This paper proposes a novel text input method based on hand gestures for wearable computers. A character is firstly written by fingertip and then recognized through B-splines recognition method. The writing procedure controlled including tracking, gesture positioning which are performed an extended CONDENSATION algorithm. We have integrated the proposed into vision system developed in our lab tested resulting Graffiti 2 alphabet. experimental result shows that promising natural

Motion tracking is attractive in what concerns a smart city environment, where citizens have to interact with Internet of Things (IoT) infrastructures spread all around one particular city. important for services and location-based cities, since it provides natural ways users the IoT infrastructures, such as ability recognize wide range hand motion real-time. Compared dedicated hardware devices, ubiquitous devices reliable speakers microphones can be developed achieve cheap acoustic-based...

Abstract Background Fetal brain magnetic resonance imaging (MRI)‐based gestational age prediction has been widely used to characterize normal fetal development and diagnose congenital malformations. Purpose The uncertainty of position external interference leads variable localization direction the brain. In addition, pregnant women typically concentrate on receiving MRI scans during anomaly scanning week, leading an imbalanced distribution data. above‐mentioned problems pose great challenges...

New energy storage has multiple values include peak shaving, reserve, frequency regulation and so on in new power systems. How to reflect the multi-scenario cost evaluation objectively is also a hot issues. This paper proposes calculation analysis model about levelized of storage, which can solve problem hybrid model. First, this analyses whole life cycle structure participating shaving scene scene, comprehensively considers influence different scenarios life, establishes systems single as...

With the increasing application demand of UAV aerial images in land surveying and mapping, agricultural production, traffic management, target detection is great significance. It difficult to detect small targets due problems low resolution, complex background, dense scale images. In this paper, YOLOv5s used as benchmark model realize vehicle unmanned vehicles. Firstly, we selected VisDrone2019-MOT dataset targets, Retinex image enhancement algorithm for preprocessing. By contrast improving...

This paper presents a novel method for image classification. For classification, use multiple instance learning (MIL) framework to describe the features as instances. By using discriminative approach, we embeded each into instance-based space and coordinates of in that are used classification k-nearest neighbor algorithm. Experiments shows approach is useful

<title>Abstract</title> Small target detection in aerial road images is challenging, and the existing models are not suitable for small targets with dense, complicated background occlusion images. Therefore, a new model based on YOLOv7 used UAV scenes. We have made four improvements： (i) head added to original detect tiny-scale targets. (ii) we use SimAM attention focus areas small-scale (iii) In downsampling stage, SPD-Conv replace CNN improve capacity identify of low resolution (iv) CARAFE...

Abstract Small target detection in aerial road images is challenging, and the existing models are not suitable for small targets with dense, complicated background occlusion images. Therefore, a new model based on YOLOv7 used UAV scenes. We have made four improvements: (i) head added to original detect tiny-scale targets. (ii) we use SimAM attention focus areas small-scale (iii) In downsampling stage, SPD-Conv replace CNN improve capacity identify of low resolution (iv) CARAFE introduced...

Human-object interaction (HOI) and human-scene (HSI) are crucial for human-centric scene understanding applications in Embodied Artificial Intelligence (EAI), robotics, augmented reality (AR). A common limitation faced these research areas is the data scarcity problem: insufficient labeled object pairs on input images, limited complexity granularity between them. Recent HOI HSI methods have addressed this issue by generating dynamic interactions with rigid objects. But more complex such as a...

A novel big data-driven approach has been proposed to address the challenges of extensive data and user application difficulties in smart grid. grid architecture based on a platform designed. The random matrix theory is applied establish algorithm model, which samples from database undergoes training learning construct model for requirements. Through calculations, macroscopic hidden transformed into microscopic users' reference, enabling identification parameters that affect normal operation...

Target detection in remote sensing images is widely used civil and military applications plays an increasingly important role. In this paper, a novel multi-scale fusion Faster RCNN model proposed to realize target optical images. First, the preprocessing operation performed on dataset by using VOC format for pretreatment data enhancement. Second, based traditional architecture, EfficientNetb0 FPN combined extract image features, so as obtain pyramid feature improve ability of network....

To explore intraretinal blood flow perfusion and nerve changes, as well the correlation between them, in diabetic patients without retinopathy (NDR).Eighty-six NDR (86 eyes) who attended ophthalmology clinic December 2019 2021 were included. Sixty-four eyes of 64 healthy examined controls same period selected control group. The underwent routine ophthalmological examination, optical coherence tomography (OCT) OCT angiography.The average thickness, minimum thickness each quadrant except for...

With the increasing consumption of electric energy in China, management power becomes more and important. It is essential for analysis appliances to make a reasonable way electricity consumption. Aiming at problem low accuracy frequency sampling recognition, this paper proposes non negative matrix decomposition method feature extraction electrical data. The peak density data calculated by algorithm, taken as initial cluster center fuzzy c-means algorithm. States each device are obtained....

With the rapid development of technology and application for Internet Things (IoT), Low-Power Wireless Personal Area Network (LoWPAN) devices are more popularly applied. Evaluation power efficiency is important to LoWPAN applications. Conventional method evaluate different as follows: first measure current under working/idle/sleep state then make an average estimation lifetime batteries, which deeply relied on accuracy testing equipment not that accurate with high cost. In this work, a...

The dynamic texture (DT) which treats the transient video process a sample from spatiotemporal model, has shown surprising performance for moving objects detection in scenes with background motions (e.g., swaying branches, falling snow, waving water). However, DT parameters estimation is based on batch-PCA, computationally inefficient method high-dimensional vectors. Besides, realm of DT, dimension state space given or set experimentally. In this work, authors present new framework to...

The recognition and classification of the remote sensing image is a key technology in application image, which has been one research hotspots. Extreme learning machine (EML), kind method, advantages fast speed good generalization capability, getting more attention from researchers. Based on two hotspots, this paper, method for fold structure information based ELM proposed. Through comparison results obtained by artificial interpretation, it shows that accuracy strong applicability, also,...

The process of civil aircraft development is a complex system, where flight test an extremely important the project. In top-level planning test, it particularly to determine quantity aircraft, which closely related total period, cost, and resources. This paper analyzes studies number performing missions, factors proposes reasonable calculation model with engineering operability based on period cost. It provides numerical data reference for overall during multi-objective optimization.

Abstract Blended-wing-body aircraft is a new type of form that has been used in military aircraft. The potential for application civil being explored. Due to its special shape, the blended-wing-body difficult adopt conventional under-wing engine mounting form, and usually applies back braced mounting. mounted on could cause surge separation problems, which affects flow field quality whole In this paper, effect height axial position overall determined by modelling calculation airflow...

This paper introduce the flight test method of civil airplane taking-off and landing in strong crosswind condition. Observe analyse relationships between pilot's inputs behaviour. Thus, establish control objectives precautionary measurements test. In test, while is accelerating during take-off, pilots' are to align center-line runway balance load main wheels with increasing velocity. While approaching landing, excessive sideslip angle or roll should be avoided prevent damage upon touchdown,...

&lt;p&gt;Reconfigurable intelligent surfaces (RISs) and millimeter-wave (mmWave) communications have been considered for providing wireless connectivity to mobile robots used in industrial plants other indoor environments. However, the existing works not sufficiently studied how number deployment locations of RISs should be optimized serving a robot. In this paper, we study RIS-assisted mmWave robot moving around fixed obstacles an environment. For total reflecting elements, formulate...

Arrhythmia is a significant cardiovascular disease that poses threat to human health, and its primary diagnosis relies on electrocardiogram (ECG). Implementing computer technology achieve automatic classification of arrhythmia can effectively avoid error, improve diagnostic efficiency, reduce costs. However, most algorithms focus one-dimensional temporal signals, which lack robustness. Therefore, this study proposed an image method based Gramian angular summation field (GASF) improved...