Teaching gesture recognition is a technique used to recognize the hand movements of teachers in classroom teaching scenarios. This technology widely education, including for evaluation, enhancing online teaching, and assisting special education. However, current research on mainly focuses detecting static gestures individual students analyzing their behavior. To analyze teacher's mitigate difficulty single-target dynamic multi-person scenarios, this paper proposes skeleton-based (ST-TGR),...

As one of the most important research fields in brain-computer interface (BCI) field, electroencephalogram (EEG) classification has a wide range application values. However, for EEG signal, it is difficult traditional neural networks to capture characteristics signal more comprehensively from time and space dimensions, which certain effect on accuracy classification. To solve this problem, we can improve via end-to-end learning dimensions EEG. In paper, new type network, separable EEGNet...

Abstract A capsule network (CapsNet) is a recently proposed neural model with new structure. The purpose of CapsNet to form activation capsules. In this paper, our team proposes dual attention mechanism (DA-CapsNet). DA-CapsNet, the first layer added after convolution and referred as Conv-Attention; second PrimaryCaps Caps-Attention. experimental results show that DA-CapsNet performs better than CapsNet. For MNIST, trained tested in testset, accuracy 100% 8 epochs, compared 25 epochs for...

Background Detection of pulmonary nodules is an important aspect automatic detection system. Incomputer-aided diagnosis (CAD) systems, the ability to detect highly important, which plays role in and early treatment lung cancer. Currently, depends mainly on doctor experience, varies. This paper aims address challenge nodule more effectively. Methods A method for detecting based improved neural network presented this paper. Nodules are clusters tissue with a diameter 3 mm 30 parenchyma....

Abstract In this paper, a variable structure pneumatic soft robot is proposed. Its in that when it grasps irregular objects, can adapt to different sizes by active expansion or contraction. range from diameter 200 300 mm, its four actuators (SPAs) be rotated independently shapes, and has high flexibility. The compliant grasping method enables capture at the best position, which improve success rate of reduce damage object being grasped. experiment proves effectiveness mechanism, proposed low...

Conventional robots have shown good positioning accuracy and reliability in multiple environments. However, the robots' lack of flexibility makes them unsuitable for application certain areas. Thus, a soft robotic system should be considered due to its adaptivity flexibility. This paper proposes novel shape memory alloy (SMA)-actuated degrees freedom (DOF) robot, which theoretically supports maximum nine-DOF movement accuracy. A motion test evaluation are conducted proposed multi-DOF robot....

Soft robots have the characteristics of good adaptability to an environment. But at present, most soft only complete a single specific task, and as result, their ability adapt complex environments is limited. In order solve weak diverse environments, this paper presents new bionic omnidirectional bending actuator (BOBA) inspired by leeches, caterpillars, other mollusks. BOBA has less radial expansion, fast response, flexibility, bending. Using modular methods, it can be quickly assembled...

In recent years, deep-learning has been gradually applied to the detection and classification of true pseudo defects in printed circuit boards (PCBs). To judge authenticity PCB defects, it is necessary not only combine shape characteristics themselves, but also add relationship information between surrounding environment defects. However, difficult for general methods extract such features. this article, we propose neighborhood correlation enhancement network (NCE-Net), which effectively...

In the field of music generation, generating structured is a highly challenging research topic. Music generation methods are currently learned in Euclidean space and usually modeled as time series without structural properties, but due to limitations representation space, hierarchical structure difficult learn, generated poorly structured. Therefore, based on hyperbolic theory, this paper proposes Hyperbolic Transformer model, which considers hierarchy models components space. Meanwhile,...

Chest computed tomography is an important method for detecting lung cancer. Because early cancer nodules are small, making them difficult to detect, the current automatic detection system can easily lead missed diagnosis when these nodules; therefore, accurate of crucial improving cure rate. To reduce rate in system, it necessary optimize extraction steps candidate nodules. Based on improvement U-Net, this paper proposes noisy U-Net (NU-Net), which enhance neural network's sensitivity small...

Background: As a control strategy of industrial robots, sliding mode has the advantages fast response and simple physical implementation, but it still problems chattering low tracking accuracy caused by chattering. This paper proposes new for application robot control, which effectively solves these problems. Methods: In this paper, deep deterministic policy gradient–nonlinear nonsingular terminal (DDPG–NNFTSMC) is proposed control. order to improve anti-interference ability, DDPG used...

As an important part of the quadruped robot, leg determines its performance. Flexible legs or flexible joints aid in buffering and adaptability robots. At present, most robots only have two-dimensional flexibility use complex parallel structures to achieve three-dimensional flexibility. This research will propose a new type structure. passive compliant reduces weight structure robot. The anti-impact performance robot is verified by side impact experiment. simulation experiments show that...

In order to improve the speed tracking accuracy of Brushless DC motor, this paper proposes a PID control method based on deep deterministic policy gradient algorithm(DDPG) compensation. The running state BLDCM is monitored by DDPG, and proportional, integral differential links are compensated make controller have learning ability. At same time, Gauss function maximum critical interval combined define reward enhance versatility strategy. Simulation results show that proposed strategy has...

An unknown nonlinear disturbance seriously affects the trajectory tracking of autonomous underwater vehicles (AUVs). Thus, it is critical to eliminate influence such disturbances on AUVs. To address this problem, paper proposes a double-loop proportional–integral–differential (PID) neural network sliding mode control (DLNNSMC). First, PID surface proposed, which has faster convergence speed than other surfaces. Second, high-order observer and are combined observe compensate for AUV system....

Residual learning and skip connections can train very deep convolutional networks with a significant performance boost. On this basis, DenseNet proposes dense connection structure in which each layer the network be directly connected to all previous layers. Compared ResNet, reduce parameter size of greater extent enhance utilization features. However, still has problem redundancy. In paper, accuracy image classification is not affected by reducing unnecessary shallow amount model. article,...

An adaptive proportional integral robust (PIR) control method based on deep deterministic policy gradient (DDPGPIR) is proposed for n-link robotic manipulator systems with model uncertainty and time-varying external disturbances. In this paper, the of nonlinear dynamic model, disturbance, friction resistance are integrated into system, term used to compensate system. addition, information as input DDPG agent search optimal parameters controller in continuous action space. To ensure agent’s...

Herein, we report a modular assembly of CF 3 -modified 3,4-dihydroquinazolinones from basic building blocks using one-pot light-driven multicomponent approach.

Summary Parallel robots are widely used in the fields of manufacturing, medical science, education, scientific research, etc. Many studies have been conducted on topic already. However, shortcomings still exist, especially certain situations. To meet demand good speed and load performances at same time, this work presents a novel 2-degree-of-freedom parallel robot. The structural design, static, stiffness, reachable workspace analysis robot given manuscript. Experiment regarding accuracy...

The incidence of cholecystitis is relatively high in developed countries and may usually be attributed to gallstones, the treatment for which involves complete surgical removal gallbladder (cholecystectomy). Bile acids produced following cholecystectomy continue flow into duodenum but are poorly absorbed by colon. Excessive bile colon stimulate mucosal secretion water electrolytes leading, severe cases, diarrhoea. acid diarrhoea (BAD) difficult diagnose, requiring a comprehensive medical...

In this paper, a soft robot driven by gas–liquid phase transition actuator with new structure is designed; The the pressure generated electrically induced ethanol transition. drive was found to be able generate larger driving force using only low voltage. Compared gas of traditional robot, transition-driven does not require complex circuit system and huge external energy supply air pump, making its overall more compact. At same time, because on has good tightness less recovery time. A...

A set of seven-stage reluctance miniaturized Hopkinson bar electromagnetic launcher has been developed in this paper. With the characteristics high precision, small size, and little noise pollution, device complies with requirements for strain rate. The is a accelerating up to 65.5 m/s. performance microcontroller used control accurately discharge capacitor sets, by means which outlet velocity projectile can be controlled within certain range.

In the field of fashion design, designing garment image according to texture is actually changing shape image, and image-to-image translation based on Generative Adversarial Network (GAN) can do this well. This help designers save a lot time energy. GAN-based has made great progress in recent years. One models––StarGAN, realized function multi-domain by using only single generator discriminator. paper details use StarGAN complete task design. Users need input an label for type generate...