shifts through the decentralized science(DeSci)movement based on decentralized autonomous organization(DAO).To this end,we design the Hanoi framework,which integrates human factors,parallel intelligence based on a combination of artificial systems and the natural world,and the DAO to inspire AI innovations.

Cross-Domain TSK Fuzzy System Based on Semi-Supervised Learning for Epilepsy Classification

Through semi-supervised learning and knowledge inheritance,a novel Takagi-Sugeno-Kang(TSK)fuzzy system framework is proposed for epilepsy data classification in this study.The new method is based on the maximum mean discrepancy(MMD)method and TSK fuzzy system,as a basic model for the classification of epilepsy data.First,formedical data,the interpretability of TSK fuzzy systems can ensure that the prediction results are traceable and safe.Second,in view of the deviation in the data distribution between the real source domain and the target domain,MMD is used to measure the distance between different data distributions.The objective function is constructed according to the MMD distance,and the distribution distance of different datasets is minimized to find the similar characteristics of different datasets.We introduce semi-supervised learning to further explore the relationship between data.Based on the MMD method,a semi-supervised learning(SSL)-MMD method is constructed by using pseudo-tags to realize the data distribution alignment of the same category.In addition,the idea of knowledge dissemination is used to learn pseudo-tags as additional data features.Finally,for epilepsy classification,the cross-domain TSK fuzzy system uses the cross-entropy function as the objective function and adopts the back-propagation strategy to optimize the parameters.The experimental results show that the new method can process complex epilepsy data and identify whether patients have epilepsy.

Identification of coexistence of biological and non-biological aerosol particles with DAPI (4′,6-diamidino-2-phenylindole) stain

The fluorescent dye 4′,6-diamidino-2-phenylindole (DAPI) has been widely used to stain microorganisms in various environment media. We applied DAPI fluorescence enumeration to airborne microorganisms and found that non-biological particles, including organic compounds, minerals, and soot, were also visible upon exposure to UV excitation under fluorescence microscope. Using laboratory-prepared biological particles as the control, we investigated the feasibility of identifying both biological and non-biological particles in the same sample with DAPI staining. We prepared biological (bacterial, fungi, and plant detritus) and non-biological (biochar, soot, mineral, metal, fly ash, salt) particles in the laboratory and enumerated the particles and their mixture with DAPI. We found that mineral particles were transparent, and biochar, soot, metals and fly ash particles were black under a filter set at excitation 350/50 nm and emission 460/50 nm bandpass (DAPI-BP), while biological particles were blue, as expected. Particles of the water-soluble salts NaCl and (NH_(4))_(2)SO_(4) were yellow under a filter set at excitation 340–380 nm and emission 425 nm long pass (DAPI-LP). Case studies with samples of dustfall, atmospheric aerosols and surface soils could allow for the quantification of the relative number of different types of particles and particles with organic matter or salt coating as well. Fluorescence enumeration with DAPI stain is thus able to identify the co-existence of biological and non-biological particles in the air, at least to the extent of those examined in this study.

Noise-Filtering Enhanced Deep Cognitive Diagnosis Model for LatentSkill Discovering

Educational data mining based on student cognitive diagnosis analysis can provide an important decision basis for personalized learning tutoring of students,which has attracted extensive attention from scholars at home and abroad and has made a series of important research progress.To this end,we propose a noise-filtering enhanced deep cognitive diagno-sis method to improve the fitting ability of traditional models and obtain students’skill mastery status by mining the interaction between students and problems nonlinearly through neural networks.First,modeling complex interactions between students and problems with multidimensional features based on cognitive processing theory can enhance the interpretability of the proposed model;second,the neural network is used to predict students’learning performance,diagnose students’skill mastery and provide immediate feedback;finally,by comparing the proposed model with several baseline models,extensive experimental results on real data sets demonstrate that the proposed Finally,by comparing the proposed model with several baseline models,the extensive experimental results on the actual data set demon-strate that the proposed model not only improves the accuracy of predicting students’learning performance but also enhances the interpretability of the neurocognitive diagnostic model.

Dynamic motion of quadrupedal robots on challenging terrain:a kinodynamic optimization approach

The dynamic motion of quadrupedal robots on challenging terrain generally requires elaborate spatial–temporal kinodynamic motion planning and accurate control at higher refresh rate in comparison with regular terrain.However,conventional quadrupedal robots usually generate relatively coarse planning and employ motion replanning or reactive strategies to handle terrain irregularities.The resultant complex and computation-intensive controller may lead to nonoptimal motions or the breaking of locomotion rhythm.In this paper,a kinodynamic optimization approach is presented.To generate long-horizon optimal predictions of the kinematic and dynamic behavior of the quadruped robot on challenging terrain,we formulate motion planning as an optimization problem;jointly treat the foot’s locations,contact forces,and torso motions as decision variables;combine smooth motion and minimal energy consumption as the objective function;and explicitly represent feasible foothold region and friction constraints based on terrain information.To track the generated motions accurately and stably,we employ a whole-body controller to compute reference position and velocity commands,which are fed forward to joint controllers of the robot’s legs.We verify the effectiveness of the developed approach through simulation and on a physical quadruped robot testbed.Results show that the quadruped robot can successfully traverse a 30°slope and 43% of nominal leg length high step while maintaining the rhythm of dynamic trot gait.

Reversible data hiding in encrypted images based on additive secret sharing and additive joint coding using an intelligent predictor

Reversible data hiding in encrypted images(RDHEI)is essential for safeguarding sensitive information within the encrypted domain.In this study,we propose an intelligent pixel predictor based on a residual group block and a spatial attention module,showing superior pixel prediction performance compared to existing predictors.Additionally,we introduce an adaptive joint coding method that leverages bit-plane characteristics and intra-block pixel correlations to maximize embedding space,outperforming single coding approaches.The image owner employs the presented intelligent predictor to forecast the original image,followed by encryption through additive secret sharing before conveying the encrypted image to data hiders.Subsequently,data hiders encrypt secret data and embed them within the encrypted image before transmitting the image to the receiver.The receiver can extract secret data and recover the original image losslessly,with the processes of data extraction and image recovery being separable.Our innovative approach combines an intelligent predictor with additive secret sharing,achieving reversible data embedding and extraction while ensuring security and lossless recovery.Experimental results demonstrate that the predictor performs well and has a substantial embedding capacity.For the Lena image,the number of prediction errors within the range of[-5,5]is as high as 242500 and our predictor achieves an embedding capacity of 4.39 bpp.

Mobility-Aware and Energy-Efficient Task Offloading Strategy for Mobile Edge Workflows

With the rapid growth of the Industrial Internet of Things(IIoT), the Mobile Edge Computing(MEC) has coming widely used in many emerging scenarios. In MEC, each workflow task can be executed locally or offloaded to edge to help improve Quality of Service(QoS) and reduce energy consumption. However, most of the existing offloading strategies focus on independent applications, which cannot be applied efficiently to workflow applications with a series of dependent tasks. To address the issue,this paper proposes an energy-efficient task offloading strategy for large-scale workflow applications in MEC. First, we formulate the task offloading problem into an optimization problem with the goal of minimizing the utility cost, which is the trade-off between energy consumption and the total execution time. Then, a novel heuristic algorithm named Green DVFS-GA is proposed, which includes a task offloading step based on the genetic algorithm and a further step to reduce the energy consumption using Dynamic Voltage and Frequency Scaling(DVFS) technique. Experimental results show that our proposed strategy can significantly reduce the energy consumption and achieve the best trade-off compared with other strategies.

Printed Surface Defect Detection Model Based on Positive Samples

For a long time, the detection and extraction of printed surfacedefects has been a hot issue in the print industry. Nowadays, defect detectionof a large number of products still relies on traditional image processingalgorithms such as scale invariant feature transform (SIFT) and orientedfast and rotated brief (ORB), and researchers need to design algorithms forspecific products. At present, a large number of defect detection algorithmsbased on object detection have been applied but need lots of labeling sampleswith defects. Besides, there are many kinds of defects in printed surface,so it is difficult to enumerate all defects. Most defect detection based onunsupervised learning of positive samples use generative adversarial networks(GAN) and variational auto-encoders (VAE) algorithms, but these methodsare not effective for complex printed surface. Aiming at these problems, Inthis paper, an unsupervised defect detection and extraction algorithm forprinted surface based on positive samples in the complex printed surface isproposed innovatively. We propose a kind of defect detection and extractionnetwork based on image matching network. This network is divided into thefull convolution network of feature points extraction, and the graph attentionnetwork using self attention and cross attention. Though the key pointsextraction network, we can get robustness key points in the complex printedimages, and the graph network can solve the problem of the deviation becauseof different camera positions and the influence of defect in the differentproduction lines. Just one positive sample image is needed as the benchmarkto detect the defects. The algorithm in this paper has been proved in “TheFirst ZhengTu Cup on Campus Machine Vision AI Competition” and gotexcellent results in the finals. We are working with the company to apply it inproduction.

A Prediction Method Based on Improved Echo State Network for COVID-19 Nonlinear Time Series

This paper proposes a prediction method based on improved Echo State Network for COVID-19 nonlinear time series, which improves the Echo State Network from the reservoir topology and the output weight matrix, and adopt the ABC (Artificial Bee Colony) algorithm based on crossover and crowding strategy to optimize the parameters. Finally, the proposed method is simulated and the results show that it has stronger prediction ability for COVID-19 nonlinear time series.

Knowledge Distillation via Hierarchical Matching for Small Object Detection

Knowledge distillation is often used for model compression and has achieved a great breakthrough in image classification,but there still remains scope for improvement in object detection,especially for knowledge extraction of small objects.The main problem is the features of small objects are often polluted by background noise and not prominent due to down-sampling of convolutional neural network(CNN),resulting in the insufficient refinement of small object features during distillation.In this paper,we propose Hierarchical Matching Knowledge Distillation Network(HMKD)that operates on the pyramid level P2 to pyramid level P4 of the feature pyramid network(FPN),aiming to intervene on small object features before affecting.We employ an encoder-decoder network to encapsulate low-resolution,highly semantic information,akin to eliciting insights from profound strata within a teacher network,and then match the encapsulated information with high-resolution feature values of small objects from shallow layers as the key.During this period,we use an attention mechanism to measure the relevance of the inquiry to the feature values.Also in the process of decoding,knowledge is distilled to the student.In addition,we introduce a supplementary distillation module to mitigate the effects of background noise.Experiments show that our method achieves excellent improvements for both one-stage and twostage object detectors.Specifically,applying the proposed method on Faster R-CNN achieves 41.7%mAP on COCO2017(ResNet50 as the backbone),which is 3.8%higher than that of the baseline.