Search

Background: Several scalp EEG epilepsy detection methods based on machine learning have achieved good detection accuracy. However, in clinical applications, different EEG acquisition equipment and experience of neurologists make the quality and style of EEG signals different, which makes previous epilepsy detection models cannot be widely used. The establishment of epilepsy detection model for a certain hospital usually depends on a large number of EEG samples, but there are usually few EEG samples from a certain hospital. Methods: To solve this problem, we proposed a small sample epilepsy detection method based on convolutional prototype learning (CPL) in this paper. CPL consists of convolutional neural network (CNN) and prototype learning. CNN is used as an adaptive feature extraction algorithm, and prototype learning is used as a small sample classification algorithm. Results: In the experiment, we select 20, 40, 60, 80, 100 and 120 samples to train and save 6 CPL-based detection models. The 6 models are used to classify the test samples, and the accuracy are 75.97%, 83.24%, 85.67%, 88.27%, 91.09% and 94.43% respectively. Conclusions: The CPL can realize automatic feature extraction of EEG signals, and solve the problem of insufficient training samples in epilepsy detection.

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.

Visual information can be easily obtained nowadays with more access to various photographic facilities. Appropriate logical analysis and processing of massive visual information could apply to flexible environment pollution detection. In this paper, we present a visual-based non-contact sensing method to detect the low frequency vibration pollution in daily life. The edge points of vibrating objects with more vibration information are extracted using bilateral filtering and Sobel operator. The Sobel operator weights the influence of the pixel position, which make the edge feature extraction effect better and the edge feature obtained more obvious. After that, the continuous frames of the vibration objects with edge information are transformed into several one-dimensional pixel level signals as the input of multi-scale network, which also improves the generalization ability. Experiments on daily objects under low frequency vibration are conducted. The detection accuracy is above 99%, which proves the reliability of the proposed method for the detection of low frequency vibration circumstance pollution.

In low-frequency vibration measurement field, optical dynamic 3-D measurement based on camera-projector system is a novel approach. In this system, linear sinusoidal grating is the most common projection mode, which can meet the requirements of some vibration measurement. However, the dynamic measurement range of amplitude using the combination of linear grating and Fourier transform profilometry (FTP) is limited in some vibration measurement environments, because the linear fringe is periodic. To solve the problem, this article proposes a low-frequency vibration measurement method based on the concentric-circle grating camera-projector system and FTP. The center of the circle is traced dynamically and the coordinate system is transformed with the center of the circle as the origin. The phase distribution of concentric-circle grating is obtained by coordinate system transformation and FTP. By exploring the relationship between phase and relative height, the time-varying curve of the object surface height is recovered, which contains the vibration information. The experiments compare the measurement ability of the concentric-circle and linear grating. The results show that the linear grating and concentric-circle grating with the similar amplitude resolution have different amplitude measurement range. In this experiment, the maximum measurement range of linear grating is about 2.4 mm, while that of concentric-circle grating is more than about 3.9 mm. It means that the concentric-circle grating projection method can measure a large dynamic range of amplitude under the condition that the center of the circle can be traced, and solve the problem of amplitude limitation of linear grating in the use of FTP.

Accurately detecting man-made intrusion from different events is of great significance for distributed optical fiber sensing perimeter security system. Most traditional algorithms lack the ability to reject various events of unknown class which are mainly from natural disturbance, and greatly decline the accuracy of intrusion recognition in field application. In order to solve this problem, we proposed a novel robust intrusion event recognition scheme based on convolutional prototype network (CPL), which realized end-to-end feature extraction and recognition based on the similarity of intrusion signals by integrating relevant variables of prototype learning into the training process of multiscale convolutional neural network (MSCNN) as trainable parameters, and had the ability to recognize and reject the unknown disturbance events. In field experiments, the average recognition accuracy of intrusion events as known class can reach 84.67%, with the rejection rate of disturbance events as unknown class is about 83.75%, which ensure the accuracy of intrusion events monitoring in complex field environments. And the recognition response time is about 17 ms, which also meets the need of real-time monitoring.

Intrusion monitoring based on optical fiber distributed sensing is a pattern recognition problem of target events. Due to the addition of uncontrollable unknown events in practical applications, the open-set problem brought about greatly reduces the accuracy of traditional events recognition methods. In order to solve this problem, this paper design a deep metric-learning network, combined with Recurrent Plot (RP) coding to improve the accuracy of target events recognition in an open environment. In this paper, the RP algorithm is used to encode the intrusion signals into images that reflect the signal’s motion. The deep metric-learning network is used to project the image into the feature space. And feature centers of each class can be calculated rely on the training samples. So, the network identifies the intrusion event according to the distance between the test sample and the feature center. By setting the appropriate threshold, the samples whose distance exceeds the threshold are identified as unknown class to solve the open-set problem. In the experiments, the dual Mach-Zehnder interferometry (DMZI) distributed optical fiber perimeter system is built to collect event signals. For 7 known event classes the highest identification accuracy can reach 99.7%. After adding 3 unknown event classes, the accuracy rate can exceed 93.3% with appropriate threshold. And the total response time is only 0.4 s. This novel method guarantees the accuracy of target identification under the introduction of unknown events and improves the robustness of the distributed optical fiber perimeter security system.

This article proposes an intrusion pattern recognition scheme based on Gramian Angular Field (GAF) and convolutional neural network (CNN) for the dual Mach–Zehnder Interference (DMZI) distributed fiber perimeter security system, which has the advantages of fast recognition speed and high recognition accuracy rate. Compared with traditional recognition algorithms lack the ability to extract deep features, GAF algorithm is used to transform the intrusion signals of 1-D time series to 2-D images which can present more deep features and maintain the time domain dependence of the signals, meanwhile, each intrusion signal corresponds to a unique fingerprint. Then, this GAF algorithm combined with the appropriate CNN model is able to extract deep features and recognize the intrusion patterns accurately. Besides, GAF algorithm is insensitive to the fluctuation of power source in the optical path, so the robustness as well as the practicability of the system are improved effectively. The experiments demonstrated that the average recognition accuracy rate of three common natural intrusion events and three human intrusion events on fence (wind blowing, light rain, heavy rain, knocking, impacting and slapping) can reach to 97.67%, and the detection response time is only about 0.58 s, which can meet the requirements of real-time emergency monitoring.

In order to enhance the image quality of multi-reflective 3D scenes during machine vision analysis, an active self-adaptive illumination system to solve the overexposure and underexposure problems is introduced. Based on projector-camera system, adaptive illumination was achieved by modulating the projection brightness at each pixel to be proportional to the pixel brightness of the target scenes, which should firstly establish the accurate pixel correspondence between the projector and camera. In this paper, the performance of adaptive illumination is improved by optimizing the correspondence stage compared with previous research. Different grating patterns with different frequency components are projected and compared to acquire the most effective correspondence, which improves the accuracy of corresponding depth map and the experiment results can achieve higher lighting accuracy reducing the number of iterations as well as the computing time. The image quality indexes, image contrast and number of saturated or dark pixels, are combined to analyze and judge the optimized lighting result. The results demonstrate that the developed technology in this research can achieve a rate of 20 frames per second by using a standard commercial CCD (charge coupled device) camera, where it can run in real time and becomes a viable solution for the industry.

A centimeter-sized bearing fault probe based on dual-fiber Bragg grating vibration sensing is proposed. The probe can provide multi-carrier heterodyne vibration measurements based on swept source optical coherence tomography technology and the synchrosqueezed wavelet transform method to obtain a wider vibration frequency response range and collect more accurate vibration data. For the sequential characteristics of bearing vibration signals, we propose a convolutional neural network with long short-term memory and transformer encoder. This method is proven in bearing fault classification under variable working conditions, and the accuracy rate reaches 99.65%.

A noncontact rapid microvibration measurement method is proposed based on multichannel projector-camera system and phase shift profilometry. The multichannel characteristics of color image can be used to calculate vibration information dynamically by collecting only one piece of image. The waveform within a frequency range from 0.1 to 2 Hz is measured accurately by means of this vibration measurement technology, and minimum vibration amplitude is about 163 um. The system can be achieved by a standard low cost commercial charge coupled device camera with 30 frames/s.

In order to modify the local luminance heterogeneity of target object image, restrain the phenomenon of overexposure and preserve enough original image information, we analyzed the adaptability of the adjustable projection which is applied to exposure compensation. In this paper, a novel iterative method by using the improved algorithm for absolute center moment (IACM) is proposed. A projector-camera system with the iterative algorithm is built to adjust the adaptive projection according to the inversion theory, the differences of depth information and the illumination intensity. Among them, the role of depth information is correspond to different light intensity at different heights. Through the experiment, the result shows that the camera could achieve uniform luminance and the index convergence is evaluated by IACM. This system can output the image with better definition based on iterative stop criterion, thereby the number of iterations is limited and objectively countable. This technique is suitable for industrial measurement, defect detection and quality control in real-time measurement.

The ability of distributed Bragg reflector (DBR) fiber laser sensor in dual-frequency ultrasound (DFUS) measurement has been demonstrated. The influences of DFUS pressure on fiber grating laser sensor were theoretically analyzed, considering the effects of relative distance of the ultrasound field and the amplitude of the DFUS on the degrees of birefringence modulation. In experiment, the birefringence in sensing fiber was modulated by ultrasound signals at 3 MHz and 5 MHz, and the test distance and driving voltage were set respectively according to the ultrasound frequency. Agreement between theoretical and experimental results was obtained for ultrasound wave propagating from different distances (40 mm to 10 mm) and different DFUS driving voltage (5 V to 20 V). The results demonstrate that the DBR fiber grating laser acoustic sensor has a multi-frequency ultrasound recognizable ability, offering a potential for ultrasound medicine and biology

An illumination optimisation technique applied to multi-reflective 3-D machine vision based on a projector-camera system is introduced, in which the projector plays a key role to compensate for surface reflectance at each pixel to be inversely proportional to the brightness of the pixel under ambient light. The adaptive illumination technology was achieved by iterations emphasising different illumination intensities according to different surface orientations and requiring an accurate correspondence between the projector pixels and the camera pixels. In order to establish the most effective correspondence to prepare for subsequent adaptive illumination, 4 kinds of grating patterns, including sinusoidal, rectangular, triangular, and dual-frequency sinusoidal grating patterns, were projected and compared. The iterations were halted when an optimally lit scene was obtained; the further experiments under weak and strong light searched for the best method of illumination optimisation and confirmed the reliability of the adaptive illumination. The proposed optimisation design could run in real time and became a viable solution for industry.

For establishing an accurate and reliable distributed optical fiber perimeter security system, this paper proposes a novel abnormity detection solution to security using Recurrent Plot (RP) and deep learning technology. Take advantage of the temporal correlation of intrusion signals, we encode the sensing signals into two-dimensional images through the RP algorithm. The RP algorithm can extract the motion characteristics of the signal from the complex time series, and it is robust to instrument noise. These encoded image signatures can reveal the deeper temporal correlation of the intrusion signals’ motion. After that, Inception network can adaptively extract the features of these images to complete the accurate identification of a series of noisy intrusion signals. We conducted experiments on three most frequent natural events and three representative man-made intrusion events, including heavy rain, light rain, wind blowing, treading, slapping, and impacting. The results show that the detection accuracy has reached 99.7%. This method can achieve 0.35 s real-time detection in the online detection of abnormal events while ensuring accuracy, providing a new intrusion pattern identification idea for perimeter security.

As a type of pollution, low-frequency vibrations in the environment have a negative effect on human life. To monitor pollution without extra complications, we propose a novel method based on visual information perception and neural network analysis. Image frame sequences can capture the state of objects, and changes in pixel values at salient vibration points may indirectly reflect ambient vibrations. In this paper, a test environment is established to simulate the influence of vibrations on daily life. A CCD camera is used to continuously sample objects via image frame sequences to monitor potential vibration pollution in the current environment through image processing and neural network analysis. During the image processing stage, a combination of image filtering and edge extraction can accurately locate the test objects’ contours. During the low-frequency vibration monitoring stage, the neural network analyzes the changes in pixel values at the contour points to monitor ambient vibration pollution. The network combines local and global features for vibration frequency classification and prediction. The proposed method's superiority is verified by the test results and a comparison of four different networks. The results demonstrate that this method accurately locates salient vibration points to monitor ambient vibration pollution.

Sputtered titanium nitride (TiN) with a novel N2 ratio adjustment and rapid thermal annealing (RTA) treatment are proposed in order to optimize pH sensing performance of electrolyte-insulator-semiconductor (EIS) structure. Selection of this methodology, which can be easily applied into standard CMOS and DRAM technology, results from the fact that TiN is a well-verified material as a buffer or a barrier layer. It was concluded that pH sensitivity of the order of 60.5 mV/pH and linearity of 99.9% could be obtained for a TiN/SiO2 EIS structure treated with N2 ratio of 20% and RTA at 800 °C, which could be a good candidate in sensor applications. This performance is stable for more than two months. Higher surface roughness shown in atomic force microscope (AFM) analysis and high oxygen level in sputtering process shown in x-ray photo photoelectron spectroscopy (XPS) could be the reasons for the high sensitivity of the fabricated TiN sensing layer.

This letter reports an analysis of the potential of distributed Bragg reflector (DBR) fiber laser sensor designed to monitor the force orientation information, and with the birefringence characteristics of optical fiber, enabling the measurement of a directional lateral force applied to the fiber. In system design, to interrogate birefringence variation of the sensor according to the change of applied force angles, a reference lasing signal was employed following the results of an analysis using heterodyning theory. Agreement between theoretical and experimental results was obtained for a constant lateral force loading from different directions corresponding to the variation of sensitivity in beat signals. The design created in this letter is based on fundamental principles and generic, thus suitable for use with any DBR fiber laser structure for vector sensor applications.

Based on the study of self-Q-switching of a short cavity erbium-doped distributed Bragg reflector (DBR) fiber laser in time domain, the influence of pressure disturbance on the self-Q-switched output is analyzed. In the experiment, the erbium-doped DBR fiber laser (EDFL) was encapsulated with epoxy resin and curing agent to protect and sensitize the fiber. The EDFL sends out self-Q-switched laser pulses when the pump power of the 980 nm laser source is more than 110 mW. The external pressure is exerted on the DBR fiber and increased gradually from 0 g (0 N) to 120 g (1.176 N) in 10 g unit with the pump power fixed at 360 mW, and the following approximately linear changes of self-Q-switched laser pulses outputs are observed by the oscilloscope: the peak voltage of output pulses reduces from 12.59 mV to 1.883 mV; the pulse repetition rate reduces from 125.0 kHz to 66.67 kHz; and the pulse width increases from 0.4083 μs to 1.28 μs.

Up to now, real-time two dimensional (2D) chemical images are not available in a simple platform, results from big data with low generating rate by complicated process including movement of light source and ultra-low current measurement. A synchronization of all inputs including coordination of scanning step, ac signal of laser spot, and dc bias of LAPS is achieved by a field-programmable gate array (FPGA)-based system and self-developed LabVIEW program. Photocurrent, pH response, 2D pH images of different resolution of Si3N4/SiO2 LAPS can be easily obtained by this system. It could be integrated with microfluidics for a real-time Lab-on-chip with biomedical sensor array.

An ultrasonic hydrophone based on a dual polarization distributed Bragg reflector (DBR) fiber laser is described, and its application to detecting the vector medical ultrasound is demonstrated. The principle of the hydrophone is based on the detection of output beat frequency signal modulated by ultrasound. The amplitude, frequency and orientation of the ultrasound can be determined by the using the upper and lower sideband frequency. It has been found that the hydrophone has an orientation recognizable ability which the piezoelectric ultrasonic immersion transducer doesn’t have. It suggests that the type of hydrophone can provide an alternative to piezoelectric hydrophone technology.

Distributed Bragg Reflector (DBR) polarimetric optical fiber laser sensors have been attracting great interests due to harsh environment capability and high signal-to-noise ratio .We demonstrate directional force measurement technology using dual-polarization DBR optical fiber laser as a sensor. The influences of external force (bending, current and ultrasonic signal) could be analyzed by inducing a DFB polarimetric laser sensor to detect the beating signals shifts when the cavity is perturbed. We present the analysis of sensing mechanism on the DBR geometric construction and demodulation of directional information by separating the dual orthogonal polarization modes. With loading angles at 35°, 45°, 55°with 10° interval, the DBR laser sensor has shown orientation recognization ability corresponding to beating signals shifts, offering a potential for vector force directional detection.

The self Q-switched characteristic of the single longitudinal-mode distributed Bragg reflector (DBR) erbium-doped fiber laser (EDFL) with narrow linewidth pulse output is demonstrated without any extra saturable absorber or electro-optic and acoustic-optic modulator, and it is provided with a simpler structure and lower cost. The proposed pulse laser operates at approximately 1550.33 nm with a 20 dB bandwidth of 0.56 nm and signal-to-noise ratio of about 55 dB. By increasing the 980 nm pump power from 80–330 mW, the repetition rate of the output pulse increases from 73.53–147.1 kHz while the pulse width decreases from 1.4–0.2 µs, which shows the feature of the narrow linewidth ideally. The maximum pulse energy of 0.379 nJ and peak power of 1897.2 µW are generated with a repetition rate and pulse width of 147.1 kHz and 0.2 µs, respectively. The result shows that the self Q-switched pulse of the DBR EDFL is obviously dependent on the applied pump power and possesses the characteristic of the single longitudinal mode and the narrow linewidth, offering potential utilization in industrial processing and scientific research.

We introduce a technique for illumination optimization based on a projector–camera system applied to a multireflective three-dimensional (3-D) scene. Simple mapping would fail if the scene is nonplane since they would translate to incoherent mapping between the projector and the camera. The correspondence between the pixels on the projector and camera in multireflective 3-D scene was indirectly established by surface orientation measurement. Once the correspondence was established, the local light was adapted by using the iteration in order to obtain the optimal illumination intensities at different surface orientations. User intervention for judging the desirable image is required in the iteration stage since there is no absolute reference for comparison to which the object must appear. According to the feedback from the camera, the projector is to compensate for surface reflectance for the most part by modulating the projection brightness at each pixel to be inversely proportional to the brightness of the pixel under uniform light. The time required for this adaptive illumination can be shortened to the inspection stage of 40 ms as long as the inspected object does not change in orientation or location. This proposed technique aims to enhance the illumination capabilities and presents a more unified system than existing methods.

A dual-frequency ultrasound measurement system based on a distributed Bragg reflector (DBR) fiber laser sensor in a liquid medium was presented. To compare the dual-frequency measurement performance of a DBR fiber laser acoustic sensor with that of a piezoelectric (PZT) ultrasound sensor, two experiments were performed. First, we fixed the driving frequencies of two ultrasound signals at 3 and 5 MHz, and decreased the driving voltage from 15 to 3 V. The outputs of the DBR acoustic sensor show flat-balanced response to dual-frequencies, compared with the PZT acoustic sensor whose response to one of the dual-frequency signals (5 MHz in this paper) has been covered by noise at low acoustic pressure. Then we increased the acoustic pressure by fixing the driving voltage at 20 V, and changed the frequency spacing between the two ultrasound signals. By analyzing the frequency response, sensitivity, signal-to-noise ratio, and noise equivalent pressure of two acoustic sensors under different frequencies, we found that the response of the DBR sensor to wideband dual-frequency is stable, while the response of the PZT sensor deteriorates sharply with increasing frequency spacing. The results demonstrate that the DBR fiber laser sensor performs better for wide bandwidth dual-frequency ultrasound measurements.

A DBR fiber grating laser acoustic sensor based on polarization beat signal modulation analysis has been demonstrated for directional acoustic signal measurement. The acoustic sensor was fabricated in birefringent erbium-doped fiber, and the influences of external-acoustic pressure on fiber grating laser sensor were analyzed, considering the effect of relative orientation of the acoustic wave on the degrees of birefringence modulation. In experiment, the birefringence in sensing fiber was modulated by ultrasonic pressure. Agreement between theoretical and experimental results was obtained for ultrasound wave propagating from different directions (0-360 degrees in 15 degrees intervals) corresponding to a nonlinearly change in beat frequency modulation rates. The results demonstrate that the DBR fiber grating laser acoustic sensor has an orientation recognizable ability, offering a potential for acoustic vector signal detection.