Visual object tracking often faces challenges such as invalid targets and decreased performance in low-light conditions when relying solely on RGB image sequences. While incorporating additional modalities like depth infrared data has proven effective, existing multimodal imaging platforms are complex lack real-world applicability. In contrast, near-infrared (NIR) imaging, commonly used surveillance cameras, can switch between NIR based light intensity. However, objects across these...

Optical clocks are the most precise measurement devices. Here we experimentally characterize one such clock based on 1S0-3P0 transition of neutral 171Yb atoms confined in an optical lattice. Given that systematic evaluation using interleaved stabilization scheme is unable to avoid noise from laser, synchronous comparisons against a second lattice system were implemented accelerate evaluation. The fractional instability falls below 4 × 10-17 after averaging over time 5,000 seconds. frequency...

Many methods have been proposed to reconstruct the moving object based on phase shifting profilometry. Quality reconstruction results can be achieved when a single or multiple objects with same movement are measured. However, errors will introduced individual movements reconstructed. This paper proposes an automated method track and movement. First, identified automatically their bounding boxes obtained. Second, objects' images before movement, tracked by KCF algorithm in successive fringe...

In the medical field, text classification based on natural language process (NLP) has shown good results and great practical application prospects such as clinical value, but most existing research focuses English electronic record data, there is less processing task for Chinese records. Most of current records are non-institutionalized texts, which generally have low utilization rates inconsistent terminology, often mingling patients’ symptoms, medications, diagnoses, other essential...

Electronic shopping has become an important way of in our daily life. A good recommendation system can greatly improve the consumer experience through alleviating and product information overload problem. In real scenario, each a wide variety interests, consumers interests are constantly changing. Many recent models usually give overall embedding for behavior sequence or only capture features customers characteristics products from single perspective. However, these methods hardly model...

In the medical field, Named Entity Recognition (NER) plays a crucial role in process of information extraction through electronic records and texts. To address problems long distance entity, entity confusion, difficulty boundary division Chinese record NER task, we propose method based on multi-head attention mechanism character-word fusion. This uses new joint feature representation pre-training model BERT self-constructed domain dictionary, which can accurately divide solve impact...

Masks cover most areas of the face, resulting in a serious loss facial identity information; thus, how to alleviate or eliminate negative impact occlusion is significant problem field unconstrained face recognition. Inspired by successful application attention mechanisms and capsule networks computer vision, we propose ECA-Inception-Resnet-Caps, which novel framework based on Inception-Resnet-v1 for learning discriminative features mask-wearing conditions. Firstly, Squeeze-and-Excitation...

Intersection over union (IoU) has been widely adopted to evaluate and select candidate regions in multi-oriented object detection. Intuitively, overlaps between candidates ground-truth boxes make more sense when assessing the quality of horizontal candidates. However, minimum bounding box (HMBB) is generally used for IoU calculation practice, bringing about biased results. In this article, we propose a novel Splicing Union (SIoU) provide preferable metric selection detecting objects. By...

Extracting entity relations from unstructured medical texts is a fundamental task in the field of information extraction. In relation extraction, dependency trees contain rich structural that helps capture long-range between entities. However, many models cannot effectively use or learn sentence adequately. this paper, we propose extraction model based on syntactic structure information. First, learns sequence by Bi-LSTM. Then, through graph convolutional networks. Meanwhile, order to remove...

Face recognition in general scenarios has been saturated recent years, but there is still room to enhance model performance extreme and fairness situations. Inspired by the successful application of Transformer ConvNet computer vision, we propose a FIN-Block, which gives more flexible composition paradigm for building novel pure convolution provides foundation constructing new framework face both FIN-Block-A uses combination stacked large-size kernels parallel branches ensure large spatial...

Unstructured Chinese medical texts are rich sources of entity and relational information. The extraction relationships from is pivotal for the construction knowledge graphs aiding healthcare professionals in making swift informed decisions. However, these presents a formidable challenge, notably due to issue overlapping relationships. This study introduces novel model that leverages RoFormer’s rotational position encoding (RoPE) technique an efficient implementation relative encoding....

The integration of miniaturized spectrometers into mobile devices offers new avenues for image quality enhancement and facilitates novel downstream tasks. However, the broader application spectral sensors in photography is hindered by inherent complexity images constraints imaging capabilities. To overcome these challenges, we propose a joint RGB-Spectral decomposition model guided framework, which consists two steps: prior-guided enhancement. Firstly, leverage complementarity between RGB...

In this paper, a novel virtual antenna array and fractional Fourier transform (FRFT)-based 2-dimension super-resolution time-of-arrival (TOA) estimation algorithm for OFDM WLAN systems has been proposed. The proposed employs channel frequency responses (CFRs) at the equi-spaced positions on line or quasi-line moving trajectory, i.e., CFRs of array, to extract multipaths' TOA information. Meanwhile, new chirp-like quadratic function is used approximate phase variation across space dimension,...

In the process of rice production, pests are one main factors that cause yield reduction. To implement prevention and control measures, it is necessary to accurately identify types diseases. However, application image recognition technologies focused on agricultural field, especially in field disease pest identification, relatively limited. Existing research diseases has problems such as single data types, low volume, accuracy. Therefore, we constructed dataset (RPDD), which was expanded...

The medical information carried in electronic records has high clinical research value, and named entity recognition is the key to extracting valuable from large-scale texts. At present, most of studies on Chinese are based character vector model or word model. Owing complexity specificity text, existing methods may fail achieve good performance. In this study, we propose a method that fuses vectors. expresses texts as vectors separately them for features. proposed can effectively avoid...

Although the neural mechanism underlying risk decision has been extensively investigated, origination of attitude and probability distortion need to be further elucidated. In this study, Rescorla–Wagner model with learning rates a+/a- upon gain/loss evaluates risky outcome forms subjective values options through process, softmax function produces choice between options. Our demonstrates that is determined by undervaluation/overvaluation outcome, standard deviation value, discrimination...

Line-focus Fresnel concentrating PV module can improve the power efficiency because of concentration ratio growing.But there is another problem which non-uniform illumination influence electrical performance parameters, such as short-circuit current Isc, open-circuit voltage Voc, fill factor FF and η.How to assess electric becoming an awkward problem.The whole equivalent circuit, including darker brighter parts, was analyzed.And forecasting algorithm designed by MATLAB software according...

In order to improve the online application ability and auxiliary decision-making of power grid dispatching operation rules, a method for identifying key information rules based on Bidirectional Long Short-Term Memory network-Conditional Random Field (BiLSTM-CRF) is proposed. A feature extraction labeling rule text proposed, BiLSTM used learn nonlinear characteristics implicit timing between regular features, CRF optimize bidirectional network encoding labels globally. The recognized combined...

<title>Abstract</title> Surrogate-assisted evolutionary algorithms (SAEAs) are one effective method for solving expensive optimization problems. However, there has been little attention to many-objective irregular To address this issue, we propose an ensemble surrogate-assisted adaptive reference point guided algorithm dealing with Firstly, a adaptation is adopted in the proposed adjust calculating indicators and guide search process. Secondly, enhanced inverted generational distance...

Visual tracking often faces challenges such as invalid targets and decreased performance in low-light conditions when relying solely on RGB image sequences. While incorporating additional modalities like depth infrared data has proven effective, existing multi-modal imaging platforms are complex lack real-world applicability. In contrast, near-infrared (NIR) imaging, commonly used surveillance cameras, can switch between NIR based light intensity. However, objects across these heterogeneous...