Weak Robustness Research Articles

A cloud-oriented siamese network object tracking algorithm with attention network and adaptive loss function

Aiming at solving the problems of low success rate and weak robustness of object tracking algorithms based on siamese network in complex scenes with occlusion, deformation, and rotation, a siamese network object tracking algorithm with attention network and adaptive loss function (SiamANAL) is proposed. Firstly, the multi-layer feature fusion module for template branch (MFFMT) and the multi-layer feature fusion module for search branch (MFFMS) are designed. The modified convolutional neural networks (CNN) are used for feature extraction through the fusion module to solve the problem of features loss caused by too deep network. Secondly, an attention network is introduced into the SiamANAL algorithm to calculate the attention of template map features and search map features, which enhances the features of object region, reduces the interference of background region, and improves the accuracy of the algorithm. Finally, an adaptive loss function combined with pairwise Gaussian loss function and cross entropy loss function is designed to increase inter-class separation and intra-class compactness of classification branches and improve the accuracy rate of classification and the success rate of regression. The effectiveness of the proposed algorithm is verified by comparing it with other popular algorithms on two popular benchmarks, the visual object tracking 2018 (VOT2018) and the object tracking benchmark 100 (OTB100). Extensive experiments demonstrate that the proposed tracker achieves competitive performance against state-of-the-art trackers. The success rate and precision rate of the proposed algorithm SiamANAL on OTB100 are 0.709 and 0.883, respectively. With the help of cloud computing services and data storage, the processing performance of the proposed algorithm can be further improved.

Multistructure Graph Classification Method With Attention-Based Pooling

Graph neural networks (GNNs) have achieved effective performance in many graph-related tasks involving recommendation systems, social networks, and bioinformatics. Recent studies have proposed several graph pooling operators to obtain graph-level representations from node representations. Nevertheless, they usually adopt a single strategy to evaluate the importance of nodes, which may generate node rankings with weak robustness. Also, they cannot capture the different substructures of a graph since they shrink the graph layer by layer. To solve the above problems, this article proposes a Multistructure graph classification method with Attention mechanism and Convolutional neural network (CNN), called MAC. In particular, we propose a novel pooling operator, which adopts multiple strategies to evaluate the importance of nodes and updates node representations through an attention mechanism. Also, we design a hierarchical architecture for MAC to capture multiple different substructures of a graph. To further reduce the loss of graph information, we utilize 2-D CNN to generate a graph-level representation. Comparative experiments are performed on public benchmark datasets deriving from social systems, and the experimental results indicate that our method outperforms a range of state-of-the-art graph classification methods.

Regularized Denoising Masked Visual Pretraining for Robust Embodied PointGoal Navigation

Embodied PointGoal navigation is a fundamental task for embodied agents. Recent works have shown that the performance of the embodied navigation agent degrades significantly in the presence of visual corruption, including Spatter, Speckle Noise, and Defocus Blur, showing the weak robustness of the agent. To improve the robustness of embodied navigation agents to various visual corruptions, we propose a navigation framework called Regularized Denoising Masked AutoEncoders Navigation (RDMAE-Nav). In a nutshell, RDMAE-Nav mainly consists of two modules: a visual module and a policy module. In the visual module, a self-supervised pretraining method, dubbed Regularized Denoising Masked AutoEncoders (RDMAE), is designed to enable the Vision Transformers (ViT)-based visual encoder to learn robust representations. The bidirectional Kullback-Leibler divergence is introduced in RDMAE as the regularization term for a denoising masked modeling task. Specifically, RDMAE mitigates the gap between clean and noisy image representations by minimizing the bidirectional Kullback-Leibler divergence. Then, the visual encoder is pretrained by RDMAE. In contrast to existing works, RDMAE-Nav applies denoising masked visual pretraining for PointGoal navigation to improve robustness to various visual corruptions. Finally, the pretrained visual encoder with frozen weights is applied to extract robust visual representations for policy learning in the RDMAE-Nav. Extensive experiments show that RDMAE-Nav performs competitively compared with state of the arts (SOTAs) on various visual corruptions. In detail, RDMAE-Nav performs the absolute improvement: 28.2% in SR and 23.68% in SPL under Spatter; 2.28% in SR and 6.41% in SPL under Speckle Noise; and 9.46% in SR and 9.55% in SPL under Defocus Blur.

Research and implementation of modulation recognition based on cascaded feature fusion

AbstractAiming at the problems of weak robustness of single feature and limited recognition range in modulation recognition, this paper proposes a modulation recognition algorithm based on cascaded feature fusion and multi‐classifier combination, in which time‐frequency map and instantaneous amplitude spectral density features are extracted from low intermediate frequency (IF) signal, constellation map and high‐order cumulant features are extracted from zero IF signal, and a three‐level recognition algorithm is designed through the decision fusion of decision tree, convolutional neural network, and support vector machine. In order to verify the performance of the algorithm, a modulation recognition system is designed and built based on USRP2974. The low IF and zero IF modulation signals are received through two channels, and the RF signals are received and processed in real time with the help of the built‐in CPU of the receiver. The recognition of 13 kinds of analog modulation and digital modulation signals is realized. Under the condition of wireless reception, the recognition rate of this system is more than 93% at 5 to 10 dB.

A double-branch graph convolutional network based on individual differences weakening for motor imagery EEG classification

The emergence of deep learning methods has driven the widespread use of brain–machine interface motor imagery classification in machine control and medical rehabilitation, and has achieved classification accuracy superior to those of traditional machine learning methods. However, models trained using current mainstream deep learning methods show a maximum variation in accuracy of over 20% when using data from different subjects in the same dataset for classification. The large variation indicates the weak robustness of such models and the difficulties in feature extraction for some subjects. As motor imagery classification is aimed at individual users, it is not conducive to the diffusion of the technique if the results vary too much from one user to another. In our research, we have found the accuracy differences between different subjects are caused by the data differ in spatial characteristics and training difficulty. Therefore, exploring the differences between different subjects’ data and weakening these differences can reduce the accuracy gap between subjects and ensure that the model can have good classification accuracy for each subject. We call this operation of reducing the accuracy gap individual differences weakening. To implement this operation, we propose a Double-branch Graph Convolutional Attention Neural Network (DGCAN), which uses a graph neural network to filter channels that are less disturbed by spatial location factors, and uses spatial–temporal domain convolution to focus on extracting features contained in the filtered channels, weakening the influence of spatial features contributes to individual differences weakening. We also design a loss function, EegLoss, which focuses on training hard samples and can effectively reduce the model-insensitive data contained in different subjects. We test model performance on the BCI Competition IV datasets 2a and 2b, achieving accuracies of 84% and 86%. We also compare the accuracy gap between subjects, showing that our model is effective in reducing the accuracy gap between subjects and has higher robustness than previous models.

Decoupled adaptive terminal sliding mode control strategy for a 6-DOF electro-hydraulic suspension test rig with RBF coupling force compensator

Aiming at the problems of poor tracking accuracy, low convergence speed, weak robustness of a 6-DOF electro-hydraulic suspension test system under system uncertainties and external disturbance, this article proposes double closed-loop control scheme. In the outer loop controller, this article designs a new decoupled adaptive fast nonsingular terminal sliding mode (DAFNTSM) control to achieve fast convergence rate, high accurate and adaptive disturbances estimation during trajectory tracking. Additionally, a coupling force compensator based on RBF neural network is employed, which can reduce online calculation quantity and guarantee accuracy. For six inner loop controllers, each hydraulic actuator can precisely follow the required force solved by outer loop controller. The hard-in-loop experiment results reveal that the control approach we proposed in this paper realize superior trajectory tracking characteristics in comparison with the conventional nonsingular terminal sliding mode (CNTSM) method.

RRIFLoc: Radio Robust Image Fingerprint Indoor Localization Algorithm Based on Deep Residual Networks

Indoor localization is one of the most exciting research areas due to the increasing demand for location-based services (LBSs) in indoor environments. The fingerprint positioning method of the received signal strength indicator (RSSI) is widely used in indoor localization due to its simple deployment and low cost. However, since the RSSI is affected by indoor environment changes and the heterogeneous nature of devices, it is easy to cause fingerprint drift and disappearance of fingerprint features, resulting in low accuracy and weak robustness of indoor localization. In this article, we propose a robust indoor localization method that is calibrated-free of Wi-Fi image fingerprints, called the radio robust image fingerprint localization (RRIFLoc) algorithm. First, the signal strength difference (SSD) fingerprint and RSSI kurtosis are derived from the RSSI fingerprint. SSD and kurtosis alleviate the low positioning accuracy and weak anti-interference caused by fingerprint drift and the disappearance of fingerprint features. Second, the fusion of the RSSI, SSD, and kurtosis is constructed into a radio robust image fingerprint (RRIF). Finally, we build the RRIFLoc model using the generated RRIF and the deep residual network for location estimation. According to experiments on a public dataset, our method reduces the average location estimation error by 56.87% compared to state-of-the-art indoor fingerprint localization methods.

Adaptive Optimal Motion Control of Uncertain Underactuated Mechatronic Systems With Actuator Constraints

Underactuated mechatronic systems are widely used in industrial production, where the control efforts and operation accuracy are both important aspects of performance evaluations. Hence, how to realize effective motion control, while reducing control efforts as much as possible, becomes an open problem for underactuated systems. Although some open loop approaches (e.g., trajectory planning) take energy optimization into account, they need linearization/approximation manipulations and exhibit weak robustness, which is prone to degrading practical control performance. To this end, this article proposes an adaptive tracking controller for uncertain multi-input-multi-output (MIMO) underactuated mechatronic systems, to fulfill accurate positioning/tracking control with saturated inputs and reduce control efforts as well. Particularly, by elaborately developing an auxiliary compensation term and a robust term, the proposed controller ensures <i>asymptotic convergence</i> of both <i>actuated</i> and <i>unactuated</i> variables. Meanwhile, the modified performance index function is approximated online and introduced into the Lyapunov function candidate to make the stability analysis process <i>more concise</i>. To the best of our knowledge, without the need of offline computation and the persistence of excitation (PE) condition, this article presents the <i>first</i> adaptive optimal controller to simultaneously achieve error elimination, control effort optimization, and actuator constraints for a class of underactuated systems. Finally, strict theoretical analysis and experimental validations show the effectiveness and robustness of the suggested controller.

Surface defect detection method for electronic panels based on attention mechanism and dual detection heads.

Automatic detection of surface defects in electronic panels is receiving increasing attention in the quality control of products. The surface defect detection of electronic panels is different from other target detection scenarios and is a meaningful and challenging problem. Its main manifestation is that surface defects of electronic panels usually exhibit extreme irregularity and small target characteristics, which bring great difficulties to the task of surface defect target detection including feature extraction and so on. The traditional methods can only detect a very small number of defect classes under specific detection conditions. And due to the weak robustness of these methods, they cannot be applied in real production scenarios on a large scale. Based on this, this paper applies the target detection technique under deep learning to the surface defect detection scenario of electronic panels for the first time. At the same time, in order to make the designed target detection network applicable to the electronic panel surface defect detection scenario and to enhance the interpretability of the designed target detection network in terms of computer mechanism. We design a deformable convolution module with a convolutional self-attentive module to learn the offset and a dual detection head incorporating the SE (Squeeze-and-Excitation) mechanism for the irregular characteristics of electronic panel surface defects and the small target characteristics, respectively. Finally, we carried out a series of experiments on our own electronic panel defect data set, including comparison with the most advanced target detection algorithms and a series of ablation experiments against our proposed method. The final experimental results prove that our method not only has better interpretability, but also has better metric performance, in which the map_0.5 metric reaches 78.257%, which is an increase of 13.506 percentage points over YOLOV5 and 33.457 percentage points higher than Retinanet. The results prove the proposed method is effective.

A Multi-Watermarking Algorithm for Medical Images Using Inception V3燼nd燚CT

Medical images are a critical component of the diagnostic process for clinicians. Although the quality of medical photographs is essential to the accuracy of a physician’s diagnosis, they must be encrypted due to the characteristics of digital storage and information leakage associated with medical images. Traditional watermark embedding algorithm embeds the watermark information into the medical image, which reduces the quality of the medical image and affects the physicians’ judgment of patient diagnosis. In addition, watermarks in this method have weak robustness under high-intensity geometric attacks when the medical image is attacked and the watermarks are destroyed. This paper proposes a novel watermarking algorithm using the convolutional neural networks (CNN) Inception V3 and the discrete cosine transform (DCT) to address above mentioned problems. First, the medical image is input into the Inception V3 network, which has been structured by adjusting parameters, such as the size of the convolution kernels and the typical architecture of the convolution modules. Second, the coefficients extracted from the fully connected layer of the network are transformed by DCT to obtain the feature vector of the medical image. At last, the watermarks are encrypted using the logistic map system and hash function, and the keys are stored by a third party. The encrypted watermarks and the original image features are performed logical operations to realize the embedding of zero-watermark. In the experimental section, multiple watermarking schemes using three different types of watermarks were implemented to verify the effectiveness of the three proposed algorithms. Our NC values for all the images are more than 90% accurate which shows the robustness of the algorithm. Extensive experimental results demonstrate the robustness under both conventional and high-intensity geometric attacks of the proposed algorithm.

Robustness Analysis of Platoon Control for Mixed Types of Vehicles

Currently, with the development of driving technologies, driverless vehicles gradually are becoming more and more available. Therefore, there would be a long period of time during which self-driving vehicles and human-driven vehicles coexist. However, for a mixed platoon, it is hard to control the formation due to the existence of the manual vehicles resulting in weak robustness and slow consensus rate on this system of platoons because of uncertainties caused by human factors for manual vehicles. In order to solve this problem, we establish models of mixed platoons with mixed types of connected and automated vehicles (CAVs), human-driven vehicles (HDVs) and HDVs without the vehicle awareness device (HDVWs). We subsequently design <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {H}_{\infty} $ </tex-math></inline-formula> controllers for the mixed platoons to realize the formation consensus. In addition, we use the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {H}_{\infty} $ </tex-math></inline-formula> norm of mixed platoons as the control objective investigating the robustness of the control algorithms in alleviating the platoon uncertainties. Furthermore, conditions are proved to maintain the stability of the mixed platoons, and the stability is analyzed based on the variation of the penetration rate of the manual vehicles. Finally, we formulate conditions for parameters according to the definition of string stability to avoid the collisions of vehicles. The results in this study are tested with simulations and suggest that the presented controllers can ensure the consensus of mixed platoons under uncertainties.

Dual-TBNet: Improving the Robustness of Speech Features via Dual-Transformer-BiLSTM for Speech Emotion Recognition

Speech emotion recognition has always been one of the topics that have attracted a lot of attention from many researchers. In traditional feature fusion methods, the speech features used only come from the data set, and the weak robustness of features can easily lead to overfitting of the model. In addition, these methods often use simple concatenation to fuse features, which will cause the loss of speech information. In this paper, to solve the above problems and improve the recognition accuracy, we utilize self-supervised learning to enhance the robustness of speech features and propose a feature fusion model(Dual-TBNet) that consists of two 1D convolutional layers, two Transformer modules and two bidirectional long short-term memory (BiLSTM) modules. Our model uses 1D convolution to take features of different segment lengths and dimension sizes as input, uses the attention mechanism to capture the correspondence between the two features, and uses the bidirectional time series module to enhance the contextual information of the fused features. We designed a total of four fusion models to fuse five pre-trained features and acoustic features. In the comparison experiments, the Dual-TBNet model achieved a recognition accuracy and F1 score of 95.7% and 95.8% on the CASIA dataset, 66.7% and 65.6% on the eNTERFACE05 dataset, 64.8% and 64.9% on the IEMOCAP dataset, 84.1% and 84.3% on the EMO-DB dataset and 83.3% and 82.1% on the SAVEE dataset. The Dual-TBNet model effectively fuses acoustic features of different lengths and dimensions with pre-trained features, enhancing the robustness of the features, and achieved the best performance.

Accurate identification and measurement of the precipitate area by two-stage deep neural networks in novel chromium-based alloys.

The performance of advanced materials for extreme environments is underpinned by their microstructure, such as the size and distribution of nano- to micro-sized reinforcing phase(s). Chromium-based superalloys are a recently proposed alternative to conventional face-centred-cubic superalloys for high-temperature applications, e.g., Concentrated Solar Power. Their development requires the determination of precipitate volume fraction and size distribution using Electron Microscopy (EM), as these properties are crucial for the thermal stability and mechanical properties of chromium superalloys. Traditional approaches to EM image processing utilise filtering with a fixed contrast threshold, leads to weak robustness to background noise and poor generalisability to different materials. It also requires an enormous amount of time for manual object measurements on large datasets. Efficient and accurate object detection and segmentation are therefore highly desired to accelerate the development of novel materials like chromium-based superalloys. To address these bottlenecks, based on YOLOv5 and SegFormer structures, this study proposes an end-to-end, two-stage deep learning scheme, DT-SegNet, to perform object detection and segmentation for EM images. The proposed approach can thus benefit from the training efficiency of CNNs at the detection stage (i.e., a small number of training images required) and the accuracy of the ViT at the segmentation stage. Extensive numerical experiments demonstrate that the proposed DT-SegNet significantly outperforms the state-of-the-art segmentation tools offered by Weka and ilastik regarding a large number of metrics, including accuracy, precision, recall and F1-score. This model forms a useful tool to aid alloy development microstructure examinations, and offers significant advantages to address the large datasets associated with high-throughput alloy development approaches.

Current Control Method of Vehicle Permanent Magnet Synchronous Motor Based on Active Disturbance Rejection Control

Due to the frequently changing working conditions and complex operating environment of electric vehicle permanent magnet synchronous motor(PMSM), the motor parameters change dramatically. However, the performance of the PI current regulator, which is the most widely used, is sensitive to motor parameters and has weak robustness, which will lead to the deterioration of motor control system performance. To address this problem, active disturbance rejection control (ADRC) technology is applied to the PMSM current loop control. Firstly, the traditional ADRC current regulator is designed, and the performance and parameter tuning laws of the extended state observer are analyzed by the method of frequency domain analysis. Then, the traditional ADRC algorithm is improved in three aspects: observation error compensation, utilization of model information and anti-windup. After that, simulations and bench test validation are performed. The simulation results show that the improved ADRC current regulator is more robust in the face of parameter changes. The torque step test results show that the improved ADRC current regulator has fast dynamic response without overshoot and has high robustness when the motor parameters change. The dynamic test results show that the improved ADRC current regulator has high robustness when the load, speed and motor parameters change, and the anti-windup measures designed can effectively overcome the integral saturation phenomenon.

Unsupervised medical image feature learning by using de-melting reduction auto-encoder

Unsupervised feature learning is a fundamental and highly prioritized problem in medical image analysis. Although it has shown considerable improvements, it remains challenging because of its weak feature expression ability, low model-learning efficiency, and weak robustness. To address these limitations, a novel unsupervised feature learning method in the medical image classification task, named de-melting reduction auto-encoder (DMRAE), is proposed in this study. A joint fusion network structure is constructed; it not only improves the expression of target features but also reduces the loss of feature decoding and parameters. To obtain a robust solution, a newly designed decomposed-reconstructed loss function is used to strengthen the semantic context between adjacent feature extractor layers, successfully avoiding the insufficient model-learning ability from the single optimization objective and improving the quality of the extracted features. Finally, extensive experiments on datasets consisting of 400 breast ultrasonographic images and 6000 lung computed tomography images are conducted to demonstrate the effectiveness of the proposed method. Experimental results reveal that the DMRAE significantly reduces the annotation effort and outperforms existing methods by a significant margin.

A Novel Capsule Network with Attention Routing to Identify Prokaryote Phosphorylation Sites.

By denaturing proteins and promoting the formation of multiprotein complexes, protein phosphorylation has important effects on the activity of protein functional molecules and cell signaling. The regulation of protein phosphorylation allows microbes to respond rapidly and reversibly to specific environmental stimuli or niches, which is closely related to the molecular mechanisms of bacterial drug resistance. Accurate prediction of phosphorylation sites (p-site) of prokaryotes can contribute to addressing bacterial resistance and providing new perspectives for developing novel antibacterial drugs. Most existing studies focus on human phosphorylation sites, while tools targeting phosphorylation site identification of prokaryotic proteins are still relatively scarce. This study designs a capsule network-based prediction technique for p-site in prokaryotes. To address the poor scalability and unreliability of dynamic routing processes in the output space of capsule networks, a more reliable way is introduced to learn the consistency between capsules. We incorporate a self-attention mechanism into the routing algorithm to capture the global information of the capsule, reducing the computational effort while enriching the representation capability of the capsule. Aiming at the weak robustness of the model, EcapsP improves the prediction accuracy and stability by introducing shortcuts and unconditional reconfiguration. In addition, the study compares and analyzes the prediction performance based on word vectors, physicochemical properties, and mixing characteristics in predicting serine (Ser/S), threonine (Thr/T), and tyrosine (Tyr/Y) p-site. The comprehensive experimental results show that the accuracy of the developed technique is close to 70% for the identification of the three phosphorylation sites in prokaryotes. Importantly, in side-by-side comparisons with other state-of-the-art predictors, our method improves the Matthews correlation coefficient (MCC) by approximately 7%. The results demonstrate the superiority of EcapsP in terms of high performance and reliability.

Robust interval type-2 kernel-based possibilistic fuzzy clustering algorithm incorporating local and non-local information

Type-2 fuzzy set theory has certain potential advantage in processing high-order uncertainty, and has been more and more widely applied in pattern recognition, image processing, and system modeling. Although it has been introduced into robust fuzzy clustering modeling, most clustering methods based on type-2 fuzzy set theory have many obvious shortcomings, such as weak robustness against noise and high sensitivity to initial value. Therefore, this paper presents a novel single fuzzifier interval type-2 kernel-based possibilistic fuzzy local and non-local information c-means clustering, and it is driven by deep neighborhood information for image segmentation in the presence of high noise. Firstly, we construct the novel deep neighborhood window structure, which consists of local neighborhood window around current pixel and deep neighborhood window around the pixel in local neighborhood window around current pixel. Secondly, using the structural similarity between local neighborhood window and deep neighborhood window around current pixel, we construct a novel fuzzy local information factor to tune the impact of local neighborhood information on current pixel clustering. Thirdly, possibility theory and fuzzy local information factor are introduced into interval type-2 kernel-based fuzzy c-means clustering, and a novel single fuzzifier type-2 possibilistic fuzzy clustering with richer local and non-local information and kernel metric is proposed. Experimental results indicate that the proposed algorithm outperforms existing state-of-the-art robust fuzzy clustering-related algorithms, and significantly improve the robustness of robust fuzzy clustering algorithms against noise.

A deep generative model based on CNN-CVAE for wind turbine condition monitoring

Auto-encoder (AE)-based condition monitoring (CM) methods for fault detection of wind turbines have received considerable attention due to their powerful feature extraction ability. However, traditional AE-based monitoring methods can only learn point-to-point features by minimizing reconstruction errors, which leads to a low sensitivity to anomaly data and weak robustness to noise data. To this end, we introduce a novel deep generative method based on the convolutional neural network (CNN)-conditional variational auto-encoder (CVAE). The key idea of CNN-CVAE is to unify the representation learning capacity of the CVAE and CNN. Specifically, CVAE can learn a probability distribution model by being trained on an anomaly-free supervisory control and data acquisition systems (SCADA) dataset; CNN and deconvolution operations are adopted for better time-series feature extraction and reconstruction performance. A statistical process control chart is applied to determine the alarm threshold. The effectiveness of the CNN-CVAE-based method is validated by datasets collected by SCADA installed in a commercial wind farm in China for impending blade breakage and gearbox failure. Abundant experiments with state-of-the-art deep learning-based CM methods are conducted, which indicate that our proposed method outperforms other methods in robustness, fault detection data sensitivity, fault warning time, and model parameters.

Fragile plant-frugivore interaction networks in tropical forest edges

Current incessant threats of environmental changes, such as habitat fragmentation, may disrupt vital mutualistic plant-frugivore networks, threatening the whole ecosystem. To advance our knowledge of the magnitude of the impacts of such disruption, we need a better understanding of its structure and robustness. We addressed this by comparing several aspects of plant-frugivore networks in the forest edge and interior habitats of a fragmented forest, using field data on fruit removal by a community of frugivorous birds and lemurs from 40 tree species over 30 months. While the plant-frugivore networks in both habitats were significantly less nested than random expectations, the network in the forest edge habitat had higher value of nestedness compared to the network in the forest interior. Also, the networks in both habitats were highly modular compared to random expectations, with a dominance of species that are highly interactive within its modules. We also found that fruit size, weight, and color as well as tree height predicted the structural role of plants in the networks and their degree of specialization. Finally, the removal/loss of best-connected plants or animals resulted in a weak robustness of the networks in both habitats; this vulnerable pattern is more pronounced in forest edges than interior habitats. Interestingly, the loss of large-bodied seed dispersers from either habitat did not affect much the robustness of the networks; in contrast, when small-sized species were eliminated first, the network became less robust, especially for the edge community. These findings highlight the need to preserve species interactions in forest edges to avoid the collapse of the network, mitigating the impacts of species extinctions on ecosystem equilibrium.

3QNet

Since the development of 3D applications, the point cloud, as a spatial description easily acquired by sensors, has been widely used in multiple areas such as SLAM and 3D reconstruction. Point Cloud Compression (PCC) has also attracted more attention as a primary step before point cloud transferring and saving, where the geometry compression is an important component of PCC to compress the points geometrical structures. However, existing non-learning-based geometry compression methods are often limited by manually pre-defined compression rules. Though learning-based compression methods can significantly improve the algorithm performances by learning compression rules from data, they still have some defects. Voxel-based compression networks introduce precision errors due to the voxelized operations, while point-based methods may have relatively weak robustness and are mainly designed for sparse point clouds. In this work, we propose a novel learning-based point cloud compression framework named 3D Point Cloud Geometry Quantiation Compression Network (3QNet), which overcomes the robustness limitation of existing point-based methods and can handle dense points. By learning a codebook including common structural features from simple and sparse shapes, 3QNet can efficiently deal with multiple kinds of point clouds. According to experiments on object models, indoor scenes, and outdoor scans, 3QNet can achieve better compression performances than many representative methods.