Search

Your search keyword '"Lyu Chengang"' showing total 100 results
Start Over Author "Lyu Chengang"
100 results on '"Lyu Chengang"'

11. An Adaptive Illumination Optimization Method for Local Overexposed Image.

Author

Lyu, Chengang, Zhang, Mengqi, and Jin, Jie

Subjects
*LIGHTING, *IMAGE processing
Abstract

In order to solve the local overexposure caused by uneven surface reflectance, this paper proposes a fast-adaptive illumination control method with a camera-projector system. At first, an image is captured by the camera and the local overexposed area is segmented using saliency detection. Then the calculated image is projected onto the object by the projector as corrective illumination. The calculation process includes the inversion of the gray value in the overexposed area and the adjustment based on the position and depth information of the object. The high-exposure saturated regional which affects the target recognition is thus reduced, and the original illumination intensity is reserved for the other regions. This process is iterated until the optimal illumination is achieved. The resulting image for each iteration is evaluated using Blind/no Reference Image Space Quality Estimator (BRISQUE). When BRISQUE value reaches the minimum, a high-quality image is achieved. The experiments show that the proposed approach can significantly improve the speed of obtaining normally exposed images, and this system provides new ideas for industry image acquisition. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

12. Visual Early Leakage Detection for Industrial Surveillance Environments

Author

Yage Liu, Jie Jin, Wang Xuekai, Lyu Chengang, Yuxin Chen, and Jiachen Yang

Subjects
Hardware_MEMORYSTRUCTURES, Computer science, Real-time computing, Hardware_PERFORMANCEANDRELIABILITY, Interference (wave propagation), Pressure sensor, Computer Science Applications, Extractor, Intrusion, Hardware_GENERAL, Control and Systems Engineering, Hardware_INTEGRATEDCIRCUITS, Sensitivity (control systems), Electrical and Electronic Engineering, Information Systems, Leakage (electronics)
Abstract

Liquid leakage can cause industrial accidents. Current liquid leakage detection methods judge the leakage state by analyzing signals from special intrusion sensors, which is not ideal for early leakage due to sensor limited sensitivity. Visual information from surveillance systems deployed in industrial environments can reflect early leakage that cannot be monitored by such pressure sensors. In this paper, we propose a visual early leakage detection system based on a visual background extractor (Vibe) and EfficientNetB0. First, we extract the translucent and small potential leakage candidates based on Vibe, which include leakage targets and environmental interference. Then, to further recognize leakage targets in potential leakage candidates, we explore CNN models and a few recently proposed methods, and compare them with two different evaluation criteria. Our model based on EfficientNetB0 performs best and achieves 99.526% accuracy. Additionally, our CNN model for leakage recognition with a smaller size is feasible for industrial application. Experiments are conducted on the leakage dataset from surveillance video from the Tianjin Binhai Heating Station, and the detection results are consistent with real leakage situations. Our leakage detection system has high sensitivity and accuracy, which meets the requirements of early leakage detection.

Published
2022

25. High-Spatial-Resolution Strain and Vibration Dual-Parameter Sensor Based on Ultra-Short FBG.

Author

Lyu, Chengang, Tian, Jiachen, Yang, Bo, Deng, Yi, Xiao, Yanping, Jin, Jie, and Ge, Chunfeng

Abstract

A method for simultaneous measurement of static strain and dynamic vibration with high spatial resolution based on millimeter sized ultra-short fiber Bragg grating (FBG) is proposed and experimented in this paper. Dual-parameter can be measured on the single unit of cascaded FBGs, which interrogate with optical coherence tomography method by a designed wavelength swept light source. In experiments, the sensing units were precisely located with mm degrees by inverse FFT of coherent signals and the phase difference calculation. And also the strain and vibration could be simultaneously measured in time and frequency domain: demodulating strain with low frequency component’s phase shift of coherent signals by extracting envelope and interpolation; demodulating vibration by finding peak positions of sidebands in FFT spectrum. Spatial resolution of positioning can reach 0.77 mm. Sensitivity of strain measurement reaches 0.811 $\boldsymbol{\times} 10^{-3} rad/\mu \varepsilon $ under a wavelength sweep rate of 0.2 nm/ms, and forces of several milli newtons can be perceived. The measurement accuracy of kHz vibration frequency is over 98%. Moreover, by comparing simultaneous and respective measurement of dual-parameter, it can be seen that there is almost no crosstalk affectting the results. The results shows great application potential in the field of mechanical sensing, especially in bionic tactile sensing for its simultaneous dynamic and static measurement with accurate positioning performance. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

32. Automatic epilepsy detection based on generalized convolutional prototype learning

Author

Chen Yuxin, Lyu Chengang, Chen Zhijuan, Liu Yuheng, and Wang Zengguang

Subjects
medicine.diagnostic_test, Prototype learning, Computer science, business.industry, Applied Mathematics, media_common.quotation_subject, Feature vector, Pattern recognition, Electroencephalography, Condensed Matter Physics, medicine.disease, Epilepsy, Discriminative model, Robustness (computer science), medicine, Quality (business), Sensitivity (control systems), Artificial intelligence, Electrical and Electronic Engineering, business, Instrumentation, media_common
Abstract

In clinical applications, different EEG acquisition equipment and experience of neurologists make the quality and pattern of EEG signals different. This hinders the promotion of some automatic epilepsy detection methods in clinical practice. In order to solve this problem, we proposed an epilepsy detection method based on generalized convolutional prototype learning (GCPL). GCPL maps EEG samples to the regions close to the prototype in the feature space, which is more suitable for the classification of clinical EEG signals with diverse quality and pattern. Moreover, it can overcome the impact of the change of discriminative features caused by different datasets on the classification performance, which shows more robustness. In the experiments, the proposed method could reach a sensitivity of 98.75%, a specificity 100% of and an accuracy 99.38%, respectively on the clinical dataset. GCPL can complete a test within 2 s, which meets the actual clinical needs.

Published
2021

39. Non-contact low-frequency vibration rapid measurement based on hue-height mapping

Author

Yang Bai, Lyu Chengang, Hao Qi, and Prabir Koirala

Subjects
Physics, Pixel, Applied Mathematics, Acoustics, 020208 electrical & electronic engineering, 010401 analytical chemistry, 02 engineering and technology, Condensed Matter Physics, 01 natural sciences, 0104 chemical sciences, Vibration, Amplitude, 0202 electrical engineering, electronic engineering, information engineering, Demodulation, RGB color model, Millimeter, Loudspeaker, Electrical and Electronic Engineering, Instrumentation, Hue
Abstract

A non-contact vibration measurement system based on camera-projector system and hue-height mapping is proposed. The CCD camera acquires the color deformed fringes with red (R) green (G) blue (B) channels, which is modulated by a vibrating loudspeaker, with the aim to demodulate the information of vibrations. In order to avoid the crosstalk between RGB channels, the sampling frames from RGB space are converted to hue (H) saturation (S) intensity (I) space. By using hue information, height value of each pixel point at different sampling frames can be restored, so as to obtain the vibration curve of the loudspeaker. The experiments compare the vibrations of different frequencies at several Hertz, and different amplitudes lower than millimeter degree. The results show that this optical image system can detect low-frequency vibrations with high accurate amplitudes, and meet the basic industrial measurement requirements.

Published
2020

41. Vehicle video surveillance system based on image fusion and parallel computing.

Author

Liu, Shan, Lyu, Chengang, and Gong, Haotian

Subjects
*VIDEO surveillance, *IMAGE fusion, *PARALLEL programming, *COMPUTER vision, *AUTONOMOUS vehicles, *DRIVERLESS cars, *PARALLEL computers
Abstract

Summary: Autonomous driving has gradually moved towards practical applications in recent years. It is particularly critical to provide reliable real‐time environmental information for autonomous driving systems. At present, vehicle video surveillance systems based on multi‐source video and target detection algorithms can effectively solve these problems. However, the previous vehicle video surveillance systems are often unable to balance the surveillance effect and the surveillance frame rate. Therefore, we will introduce a vehicle video surveillance system based on parallel computing and computer vision in this article. First, multiple fisheye cameras are used to collect surround‐view environmental information. Second, we will use a low‐light camera, infrared thermal imager, and millimeter‐wave radar to provide forward‐view environmental information at night. Correspondingly, we designed the surround‐view image fusion algorithm and the forward‐view image fusion algorithm based on parallel computing. At last, a monocular camera and detection algorithms are used to provide forward‐view detection results. In a word, this vehicle video surveillance system will benefit the practical application of autonomous driving. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

48. Low-frequency vibration measurement based on camera-projector system

Author

Chang Yuqing, Jiachen Yang, Liu Yuxiang, Lyu Chengang, Jie Jin, Bao Zhiqiang, Gao Jiale, and Shuang Gao

Subjects
Physics, Pixel, Field (physics), Applied Mathematics, Acoustics, 010401 analytical chemistry, Process (computing), 01 natural sciences, 0104 chemical sciences, law.invention, 010309 optics, Vibration, Amplitude, Projector, law, 0103 physical sciences, Trajectory, Point (geometry), Instrumentation, Engineering (miscellaneous)
Abstract

A low-frequency vibration measurement based on a camera-projector system is proposed and verified by experiment at several hertz. A CCD camera gathered deformed fringes modulated by a vibrating trumpet at a constant sampling rate. Information on amplitude and frequency are contained in a series of observed images. The height value of each pixel point in the image can be restored by using Fourier transform profilometry. Connecting the height value of every sampling frame into a motion trajectory, the frequency and amplitude of the vibrating object can be obtained from the trajectory curve. The experiment restored the vibration process of a trumpet in half a cycle while the vibrating frequency was 3 Hz. Subsequently, the measurement of different frequencies and different amplitudes are comparatively studied. The results show that slight vibration at 35 µm and above can be detected by the system, which confirms the feasibility of the proposed system in the field of industrial vibration measurement.

Published
2017

Catalog

Books, media, physical & digital resources
See catalog results