High-Precision Digital Surface Model Extraction from Satellite Stereo Images Fused with ICESat-2 Data

The lack of ground control points (GCPs) affects the elevation accuracy of digital surface models (DSMs) generated by optical satellite stereo images and limits the application of high-resolution DSMs. It is a feasible idea to use ICESat-2 (Ice, Cloud, and land Elevation Satellite-2) laser altimetry data to improve the elevation accuracy of optical stereo images, but it is necessary to accurately match the two types of data. This paper proposes a DSM registration strategy based on terrain similarity (BOTS), which integrates ICESat-2 laser altimetry data without GCPs and improves the DSM elevation accuracy generation from optical satellite stereo pairs. Under different terrain conditions, Worldview-2, SV-1, GF-7, and ZY-3 stereo pairs were used to verify the effectiveness of this method. The experimental results show that the BOTS method proposed in this paper is more robust when there are a large number of abnormal points in the ICESat-2 data or there is a large elevation gap between DSMs. After fusion of ICESat-2 data, the DSM elevation accuracy extracted from the satellite stereo pair is improved by 73~92%, and the root mean square error (RMSE) of Worldview-2 DSM reaches 0.71 m.

Map Construction Based on LiDAR Vision Inertial Multi-Sensor Fusion

In order to make up for the shortcomings of independent sensors and provide more reliable estimation, a multi-sensor fusion framework for simultaneous localization and mapping is proposed in this paper. Firstly, the light detection and ranging (LiDAR) point cloud is screened in the front-end processing to eliminate abnormal points and improve the positioning and mapping accuracy. Secondly, for the problem of false detection when the LiDAR is surrounded by repeated structures, the intensity value of the laser point cloud is used as the screening condition to screen out robust visual features with high distance confidence, for the purpose of softening. Then, the initial factor, registration factor, inertial measurement units (IMU) factor and loop factor are inserted into the factor graph. A factor graph optimization algorithm based on a Bayesian tree is used for incremental optimization estimation to realize the data fusion. The algorithm was tested in campus and real road environments. The experimental results show that the proposed algorithm can realize state estimation and map construction with high accuracy and strong robustness.

SELF-SUPERVISED ACOUSTIC ANOMALY DETECTION VIA CONTRASTIVE LEARNING

We propose an acoustic anomaly detection algorithm based on the framework of contrastive learning. Contrastive learning is a recently proposed self-supervised approach that has shown promising results in image classification and speech recognition. However, its application in anomaly detection is underexplored. Earlier studies have demonstrated that it can achieve state-of-the-art performance in image anomaly detection, but its capability in anomalous sound detection is yet to be investigated. For the first time, we propose a contrastive learning-based framework that is suitable for acoustic anomaly detection. Since most existing contrastive learning approaches are targeted toward images, the effect of other data transformations on the performance of the algorithm is unknown. Our framework learns a representation from unlabeled data by applying audio-specific data augmentations. We show that in the resulting latent space, normal and abnormal points are distinguishable. Experiments conducted on the MIMII dataset confirm that our approach can outperform competing methods in detecting anomalies.

SELF-SUPERVISED ACOUSTIC ANOMALY DETECTION VIA CONTRASTIVE LEARNING

We propose an acoustic anomaly detection algorithm based on the framework of contrastive learning. Contrastive learning is a recently proposed self-supervised approach that has shown promising results in image classification and speech recognition. However, its application in anomaly detection is underexplored. Earlier studies have demonstrated that it can achieve state-of-the-art performance in image anomaly detection, but its capability in anomalous sound detection is yet to be investigated. For the first time, we propose a contrastive learning-based framework that is suitable for acoustic anomaly detection. Since most existing contrastive learning approaches are targeted toward images, the effect of other data transformations on the performance of the algorithm is unknown. Our framework learns a representation from unlabeled data by applying audio-specific data augmentations. We show that in the resulting latent space, normal and abnormal points are distinguishable. Experiments conducted on the MIMII dataset confirm that our approach can outperform competing methods in detecting anomalies.

Intrusion Detecting System Based on Temporal Convolutional Network for In-Vehicle CAN Networks

In recent years, deep learning theories, such as Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN), have been applied as effective methods for intrusion detection in the vehicle CAN network. However, the existing RNNs realize detection by establishing independent models for each CAN ID, which are unable to learn the potential characteristics of different IDs well, and have relatively complicated model structure and high calculation time cost. CNNs can achieve rapid detection by learning the characteristics of normal and attack CAN ID sequences and exhibit good performance, but the current methods do not locate abnormal points in the sequence. To solve the above problems, this paper proposes an in-vehicle CAN network intrusion detection model based on Temporal Convolutional Network, which is called Temporal Convolutional Network-Based Intrusion Detection System (TCNIDS). In TCNIDS, the CAN ID is serialized into a natural language sequence and a word vector is constructed for each CAN ID through the word embedding coding method to reduce the data dimension. At the same time, TCNIDS uses the parameterized Relu method to improve the temporal convolutional network, which can better learn the potential features of the normal sequence. The TCNIDS model has a simple structure and realizes the point anomaly detection at the message level by predicting the future sequence of normal CAN data and setting the probability strategy. The experimental results show that the overall detection rate, false alarm rate, and accuracy rate of TCNIDS under fuzzy attack, spoofing attack, and DoS attack are higher than those of the traditional temporal convolutional network intrusion detection model.

Improved sensitivity of microperimetric outcomes for clinical studies in age-related macular degeneration

To investigate sensitive outcome measures based exclusively on abnormal points in microperimetry testing of eyes with intermediate age-related macular degeneration (iAMD). 25 eyes of 25 subjects with iAMD had undergone 2 successive tests of mesopic microperimetry with the Macular Integrity Assessment Microperimeter (MAIA), using a custom grid of 33 points spanning the central 14 degrees of the macula. Each point was defined as abnormal if its threshold sensitivity was lower than 1.65 standard deviations (SD) (5%) or 2 SD (2.5%) than the expected threshold in healthy eyes according to the MAIA internal database. Among the 25 eyes there were 11.8 ± 9 and 8.4 ± 8.2 abnormal points at < 5% and < 2.5%, with mean deviation of thresholds from normal − 4.9 ± 1.2 dB and − 5.8 ± 1.5 dB, respectively. These deviations were greater, and their SD smaller, compared with the complete microperimetry grid, − 2.3 ± 2.0 dB. The 95% limits of agreement for average threshold between the 2 successive tests were smaller when only abnormal points were included. Analyzing only abnormal grid points yields an outcome parameter with a greater deviation from normal, a more homogenous dataset, and better test–retest variability, compared with analysis of all grid points. This parameter may thus be more sensitive to change, while moderately limiting the number of potential recruits. The proposed outcome measures should be further investigated as potential endpoints in clinical trials in iAMD.

A detection and restoration approach for vessel trajectory anomalies based on AIS

In view of the continuous increase in the amount of AIS data at sea and the existence of more abnormal points, it is difficult to construct ship trajectories based on AIS data. Aiming at this problem, a new method for identifying and repairing abnormal points in trajectories only based the AIS data of the ship itself is proposed. Longitude and latitude, speed, acceleration, direction and other parameters are comprehensively used to identify and repair the abnormal points in the method proposed. Compared with the methods based on single location data, it can effectively reduce the missed judgement of outliers. Compared with the methods based on trajectories clustering to judge singular point, this method does not require the data of historical trajectories to expand the application scope. The cubic spline method is used to interpolate points for the discontinuous segments to further improve the continuity and integrity of the ship trajectory. The results of AIS data processing and analysis on ships in actual sea areas verify the feasibility and effectiveness of the proposed method.

Real-Time Monitoring and Reminding of Remote Peritoneal Dialysis System Based on the Principle of Least Squares

Peritoneal dialysis has been widely studied and applied for kidney disease because of its low cost and easy operation. Given the development of chronic kidney disease worldwide, peritoneal dialysis has attracted more and more attention. At the same time, with the development and popularization of mobile network technology, mobile telematics has begun to become a mainstream trend. By integrating the experience of clinicians, the remote diagnosis and treatment system of the peritoneal dialysis developed by Shenzhen Traditional Chinese Medicine Hospital can monitor the entire peritoneal dialysis data of patients. The peritoneal dialysis data were analyzed by statistical methods. In this paper, we designed a data acquisition device with Bluetooth transmission protocol and a user APP to collect peritoneal dialysis data from experimental patients, and built a regression model based on the least square principle according to the clinical data of real patients. Through the model, abnormal or discrete points can be identified in real time. In clinical practice, by analyzing the possible medical risks and adverse events of patients according to the abnormal points, we realize the function of prediction and early reminding. The system indicates the results to patients according to the confidence interval of regression prediction, which greatly strengthens the interaction of the system and improves patient compliance.

Optimization of Abnormal Point Cloud Recognition in Robot Vision Grinding System Based on Multidimensional Improved Eigenvalue Method (MIEM)

To improve the accurate and sufficient recognition of abnormal points on the workpiece, a multidimensional anomaly point identification approach based on an improved eigenvalue method is proposed in this paper. Whether a point is normal or not depends on the angle between the two adjacent vectors which consisted of four adjacent points around the current focus. The comprehensive judgment is carried out by multidimensional approximation. The numerical simulation and actual experiment validate the efficiency of the proposed method to quickly and accurately identify the abnormal point cloud in the surface point cloud data.

System load trend prediction method based on IF-EMD-LSTM

Accurately predicting the load change of the information system during operation has important guiding significance for ensuring that the system operation is not interrupted and resource scheduling is carried out in advance. For the information system monitoring time series data, this article proposes a load trend prediction method based on isolated forests-empirical modal decomposition-long-term (IF-EMD-LSTM). First, considering the problem of noise and abnormal points in the original data, the isolated forest algorithm is used to eliminate the abnormal points in the data. Second, in order to further improve the prediction accuracy, the empirical modal decomposition algorithm is used to decompose the input data into intrinsic mode function (IMF) components of different frequencies. Each intrinsic mode function (IMF) and residual is predicted using a separate long-term and short-term memory neural network, and the predicted values are reconstructed from each long-term and short-term memory model. Finally, experimental verification was carried out on Amazon’s public data set and compared with autoregressive integrated moving average and Prophet models. The experimental results show the superior performance of the proposed IF-EMD-LSTM prediction model in information system load trend prediction.