Abstract Human faces contain useful information that can be used in the identification of age, gender, weight etc. Among these biometrics, body mass index (BMI) and are good indicators a healthy person. Motivated by recent health science studies, this work investigates ways to identify malnutrition affected people obese analyzing BMI from facial images proposing regression method based on 50‐layers Residual network architecture. For face detection, Multi‐task Cascaded Convolutional Neural...

Railway signaling facilitates two main functions, namely, train detection and control, in order to maintain safe separations among the trains. Track circuits are most commonly used means with simple open/close circuit principles; subsequent adoption of axle counters further allows trains under adverse track conditions. However, electrification power electronics traction drive systems, aggravated by electromagnetic interference vicinity system, railway engineers often find unstable or even...

Wheel defects on trains, such as flat wheels and out-of-roundness, inevitably jeopardize the safety of railway operations. Regular visual inspection checking by experienced workers are commonly adopted practice to identify wheel defects. However, may not be spotted in time. Therefore, an automatic, remote-sensing, reliable, accurate monitoring system for condition is always desirable. The paper describes a real-time monitor based fiber Bragg grating sensors. Track strain response upon...

In the processing of recognizing Φ-OTDR event signals by deep learning models, a large amount annotated data is required for model training. However, annotation major challenge lot unlabeled and labeling cost are high. This article proposes valuable signal selection method sensing system. The samples that beneficial training from set selected based on their uncertainty values. Through only these classification with them, can be saved keep its accuracy. experiments show this save 83.77%...

Strain and acceleration signals are essential for accurate event recognition along submarine cables. However, concise identification still poses challenges the prompt classification of anchor smashing hooking events with multiple types sensors locations. For these reasons, this paper proposes a hybrid model that combines convolutional neural networks-bidirectional long short-term memory (CNN-BiLSTM) block attention module (CBAM) to instantaneously identify organise events. Eight different...

Urban villages are a characteristic settlement type characterized by preserving their morphological characteristics embedded in sharp contrast modern, high-rise developments found especially fast growing urban agglomerations of China. They serve very important socioeconomic functions terms the provision cheap housing for rural-urban migrants, but they also considered controversial local governments. Due to unprecedented pace growth, Pearl River Delta region (PRD), up-to-date information on...

This paper deals with uncertainty, asymmetric information, and risk modelling in a complex power system. The uncertainty is managed by using probability decision theory methods. Multi-attribute decision-making (MADM) technique very effective well-known tool to investigate fuzzy information more effectively. However, the selection of houses cannot be carried out utilizing symmetry because enterprises does not have complete so should used when selecting enterprises. Hamy mean (HM) operator...

Different signal representations show different unique features for classification. In this paper, a feature fusion method with attention mechanism based on multiple is proposed Φ-OTDR event classification buried optical fiber. Each representation fused after extraction to get richer and better features. With the help of layer pruning mechanism, network size can be kept avoid computation increase. Experiment results that 3 improve recognition accuracy 97.93%, 3.52% improvement compared...

An efficient and robust meter readings location recognition algorithm plays a very important role in remote reading systems automatic inspection systems. A novel electric method is presented this study. It includes the following three steps: coarse-to-fine detection of readings, projection-based character segmentation BP neural network recognition. The was tested on HUALI, KEYUAN, JINLING, LANJIER SANXING images. Experimental results show that proposed has satisfying performance good robustness.

Due to the numerous important applications of video surveillance and monitoring, object tracking has been an active research topic in last decade. This paper makes a survey approaches high quality by looking at theoretical backgrounds practical results, which are categorized into four groups. The principle, evolution processes latest progresses these identified form conclusion for future directions algorithms.

Laying power cables along the bridge is a new way of laying submarine across sea. Monitoring health status and their telescopic compensation devices necessary. In this study, fiber grating sensing technology was used to monitor strain, temperature, vibration cable Zhoushan–Daishan Bridge in Zhoushan, Zhejiang Province, its device. Two typhoons one invasion event happened during monitoring period. Temperature signals, strain time domain time–frequency signals were analyzed. The results showed...

Ankle joint flexion and extension movements play an important role in the rehabilitation training of patients who have been injured or bedridden for a long time before after surgery. Accurately guiding to perform ankle can significantly reduce deep vein thromboembolism. Currently, most devices focus on assisting with movements, there is lack effectively monitoring these movements. In this study, we designed movement-monitoring device based pressure sensor. It was composed STM32...

Automated steel bar counting and center localization plays an important role in the factory automation of bars. Traditional methods only focus on their performances are often limited by complex industrial environments. Convolutional neural network (CNN), which has great capability to deal with tasks challenging environments, is applied this work. A framework called CNN-DC proposed achieve automated simultaneously. The first detects candidate points a deep CNN. Then effective clustering...

Automatic and accurate estimation of choroidal thickness plays a very important role in computer-aided system for eye diseases. One the most common methods automatic is segmentation-based methods, which boundaries choroid are first detected from optical coherence tomography (OCT) images. The then computed based on boundaries. A shortcoming that estimating precision greatly depends segmentation results. To avoid dependence step, this paper, we propose direct method convolutional neural...

To measure the temperature profile of various critical parts on a train bogie to fore-warn drivers potential failures in those mechanical parts, such as axle and motor bearing, gear box braking mechanisms, which could go undetected under existing maintenance procedure, so reduce possibility sudden breakdowns. Eighteen fiber Bragg grating sensors are installed an interrogation system is used capture sensor data, then data processing presentation software consisting graphical user interface...

In recent years, CNN models have become a mainstream deep learning framework and been widely used in research on improving the accuracy of $\Phi$-OTDR event recognition systems. However, new framework, Transformer model, has recently proposed to replace CNN. this article, an method based transformer model is proposed. It compared with experiments shown that performs better than pretrained models, rate 99.64%. When pruned only 2 layers, its can still be saved, size reduced 1.49M.

In this paper, some novel converters are proposed to solve the drawbacks of traditional phase-shifted full-bridge (PSFB) converter. primary side converters, FB inverter is divided into two half-bridge inverters and large-sized transformer replaced by small-sized transformers. By employing structure, zero voltage switching range can be extended circulating current existing in additional PSFB converter removed. secondary side, various rectifiers adopted suit for different applications....