Adjustment Module Research Articles

Robust Template Adjustment Siamese Network for Object Visual Tracking.

Most of the existing trackers address the visual tracking problem by extracting an appearance template from the first frame, which is used to localize the target in the current frame. Unfortunately, they typically face the model degeneration challenge, which easily results in model drift and target loss. To address this issue, a novel Template Adjustment Siamese Network (TA-Siam) is proposed in this paper. The proposed framework TA-Siam consists of two simple subnetworks: The template adjustment subnetwork for feature extraction and the classification-regression subnetwork for bounding box prediction. The template adjustment module adaptively uses the feature of subsequent frames to adjust the current template. It makes the template adapt to the target appearance variation of long-term sequence and effectively overcomes model drift problem of Siamese networks. In order to reduce classification errors, the rhombus labels are proposed in our TA-Siam. For more efficient learning and faster convergence, our proposed tracker uses a more effective regression loss in the training process. Extensive experiments and comparisons with trackers are conducted on the challenging benchmarks including VOT2016, VOT2018, OTB50, OTB100, GOT-10K, and LaSOT. Our TA-Siam achieves state-of-the-art performance at the speed of 45 FPS.

System integration for predictive process adjustment and cloud computing-based real-time condition monitoring of vibration sensor signals in automated storage and retrieval systems

As automation and digitalization are being increasingly implemented in industrial applications, manufacturing systems comprising several functions are becoming more complex. Consequently, fault analysis (e.g., fault detection, diagnosis, and prediction) has attracted increased research attention. Investigations involving fault analysis are usually performed using real-time, online, or automated techniques for fault detection or alarming. Conversely, recovery of faulty states to their healthy forms is usually performed manually under offline conditions. However, the development of intelligent systems requires that appropriate feedback be provided automatically, to facilitate faulty-state recovery without the need for manual operator intervention and/or decision-making. To this end, this paper proposes a system integration technique for predictive process adjustment that determines appropriate recovery actions and performs them automatically by analyzing relevant sensor signals pertaining to the current situation of a manufacturing unit via cloud computing and machine learning. The proposed system corresponds to an automated predictive process adjustment module of an automated storage and retrieval system (ASRS). The said integrated module collects and analyzes the temperature and vibration signals of a product transporter using an internet-of-things-based programmable logic controller and cloud computing to identify the current states of the ASRS system. Upon detection of faulty states, the control program identifies corresponding process control variables and controls them to recover the system to its previous no-fault state. The proposed system will facilitate automatic prognostics and health management in complex manufacturing systems by providing automatic fault diagnosis and predictive recovery feedback.

Adaptive spatial scale person reidentification

Person reidentification (ReID) requires the discriminative features of an entire pedestrian image to handle the problems of inaccurate person bounding box detection, background confusion, and occlusion. Many recent person ReID methods have attempted to learn the feature information of an entire pedestrian image through parts feature representations, but it is often too time consuming. Person ReID relies on discriminative pedestrian features, and different spatial scales can distinguish features by differing degrees. We propose an innovative and effective adaptive spatial scale person ReID network model based on the residual neural network (ResNet) of an instance batch normalization. Through experimental visualizations, pedestrian features extracted by ResNet from four layers are analyzed, and two layers with discriminative features are selected. Using an adaptive dimension adjustment module, different spatial scale features are aggregated and merged by the aggregation layer. To effectively learn spatial channel correlations and prevent overfitting, a multilayer distribution normalization processing module is designed to implement end-to-end training and evaluate the person ReID networks. Compared with other methods, this network model showed excellent performance on four public person ReID datasets and is superior to most current methods.

Dynamic Modification Neural Network Model for Short-term Traffic Prediction

Short-term precise prediction of network traffic can effectively help operators reasonably allocate network resources, improve network service quality and scheduling efficiency. Our model proposed in this paper uses dynamic parameters to adjust the predicted value and refit the predicted sequence for short-term traffic prediction which is called Dynamic Modification Neural Network (DMNN) model. The architecture of our model consists of a prediction module, an adjustment module and a series re-fitting module. The prediction module learns time closeness of the traffic data before the predicted point. The adjustment module takes dynamic period characteristics of data into account and generates adjusted value by linear discrete dynamic parameters. The series re-fitting module refits predicted series from the first two modules with a new hybrid loss function. In this paper, we take different neural networks in the prediction module, such as Long Short-Term Memory (LSTM), Bi-directional Long Short-Term Memory (BiLSTM), Convolutional Long Short-Term Memory (ConvLSTM) and Convolutional Bi-directional Long Short-Term Memory (ConvBiLSTM). We evaluate the performance of the proposed model by using two real-world datasets for short-term traffic flow prediction. Moreover, Experimental results show that the proposed model with dynamic modification has much higher accuracy than other models and decreases the prediction value error and time skew of the inflection points.

Adaptive Component Discrimination Network for Airplane Detection in Remote Sensing Images

Airplane detection and recognition in the high-resolution remote sensing images (RSIs) remain a challenging task due to the factors of multiple view angles, multiple scales, multiple orientations, etc. This article proposes an adaptive component discrimination network for airplane detection and recognition in RSIs, which focuses on various scales from global to local, making full use of the overall contour as well as the dominant component features of airplanes. First, a standardization processing module is proposed for image projection conversion and resolution uniform to alleviate the confusion of different types of airplanes in different resolution images. Second, the rotatable boundingbox-based pyramid network is utilized to extract candidate airplane coordinates and categories. Furthermore, an adaptive aircraft component discrimination method is established for confusing few-shot airplane targets recognition, which consists of a target orientation adaptive adjustment module (OAAM) and a component discrimination module (CDM). OAAM obtains airplanes with the same orientations by predicting the orientation of the slices and rotating them adaptively. All the uniformed slices are then fed into the CDM for dominant components detection, which corrects the target preclassification results, improving the classification performance. Experiments conducted on the 2020 Gaofen Challenge demonstrate the efficacy and superiority of the proposed method.

Application of Monocular Direct Vision Odometry in Augmented Reality

In recent years, the unlabeled augmented reality system has been gradually applied to various mobile devices, among which stable, accurate, and fast registration is the key to realizing this function. For this technique, this paper introduces camera exposure parameters and puts the data association and pose estimation into a unified nonlinear optimization problem. Moreover, the direct monocular vision odometer is transplanted into the augmented reality system through the position adjustment module. We compare it with the traditional visual odometry method that matches the feature points. The results show that this improved method can be used to track more quickly and build a more visual semi-dense point cloud map, which can be used to support the registration and tracking of virtual objects in augmented reality.

Design and Experiment of Assistive Mechanism for Adjustment of Transrectal Ultrasound Probe.

Transrectal ultrasound prostate biopsy is the most commonly used method for the diagnosis of prostate cancer. During the operation, the doctor needs to manually adjust the ultrasound probe for repeated adjustments, which is difficult to ensure the efficiency, accuracy, and safety of the operation. This paper presents a passive posture adjusting mechanism of transrectal ultrasound probe. The overall mechanism has 7 degrees of freedom, consisting of a position adjustment module, a posture adjustment module, and an ultrasonic probe rotation and feed module. In order to achieve the centering function, the posture adjustment module is designed based on the double parallelogram. Centering performance is verified based on SimMechanics, and remote center point error of physical prototypes is evaluated. The maximum error of the azimuth remote center point motion and the maximum error of the remote center point motion of the ultrasonic probe are 4 mm and 3.4 mm, respectively, which are less than the anus that can withstand 6 mm. Meanwhile, the analysis of measurement error shows that the random error correlation is weak in different directions, the systematic error confidence intervals of azimuth and elevation angle are less than 2.5 mm, and the maximum relative fixed point error and the maximum relative standard error are 14.73% and 14.98%, respectively. The simulation and testing results have shown the effectiveness and reliability of the propose mechanism.

Design and experimental study of a novel 7-DOF manipulator for transrectal ultrasound probe.

Traditional hand-held ultrasound probe has some limitations in prostate biopsy. Improving the localization and accuracy of ultrasound probe will increase the detection rate of prostate cancer while biopsy techniques remain unchanged. This paper designs a manipulator for transrectal ultrasound probe, which assists doctors in performing prostate biopsy and improves the efficiency and accuracy of biopsy procedure. The ultrasound probe manipulator includes a position adjustment module that can lock four joints at the same time. It reduces operating time and improves the stability of the mechanism. We use the attitude adjustment module designed by double parallelogram RCM mechanism, the ultrasound probe can realize centering and prevent its radial motion. The self-weight balance design helps doctors operate ultrasound probe without weight. Using MATLAB to analyze the manipulator, the results show that the workspace of the mechanism can meet the biopsy requirements. And simulate the centering effect of the ultrasound probe when the attitude is adjusted at different feeding distances, the results show that the ultrasound probe is centering stability. Finally, the centering and joint interlocking tests of the physical prototype are completed. In this paper, a 7-DOF manipulator for transrectal ultrasound probe is designed. The mechanism is analyzed for kinematics, workspace analysis, simulation of centering effects, development of a physical prototype and related experimental research. The results show that the surgical demand workspace is located inside the reachable workspace of the mechanism and the joint locking of the manipulator is reliable.

Enhancement in the World of Artificial Intelligence

Artificial Intelligence is a developing zone in the field of innovation and furthermore attempts to show that the eventual fate of AI gains ground so that machines would function according to a human and would likewise convey the action of the person. It is difficult to create a machine like individuals who can appear feelings or think like individuals in different conditions. Directly we have recognized that AI is the examination of how to form things which can accurately fill in as people do. A working framework that utilizes AI reasoning procedures has a computerized reasoning motor, and experience scientific and Statistical module, an adjustment module and a UI. The computerized reasoning motor processes an accomplished expository boundary from a front code and a back code. The experience of scientific and Statistical module records and changes the experience's systematic boundary. The alteration module changes the front code and the back code as per the consequence of the experience logical and Statistical module computation of the experience systematic boundary. The UI inputs information or showcases the consequence of the computation. In the man-made consciousness motor, the experience diagnostic boundary is then again added to either the front code or the back code to register another experience investigative boundary. Such a game plan, the working framework can consequently change the consequence of the computation as per the decision or past decisions of the client.

Ultra-broadband terahertz absorber based on a multilayer graphene metamaterial

An ultra-broadband absorber is introduced based on a multilayer graphene structure. Wave absorption of the optimized device is >90% in the bandwidth between 1.12 THz and 3.78 THz. Electric field distributions are studied to analyze the working mechanism. Because of the symmetrical structure, the designed absorber is not sensitive to the polarization state of the incident wave under small angles. The absorber has a reasonable tolerance for incident angles. The working intensity and bandwidth of the absorber are modulated with a wide range of tunability by different Fermi energy levels of graphene. This kind of ultra-broadband terahertz absorbers could have potential applications in a variety of areas, such as terahertz imaging, anti-radar stealth, and electrically adjustable modulation.

A Multiscale-Based Adjustable Convolutional Neural Network for Multiple Organ Segmentation

Accurate segmentation ofs organs-at-risk (OARs) in computed tomography (CT) is the key to planning treatment in radiation therapy (RT). Manually delineating OARs over hundreds of images of a typical CT scan can be time-consuming and error-prone. Deep convolutional neural networks with specific structures like U-Net have been proven effective for medical image segmentation. In this work, we propose an end-to-end deep neural network for multiorgan segmentation with higher accuracy and lower complexity. Compared with several state-of-the-art methods, the proposed accuracy-complexity adjustment module (ACAM) can increase segmentation accuracy and reduce the model complexity and memory usage simultaneously. An attention-based multiscale aggregation module (MAM) is also proposed for further improvement. Experiment results on chest CT datasets show that the proposed network achieves competitive Dice similarity coefficient results with fewer float-point operations (FLOPs) for multiple organs, which outperforms several state-of-the-art methods.

Parallel Dual Networks for Visual Object Tracking

Visual Object Tracking plays an essential role in solving many basic problems in computer vision. In order to improve the tracking accuracy, the previous methods have prevented tracking failures from occurring by improving the ability to describe the target. However, few of them consider ways to relocate and track the target after a tracking failure. In this paper, we propose the use of a parallel dual network for visual object tracking. This is constructed from two networks and an adjustment module to enable judgement of tracking failures, as well as target relocation and tracking. Firstly, we employ the Siamese matching method and correlation filter method to build tracking network and inspecting network. Both networks track the target simultaneously to obtain two tracking results. Secondly, an adjustment module is constructed, which compares the overlap ratio of the two tracking results with a set threshold, then fuses them or selects the best one. Finally, the fusion or selection result is output and the tracker is updated. We perform comprehensive experiments on five benchmarks: VOT2016, UAV123, Temple Color-128, OTB-100 and OTB-50. The results demonstrate that, compared with other state-of-the-art algorithms, the proposed tracking method improves tracking precision while maintaining real-time performance.

Ligand-Induced Degradation of a CAR Permits Reversible Remote Control of CAR T Cell Activity In Vitro and In Vivo

Chimeric antigen receptor (CAR)-modified Tcells are endowed with novel antigen specificity and are most often administered to patients without an engineered mechanism to control the CAR Tcells once infused. "Suicide switches" such as the small molecule-controlled, inducible caspase-9 (iCas9) system afford the ability to selectively eliminate engineered Tcells; however, these approaches are designed for all-or-none, irreversible termination of an ongoing immune response. In order to permit reversible and adjustable modulation, we have created a CAR that is capable of on-demand downregulation by fusing the CAR to a previously developed ligand-induced degradation (LID) domain. Addition of a small molecule ligand triggers exposure of a cryptic degron within the LID domain, resulting in proteasomal degradation of the CAR-LID fusion protein andloss of CAR on the surface of Tcells. This fusion construct allowed for reversible and "tunable" inhibition of CAR Tcell activity invitro. Delivery of the triggering molecule in CAR-LID-treated tumor-bearing mice temporarily reduced CAR activity through modulation of CAR surface expression. The ability to more flexibly modulate CAR Tcell expression through a small molecule provides a platform for controlling possible adverse side effects, as well as preclinical investigations of CAR Tcell biology.

A Multimodal, Adjustable Sensitivity, Digital 3-Axis Skin Sensor Module.

This paper presents major improvements to a multimodal, adjustable sensitivity skin sensor module. It employs a geomagnetic 3-axis Hall effect sensor to measure changes in the position of a magnetic field generated by an electromagnet. The electromagnet is mounted on a flexible material, and different current values can be supplied to it, enabling adjustments to the sensitivity of the sensor during operation. Capacitive sensing has been added in this iteration of the module, with two sensing modalities: “pre-touch” detection with proximity sensing and normal force capacitive sensing. The sensor has been designed to be interconnected with other sensor modules to be able to cover large surfaces of a robot with normal and shear force sensing and object proximity detection. Furthermore, this paper introduces important size reductions of the previous sensor design, calibration results, and further analysis of other sensor characteristics.

Stepwise PathNet: a layer-by-layer knowledge-selection-based transfer learning algorithm

Some neural network can be trained by transfer learning, which uses a pre-trained neural network as the source task, for a small target task’s dataset. The performance of the transfer learning depends on the knowledge (i.e., layers) selected from the pre-trained network. At present, this knowledge is usually chosen by humans. The transfer learning method PathNet automatically selects pre-trained modules or adjustable modules in a modular neural network. However, PathNet requires modular neural networks as the pre-trained networks, therefore non-modular pre-trained neural networks are currently unavailable. Consequently, PathNet limits the versatility of the network structure. To address this limitation, we propose Stepwise PathNet, which regards the layers of a non-modular pre-trained neural network as the module in PathNet and selects the layers automatically through training. In an experimental validation of transfer learning from InceptionV3 pre-trained on the ImageNet dataset to networks trained on three other datasets (CIFAR-100, SVHN and Food-101), Stepwise PathNet was up to 8% and 10% more accurate than finely tuned and from-scratch approaches, respectively. Also, some of the selected layers were not supported by the layer functions assumed in PathNet.

Analysis on energy consumption of blended-wing-body underwater glider

As a new type of underwater observation platform, underwater glider is widely used in marine explorations and military surveys, and most gliders are powered by their own batteries whose capacities are limited. It is therefore necessary to analyze the energy consumption of underwater glider. In this article, the variation law of seawater density changing with depth is considered; based on the theory of rigid body dynamics, the motion model of blended-wing-body underwater glider is established; the energy consumption model of each component module is accounted by analyzing the energy consumption composition in the working process of blended-wing-body underwater glider; and the energy consumption under different navigation depths, different glide ratios, and different buoyancy adjustments regulation is simulated. The results demonstrate that as the glide depth is increased, the total energy consumption increases in a single cycle and decreases per gliding distance, leading to a smaller energy consumption ratio for the attitude adjustment module; on the other hand, as the buoyancy adjustment is increased, more energy is consumed in a single cycle and less energy is consumed per gliding distanced, resulting in a larger energy consumption ratio for the attitude adjustment module. As the glide ratio increases, the total energy consumption in a single cycle first increases and then decreases, while the energy consumption of per gliding distance and the energy consumption ratio of the attitude adjustment module are decreased.

Active modulation of intracavity laser intensity with the Pound-Drever-Hall locking for photoacoustic spectroscopy.

Here we report a novel, to the best of our knowledge, method of active intracavity intensity modulation for cavity-enhanced photoacoustic spectroscopy (PAS) without the need for any external optical modulators. Based on the Pound-Drever-Hall (PDH) locking technique, a dither is added to the PDH error signal to periodically vary the locking point between the laser frequency and optical cavity within a sub-MHz frequency range. While significantly enhancing the intracavity laser intensity, the optical cavity also acts as an intensity modulator. As a proof-of-principle, we demonstrated the PAS of ${{\rm C}_2}{{\rm H}_2}$C2H2 by placing a photoacoustic cell ($Q$Q-factor $\sim{10}$∼10) inside a Fabry-Perot cavity (finesse $\sim{628}$∼628) and adopting the proposed intracavity intensity modulation scheme. By detecting the weak ${{\rm C}_2}{{\rm H}_2}$C2H2 line at ${6412.73}\;{{\rm cm}^{ - 1}}$6412.73cm-1, the sensor achieves a normalized noise equivalent absorption (NNEA) coefficient of ${1.5} \times {{10}^{ - 11}}\;{{\rm cm}^{ - 1}}{{\rm WHz}^{ - 1/2}}$1.5×10-11cm-1WHz-1/2. This method enables the continuous locking of laser frequency and optical cavity, and it achieves the intracavity intensity modulation with an adjustable modulation depth as well.

A Modular Multilevel Dual Buck Inverter With Adjustable Discontinuous Modulation

Conventional single-cell dual buck inverters present high reliability, however they cannot be directly used in high voltage applications, because of the limited voltage rating of commercial switching devices. Therefore, a novel modular multilevel dual buck inverter (MMDBI), capable of processing voltage and power at higher levels, requiring a single dc input voltage, is proposed in this paper. Non-interleaved phase-shifted discontinuous modulation (NIPSDM) and interleaved phase-shifted discontinuous modulation (IPSDM) schemes are applied to the proposed inverter to reduce the current stress in the switching devices. Despite beneficial to the converter efficiency, the use of these discontinuous modulation approaches results in current zero-crossing distortion, which needs to be appropriately managed. In this work, theoretical values for the common-mode and differential-mode current ripples under NIPSDM and IPSDM modulation strategies are derived, and each switching operating mode is fully analyzed. From that, an adjustable current threshold value, which depends on the theoretical current ripple magnitude, is incorporated into the modulation strategy, producing a minimum switching overlap period between the two buck cells that eliminates the current zero-crossing distortion and produces a high-quality output current. Extensive simulation and experimental investigations are presented to demonstrate the feasibility and effectiveness of proposed topology and adjustable discontinuous modulation strategy.

Deriving and Updating Optimal Transmission Configurations for Lora Networks

LoRa-based networks exhibit good flexibility in terms of configurable parameters and adjustable modulation properties. Thanks to this, wireless nodes can be tuned to improve their communication behavior. In fact, optimal network-level transmission configurations (C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">opt</sub> ) can be derived in such a way that the global network performance is maximized. To derive this C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">opt</sub> , one must know the radio-propagation behavior of each node beforehand. Traditionally, this has been pursued by using general, low-precision, propagation models due to the infeasibility (in terms of time and energy) of deriving each individual node propagation behavior. In this work we propose a straightforward bounding technique that reduces up to 73% the energy and time required to obtain the radio-propagation behavior of each individual node in the network, enabling the derivation of network-level optimal transmission configurations. Also, we provide mechanisms to keep this knowledge updated, swiftly reacting to changes in the environment and leading to network performance improvements of 15% when compared to traditional alternatives like LoRaWAN ADR. Furthermore, by means of a testbed we demonstrate that this mechanism can also provide resistance to Denial-of-service attacks. Finally, we incorporate the power consumption into the proposed Copt formulation and provide a generalizable power-consumption determination methodology. This way we can limit the set of eligible transmission configurations to help extending LoRa network lifetimes more than 40%.

Bioreduction of nitrate in high-sulfate water using a hydrogen-based membrane biofilm reactor equipped with a separate carbon dioxide module

A hydrogen-based membrane biofilm reactor (MBfR) equipped with separate hydrogen (H2) and carbon dioxide (CO2) adjustable gas-transfer modules was used to explore denitrification in a water containing a high concentration of sulfate (SO42−) (500 mg S/L). CO2 was used for pH control and as the sole carbon source for long-term operation of the MBfR, during which >90% denitrification was achieved. Short-term experiments demonstrated that increasing the SO42− loading had negligible effect on nitrate (NO3−) reduction, but SO42− reduction was strongly suppressed by high NO3− loading regardless of H2 availability. Sulfide oxidation accompanied with SO42− reduction occurred in the biofilm, and the intermediate element sulfur was further oxidized to SO42−. Results of high-throughput sequencing suggest that sulfide oxidation was carried out by the known sulfide-oxidizing bacteria: Starkeya and Xanthobacter. And sulfide oxidation coupled to denitrification was a primary reason for minimal net SO42− reduction when the NO3− loading was high enough. This study demonstrates the promise of effective denitrification without net SO42− reduction in MBfR systems and documents the important sulfur-conversion processes.