Electrocardiograms (ECGs), which are recordings of the electrical activity of the heart with electrodes placed on the body, have many clinical applications. Each phase of the heart's mechanical function is represe...
详细信息
This paper describes a new approach to the problem of interception of wireless communication channels between the legitimate users. Physical PHY Layer Security (PLS) is new topic enhancing the secrecy performance of a...
详细信息
At low sampling to fundamental frequency ratios, the time delay introduced by the Pulse Width Modulation (PWM) calculation and the duty cycle update deteriorates the decoupling dynamic between d and q-axis currents. T...
详细信息
In this work, an attempt is made for the first time to use the measurement pattern generated by morphological transformation quantified by Hausdorff fractal dimension (HFD) and classified with ensemble learning based ...
详细信息
The k-nearest neighbor learning algorithm is indeed one of the most commonly used supervised algorithms in machine learning. KNN is a lazy learner, and one of the most important similarity features used in this algori...
详细信息
The quality of the airwe breathe during the courses of our daily lives has a significant impact on our health and well-being as ***,personal air quality measurement remains *** this study,we investigate the use of fir...
详细信息
The quality of the airwe breathe during the courses of our daily lives has a significant impact on our health and well-being as ***,personal air quality measurement remains *** this study,we investigate the use of first-person photos for the prediction of air *** main idea is to harness the power of a generalized stacking approach and the importance of haze features extracted from first-person images to create an efficient new stacking model called AirStackNet for air pollution *** consists of two layers and four regression models,where the first layer generates meta-data fromLight Gradient Boosting Machine(Light-GBM),Extreme Gradient Boosting Regression(XGBoost)and CatBoost Regression(CatBoost),whereas the second layer computes the final prediction from the meta-data of the first layer using Extra Tree Regression(ET).The performance of the proposed AirStackNet model is validated using public Personal Air Quality Dataset(PAQD).Our experiments are evaluated using Mean Absolute Error(MAE),Root Mean Square Error(RMSE),Coefficient of Determination(R2),Mean Squared Error(MSE),Root Mean Squared Logarithmic Error(RMSLE),and Mean Absolute Percentage Error(MAPE).Experimental Results indicate that the proposed AirStackNet model not only can effectively improve air pollution prediction performance by overcoming the Bias-Variance tradeoff,but also outperforms baseline and state of the art models.
作者：
Bendiak, IstvánSándor, TamásObuda University
Institute of Automation and Energy Systems Kandó Kádmán Faculty of Electrical Engineering Department of Automation Budapest Hungary
The target area of the research is the measurement of the gear change transient of a synchronous motor drive mounted on a common output shaft. The two synchronous motors are a type of motor developed for permanent mag...
详细信息
Optical camera communication (OCC) has been widely employed in various applications as a flexible and cost-effective means of communication both on land and underwater. However, the performance of the OCC system throu...
详细信息
Optical camera communication (OCC) has been widely employed in various applications as a flexible and cost-effective means of communication both on land and underwater. However, the performance of the OCC system through the water-air interface has not been thoroughly investigated. In this paper, we explore the performance of the OCC system in a water-air environment and propose a bubble-wave-mitigation algorithm to pre-process the captured frames of received video. Moreover, we propose a transformer-based neural network to demodulate the transmitted signal, mitigating the deterioration in transmission performance caused by inter-symbol interference (ISI). The experimental results demonstrate that a robust transmission can be achieved in the water-air environment by applying our proposed algorithms and neural network demodulator. Author
The aim of the paper is to examine the steady-state operation of asynchronous motors using the electrical signal analysis method. The operating conditions of rotary machines have been significantly transformed by the ...
详细信息
The visual quality of images and videos is significantly degraded by atmospheric particles such as smoke and dust, leading to the haze problem, which is characterized by low contrast and a whitish veil obscuring the c...
详细信息
The visual quality of images and videos is significantly degraded by atmospheric particles such as smoke and dust, leading to the haze problem, which is characterized by low contrast and a whitish veil obscuring the content. Dehazing techniques have been developed to mitigate these effects and restore sharp, visually appealing imagery. This paper presents an enhanced dehazing approach, referred to as the Enhanced DehazeFormer technique, which integrates both preprocessing and dehazing stages to produce high-quality dehazed images. The proposed method incorporates a three-step preprocessing phase aimed at reducing noise and enhancing dynamic range, issues commonly introduced by measurement device errors and other external factors. Prior to dehazing, each image or video frame undergoes homomorphic filtering, followed by Contrast Limited Adaptive Histogram Equalization (CLAHE) and a fast dehazing algorithm to further improve visual quality. The dehazing stage utilizes an extended and customized Swin Transformer architecture, known as DehazeFormer, which is tailored specifically for haze removal tasks. The preprocessed frames are input into the modified Swin Transformer to generate dehazed outputs of superior visual quality. The proposed technique is thoroughly evaluated on visible images, Near-Infrared (NIR) frames, and real-world hazy datasets to assess its effectiveness. Evaluation metrics include entropy, Peak Signal-to-Noise Ratio (PSNR), Feature Similarity Index (FSIM), Feature Similarity Index Chromatic (FSIMC), edge intensity, average gradient, and correlation, all of which are used to quantitatively measure dehazing performance. Furthermore, histograms and spectral entropy analyses are employed to compare the proposed method against other dehazing techniques. A comparative analysis is conducted using five frames from each type of visible and NIR videos to assess the performance of the baseline DehazeFormer and the enhanced DehazeFormer technique. Additional eva
暂无评论