Remimazolam is a new ultrashort acting benzodiazepine anesthetic which has predictable sedative duration and rapid recovery in gastrointestinal endoscopy. Propofol commonly used intravenous clinical work also action, short action time recovery. To date, there have been relatively few articles comparing the two for general anesthesia induction. So, we conducted randomized trial to evaluate whether remimazolam superior propofol during induction terms of efficacy safety.One hundred eighty nine...

Nearshore multi-object tracking (NMOT) aims to locat and identify nearshore objects. Most approaches accomplish this task using radar remote-sensing technologies. In contrast, video data can describe the visual appearance of objects without prior information, such as identity, location, or movement. study, we introduce a cross-modality pipeline address four major challenges NMOT. First, propose introducing bi-attention transformer (CBT) manage information interaction between RGB thermal...

Pavement damage is the main factor affecting road performance. cracking, a common type of damage, key challenge in maintenance. In order to achieve an accurate crack classification, segmentation, and geometric parameter calculation, this paper proposes method based on deep convolutional neural network fusion model for pavement identification, which combines advantages multitarget single-shot multibox detector (SSD) U-Net model. First, classification detection applied classify cracks obtain...

Sorting gangue from raw coal is an essential concern in mining engineering. Prior to separation, the location and shape of should be extracted image. Several approaches regarding automatic detection have been proposed date; however, none them satisfying. Therefore, this paper aims conduct segmentation using a U-shape fully convolutional neural network (U-Net). The trained segment images collected under complex environmental conditions. probability map outputted by was used obtain information...

It is critical for intelligent vehicles to be capable of monitoring the health and well-being drivers they transport on a continuous basis. This especially true in case autonomous vehicles. To address issue, an automatic system developed driver’s real emotion recognizer (DRER) using deep learning. The emotional values indoor are symmetrically mapped image design order investigate characteristics abstract expressions, expression principles, experimental evaluation conducted based existing...

Car crashes are among the top ten leading causes of death; they could mainly be attributed to distracted drivers. An advanced driver-assistance technique (ADAT) is a procedure that can notify driver about dangerous scenario, reduce traffic crashes, and improve road safety. The main contribution this work involved utilizing driver's attention build an efficient ADAT. To obtain "attention value", gaze tracking method proposed. direction critical toward understanding/discerning fatal...

Abstract Short‐term forecasting of traffic flow is an indispensable part easing pressure. Considering that different patterns will affect the short‐term prediction results, a combined method based on K‐means clustering algorithm and gated recurrent unit (GRU) proposed to build model overcome above problems. The used cluster historical data establish pattern libraries. K‐nearest neighbour (KNN) classification determine most similar change trend date be predicted. All in this category training...

The lack of skid resistance performance is a crucial factor leading to road traffic accidents. pavement surface friction an essential indicator for measuring the resistance. texture structure significantly affects between tire and pavement, determining To deeply study relationship performance, two types asphalt mixture specimens, concrete (AC) open-graded course (OGFC), are prepared test. Firstly, high-precision 3D smart sensor Gocator 3110 used collect point cloud data texture. British...

To address the problem that traditional models are not effective in predicting atmospheric temperature, this paper proposes an temperature prediction model based on symmetric BiLSTM (bidirectional long short-term memory)-Attention model. Firstly, meteorological data from five major stations Beijing were integrated, cleaned, and normalized to build dataset containing multiple feature dimensions; then, a memory network was used construct with forward backward information time dimension. And...

• Based on fuzzy trend forecasting, automatic clustering and PSO techniques. IRI is regarded as time series, which are divided into different granular spaces. prediction in this research can utilize multiple factors. technique employed incorporating with multi-granularity analysis. The effective of pavement performance trends help achieving the cost-effective management pavements over their service life. international roughness index (IRI) a widely used index, be considered time-dependent...

To solve the problem of low accuracy pavement crack detection caused by natural environment interference, this paper designed a lightweight framework named PCDETR (Pavement Crack DEtection TRansformer) network, based on fusion convolution features with sequence and proposed an efficient method. Firstly, scalable Swin-Transformer network residual are used as two parallel channels backbone to extract long-sequence global underlying visual local cracks, respectively, which concatenated fused...

A super-resolution reconstruction approach based on an improved generative adversarial network is presented to overcome the huge disparities in image quality due variable equipment and illumination conditions image-collecting stage of intelligent pavement detection. The nonlinear generator first improved, Residual Dense Block (RDB) created serve as Batch Normalization (BN). Attention Module then formed by combining RDB, Gated Recurrent Unit (GRU), Conv Layer. Finally, a loss function L1 norm...