On-line Detection System Research Articles

Application Research of Portable On-line Inspection System for Car Lights

In this article, a portable automotive lighting online detection system is designed, which can test various signals of car lamps online. It also can perform function and performance tests on multiple car lamps at the same time. The system has a compact structure in order to be easy to carry. It can be tested in the car light laboratory or be used for on-line testing of the production line. The system can automatically complete data collection and processing online. It can discriminate test results and print test reports easily. It has strong versatility and high measurement accuracy. This portable online detection system is convenient for detection and has good application value.

Bloodstain Identification Based on Visible/Near-Infrared Hyperspectral Imaging With Convolutional Neural Network

Blood samples are easily damaged in traditional bloodstain detection and identification. In complex scenes with interfering objects, bloodstain identification may be inaccurate, with low detection rates and false-positive results. In order to meet these challenges, we propose a bloodstain detection and identification method based on hyperspectral imaging and mixed convolutional neural networks, which enables fast and efficient non-destructive identification of bloodstains. In this study, we apply visible/near-infrared reflectance hyperspectral imaging in the 380-1000 nm spectral region to analyze the shape, structure, and biochemical characteristics of bloodstains. Hyperspectral images of bloodstains on different substrates and six bloodstain analogs are experimentally obtained. The acquired spectral pixels are pre-processed by Principal Component Analysis (PCA). For bloodstains and different bloodstain analogs, regions of interest are selected from each substance to obtain pixels, which are further used in convolutional neural network (CNN) modeling. After the mixed CNN modeling is completed, pixels are selected from the hyperspectral images as a test set for bloodstains and bloodstain analogs. Finally, the bloodstain recognition ability of the mixed 2D-3D CNN model is evaluated by analyzing the kappa coefficient and classification accuracy. The experimental results show that the accuracy of the constructed CNN bloodstain identification model reaches 95.4%. Compared with other methods, the bloodstain identification method proposed in this study has higher efficiency and accuracy in complex scenes. The results of this study will provide a reference for the future development of the bloodstain online detection system.

Real-Time Terahertz Characterization of Minor Defects by the YOLOX-MSA Network

Terahertz (THz) imaging has been widely used in industrial non-destructive testing (NDT) of nonpolar materials owing to its unique property. Minor defect detection via THz NDT at high accuracy and fast speed is essential for industrial on-line detection systems. However, traditional defect detection algorithms can’t meet the demand of real-time high-precision detection of minor defects. Therefore, based on the YOLOX algorithm and multi scale attention (MSA) mechanism, the modified YOLOX network called YOLOX-MSA is proposed as a one-stage minor defect detection framework to improve the detection accuracy while supporting the real-time operation. The proposed YOLOX-MSA network improves the mean average precision (mAP) by at least 11.65% on the PCBs dataset with THz characteristics when the Intersection over Union (IoU) is 0.75. Additionally, the proposed algorithm can reach to the detection speed as 24~25 Frame Per Second (FPS). Overall, our proposed method can be beneficial to generalize the THz NDT in the frequency domain on the minor defects of nonpolar material, which will fulfill the impending requirements of real-time defect detection for industrial applications.

An on-line detection system for screening small molecule inhibitors of α-Amylase and α-Glucosidase in Prunus mume

High-throughput screening of inhibitors from natural products is an efficient approach to target key enzymes in diabetes progression. In this study, an on-line detection system was established for the first time to rapidly screen inhibitors of α-amylase and α-glucosidase from Prunus mume. Among 28 identified compounds, 26 and 21 compounds showed strong inhibitory effect against α-amylase and α-glucosidase, respectively. Their inhibitory effects were validated by in vitro enzyme assay and fluorescence quenching which demonstrated that these inhibitors effectively interfered enzyme active sites. The inhibition kinetics suggested that chemical structures are of great importance for interfering the enzyme structures and their microenvironment polarity. Among evaluated compounds, isorhamnetin-3-O-glucoside (19) showed the strongest binding activities to α-amylase and α-glucosidase (6.34×106·nmol−1 and 6.28×106·nmol−1, respectively) by the on-line detection system. Its IC50 values were 0.16 ± 0.06 and 0.09 ± 0.01 µM against α-amylase and α-glucosidase, respectively. 19 gave a much higher Ki for α-amylase (0.1307 mM) than α-glucosidase (0.0063 mM), indicating its selectivity towards α-glucosidase. This reported method was rapid and reliable to identify prototype inhibitors against key enzymes in diabetes, and thus might serve as a general platform to screen enzyme inhibitors from natural products.

Online detection of blood‐spot eggs based on a novel method of spectral amplitude space conversion

AbstractTo improve the robustness of the online detection system for blood‐spot eggs, a linear converter was established based on Levenberg–Marquardt (LM) algorithm to transform spectral signals in the amplitude space, so that error introduced from the variation of sample, mechanical device, and other factors can be reduced to a certain degree. First, the transmission spectrum in the 500–599 nm bandwidth of each sample was selected. After preprocessed with the moving average filtering, the spectrum was transformed in the space of spectral amplitude using the LM algorithm. Then, the converted spectrum was utilized to establish the backpropagation neural network model. As the result, the classification model using the transformed spectra realized more than 96% accuracy. When the detection system dealt with different batches of samples produced on different dates, the discriminant accuracy using LM converter was significantly higher than that without (p value < .05). Meanwhile, when the detection system ran at different speeds, the model still showed good robustness, and the accuracy was increased to some degree with LM converter. The research indicated the technical superiority of the model built with the proposed spectra converter over the model without and provided a feasible method to detect blood‐spot eggs online for commercial usage.Practical ApplicationsBlood‐spot eggs need to be detected and removed before sale or export. During the task of online detection, the variation of sample batch, conveyor speed, and other factors can introduce error to the predicting outcomes of the classification model. Thus, it is essential for the detection system to stay robust and realize high‐accuracy identification under variable external factors. To this end, a novel linear converter for transforming spectral signals based on Levenberg–Marquardt algorithm was studied in this paper. The results showed that the proposed converter could increase the classification accuracy and robustness of detection system to a certain degree, thus realizing accurate and fast online detection of blood‐spot eggs.

Detection and Classification System for Rail Surface Defects Based on Eddy Current.

The prospect of growth of a railway system impacts both the network size and its occupation. Due to the overloaded infrastructure, it is necessary to increase reliability by adopting fast maintenance services to reach economic and security conditions. In this context, one major problem is the excessive friction caused by the wheels. This contingency may cause ruptures with severe consequences. While eddy’s current approaches are adequate to detect superficial damages in metal structures, there are still open challenges concerning automatic identification of rail defects. Herein, we propose an embedded system for online detection and location of rails defects based on eddy current. Moreover, we propose a new method to interpret eddy current signals by analyzing their wavelet transforms through a convolutional neural network. With this approach, the embedded system locates and classifies different types of anomalies, enabling an optimization of the railway maintenance plan. Field tests were performed, in which the rail anomalies were grouped in three classes: squids, weld and joints. The results showed a classification efficiency of ~98%, surpassing the most commonly used methods found in the literature.

An online detection system of energy efficiency for intelligent manufacturing

Online detection of energy efficiency automatically has important significance for evaluating new production lines and detecting the performance of existing production lines in modern intelligent manufacturing and smart industry. To solve the shortcomings of existing detection systems, a novel online detection system of energy efficiency is proposed in this paper. Within a time interval, the proposed system detects the production quantity and the energy consumption of the production line by a production quantity detection module and a power consumption sensing module, then calculates and obtains the production efficiency during the period and displays to users by output module. Experimental results show that the proposed system has high practical value.

Study on Size Distribution and Flow Characteristics of Condensed Products in Solid Rocket Motor

The size distribution of condensed products during the combustion of aluminized propellants and flow characteristics of the gas-solid two-phase flow in solid rocket motor were studied in this paper. Firstly, based on the laser scattering technology, an online detection system for condensed products in plume was established, and the size detection of condensed products in the plume of solid rocket motor is carried out. Secondly, a numerical model of two-phase flow in solid rocket motor is established by combining the real size distribution of products in the plume with discrete phase model through the Rosin-Rammler distribution function. Besides, numerical simulation research is carried out under the same experimental conditions, focusing on the influence of condensed products with real size on the characteristics of solid rocket motor. The results show that the innovation measurement system can be used to obtain the size distribution characteristic of condensed products in the plume. At the particle size of stable stage, the mean size, D v 50 , is 104 μm, which is the smallest among all stages. It is also suggested that condensed products at the end stage have the most impact on the flow behavior in solid rocket motor, in that the shock structure, Mach number, and temperature distribution in the near field of plume are significantly changed.

On-line concurrent control chart pattern recognition using singular spectrum analysis and random forest

A scheme integrated with engineer process control (EPC) is a useful approach to identify the disturbances for statistical process control (SPC) charting of an auto-correlated process. Although the EPC is able to compensate for the underlying disturbance, the underlying disturbance is embedded in the control chart. Thus, this makes the control chart patterns (CCPs) difficult to be identified. As documented in the literature, much efforts have been put in the recognition of single basic patterns of unnatural variation. However, there could exist a combination of two unnatural patterns simultaneously in a real-world manufacturing process. Also for an automated real-time production line, data are collected automatically and monitored by a computer-based system. Thus, the early detection of abnormality is of high importance for the aforesaid problem. This study aims to establish an on-line detection system used for monitoring the mixture unnatural CCPs for a SPC-EPC process. This paper presents a hybrid approach based on singular spectrum analysis (SSA) and random forest (RF) to identify the concurrent CCPs in an on-line SPC-EPC process. A total of fifteen types of concurrent CCPs were utilized to validate the proposed method. The SSA method was also used to decompose the mixture patterns into single patterns. The RF was employed to identify the types of patterns to which it belongs. The results showed that the proposed method was able to handle most of the concurrent CCPs very successfully with an average accurate identification rate of 91.8%. Also, the proposed method was found to be more accurate and efficient than the use of the hybrid method of SSA and support vector machine (SVM). It is suggested that this proposed system could be possibly applied for monitoring an on-line process to a greater extent.

Research on Key Technologies of Robot for Visual Detection of Foreign Matter in Liquid Medicine Based on Machine Learning Algorithm

At present, in the pharmaceutical industry, people pay more and more attention to the quality of drugs, and therefore put forward higher requirements for the detection of drugs. In this paper, based on a large number of theories and experiments, an intelligent on-line detection system for visible foreign bodies in medicated liquor solution is designed based on machine learning algorithm. From the angle of mechanics and kinematics, the force analysis of moving objects in liquid after high-speed rotation and sudden stop is discussed, which provides an important reference standard for distinguishing visible foreign objects from random noise. The experimental results show that the algorithm can effectively eliminate the vibration interference of electromechanical system, and make the foreign body detection equipment of liquid medicine successfully migrate from single-station detection equipment to automatic online detection robot, which is of great significance to the research and development of the whole equipment.

Identification of soybean varieties based on hyperspectral imaging technology and one‐dimensional convolutional neural network

AbstractVariety identification of seeds is essential to ensure the purity and yield of the variety. A model based on hyperspectral imaging technology and one‐dimensional convolutional neural network (1D CNN) was proposed to distinguish soybean seed varieties in this article. A total of 3,600 soybean seeds (900 seeds per variety) of hyperspectral images in the spectral range of 866.4–1,701.0 nm were collected. Traditional machine learning models (k‐nearest neighbor, support vector machines, partial least squares discriminant analysis) and 1D CNN model were established based on different numbers of sample sets. The 1D CNN model was the most stable and had the highest classification accuracy, higher than 98%. Fivefold cross validation was used to evaluate the model, and the model achieved an accuracy of more than 95% in both the training set and the validation set. Finally, the t‐distributed stochastic neighbor embedding was used to visualize the feature values extracted by 1D CNN. The research results demonstrated that the 1D CNN model maintained good classification performance in soybean seed classification and had good application prospects in hyperspectral imaging technology.Practical ApplicationsVariety identification of soybean seeds is essential to ensure the purity and yield of the variety. Different soybean varieties have different genetic purity, physical purity, germination ability and vigor, which are related to quality attributes, such as nutritional value, stress resistance, and final yield. In the past, most soybeans were identified using traditional machine learning algorithms combined with near‐infrared hyperspectral imaging technology to build models. This requires a series of operations, such as smoothing preprocessing and dimensionality reduction on hyperspectral data. These steps are too cumbersome and are not conducive to online hyperspectral monitoring systems. The results of this study show that it is feasible to use one‐dimensional convolutional neural network combined with hyperspectral technology to identify soybean seeds, and it also provides a new idea for building an online detection system in the future.

Partial Discharge Online Detection for Long-Term Operational Sustainability of On-Site Low Voltage Distribution Network Using CNN Transfer Learning

Partial discharge (PD) detection studies aiming at the fault diagnosis for facilities and power cables in transmission networks have been conducted over the years. Recently, the deep learning models for PD detection have been used to diagnose the PD fault of facilities and cables. Most PD studies have been conducted in the field, such as gas-insulated switchgear (GIS) and power cables for high voltage transmission networks. There are few studies of PD fault detection for on-site low-voltage distribution networks. Additionally, there are few studies of PD detection algorithms for improving the accuracy of the deep learning models using small real PD data only. In this study, a PD online detection system and a model for long-term operational sustainability of on-site low voltage distribution networks are proposed using convolutional neural network (CNN) transfer-learning. The proposed PD online system makes it possible to acquire as many real PD data as possible through continuous monitoring of PD occurrence. The PD detection accuracy results showed that the proposed CNN transfer-learning models are more effective models for obtaining improved accuracy (97.4%) than benchmark models, such as CNN and support vector machine (SVM) using only small real PD data acquired from PD online detection system.

Flexible Noncontact Sensing for Human–Machine Interaction

From typical electrical appliances to thriving intelligent robots, the exchange of information between humans and machines has mainly relied on the contact sensor medium. However, this kind of contact interaction can cause severe problems, such as inevitable mechanical wear and cross-infection of bacteria or viruses between the users, especially during the COVID-19 pandemic. Therefore, revolutionary noncontact human-machine interaction (HMI) is highly desired in remote online detection and noncontact control systems. In this study, a flexible high-sensitivity humidity sensor and array are presented, fabricated by anchoring multilayer graphene (MG) into electrospun polyamide (PA) 66. The sensor works in noncontact mode for asthma detection, via monitoring the respiration rate in real time, and remote alarm systems and provides touchless interfaces in medicine delivery for bedridden patients. The physical structure of the large specific surface area and the chemical structure of the abundant water-absorbing functional groups of the PA66 nanofiber networks contribute to the high performance synergistically. This work can lead to a new era of noncontact HMI without the risk of contagiousness and provide a general and effective strategy for the development of smart electronics that require noncontact interaction.

Development of on-line lateral stiffness measurement system for anterior cruciate ligament and its influence on anterior cruciate ligament reconstruction

The anterior cruciate ligament (ACL) reconstruction mostly relies on the experience of surgeons. To improve the effectiveness and adaptability of the tension after ACL reconstruction in knee joint rehabilitation, this paper establishes a lateral force measurement model with relaxation characteristics and designs an on-line stiffness measurement system of ACL. In this paper, we selected 20 sheep knee joints as experimental material for the knee joint stability test before the ACL reconstruction operation, which were divided into two groups for a comparative test of single-bundle ACL reconstruction through the anterolateral approach. The first group of surgeons carried out intraoperative detection with routine procedures. The second group used ACL on-line stiffness measurement system for intraoperative detection. After that, the above two groups were tested for postoperative stability. The study results show that the tension accuracy is (- 2.3 ± 0.04)%, and the displacement error is (1.5 ± 1.8)%. The forward stability, internal rotation stability, and external rotation stability of the two groups were better than those before operation ( P < 0.05). But the data of the group using the system were closer to the preoperative knee joint measurement index, and there was no significant difference between them ( P > 0.05). The system established in this paper is expected to help clinicians judge the ACL reconstruction tension in the operation process and effectively improve the surgical effect.

Classification of Lingwu long jujube internal bruise over time based on visible near-infrared hyperspectral imaging combined with partial least squares-discriminant analysis

Early detection of internal bruise is one of the major challenges in postharvest quality sorting processes in Lingwu long jujube. In this study, the visible/near infrared (VIS/NIR) hyperspectral imaging system (400–1000 nm) was used to rapidly detect the intact and damaged jujube at five time points after mechanical damage (2 h, 4 h, 8 h, 12 h and 24 h). The region of interest of samples was selected by ENVI software, and the average spectrum was calculated for modelling. Different preprocessing methods were used to transform and enhance the spectral signal. Partial least squares-discriminant analysis (PLS-DA) classification models of the original and preprocessed spectra were established. The de-trending-PLS-DA model had the best effect, the accuracy of the calibration set and prediction set were 85.56% and 92.22%, respectively. The spectra were pre-processed by the de-trending algorithm, and the successive projection algorithm (SPA), uninformative variable elimination (UVE), competitive adaptive reweighted sampling (CARS), interval variable iterative space shrinkage approach (iVISSA), cluster analysis (CA), two-dimensional correlation spectra (2D-COS), UVE-SPA, CARS-SPA, iVISSA-SPA, CA-SPA and 2D-COS-SPA method were used to select characteristic variables. The PLS-DA model based on feature variables was established. The de-trending-CARS-PLS-DA model with 63 variables was found to be the optimal model. In the de-trending-CARS-PLS-DA model, the accuracy of calibration set and prediction set were 86.67% and 91.11%, respectively. It was found that the model accurately detected bruising in jujube at 8 h after bruising, and the accuracy of calibration set and prediction set were 100%. The results showed that VIS/NIR hyperspectral imaging technology combined with PLS-DA could discriminate different stages of bruising in Lingwu long jujube. This study may help develop an online detection system of bruising in Lingwu long jujube.

Classification of COVID-19 by Compressed Chest CT Image through Deep Learning on a Large Patients Cohort.

Corona Virus Disease (COVID-19) has spread globally quickly, and has resulted in a large number of causalities and medical resources insufficiency in many countries. Reverse-transcriptase polymerase chain reaction (RT-PCR) testing is adopted as biopsy tool for confirmation of virus infection. However, its accuracy is as low as 60-70%, which is inefficient to uncover the infected. In comparison, the chest CT has been considered as the prior choice in diagnosis and monitoring progress of COVID-19 infection. Although the COVID-19 diagnostic systems based on artificial intelligence have been developed for assisting doctors in diagnosis, the small sample size and the excessive time consumption limit their applications. To this end, this paper proposed a diagnosis prototype system for COVID-19 infection testing. The proposed deep learning model is trained and is tested on 2267 CT sequences from 1357 patients clinically confirmed with COVID-19 and 1235 CT sequences from non-infected people. The main highlights of the prototype system are: (1) no data augmentation is needed to accurately discriminate the COVID-19 from normal controls with the specificity of 0.92 and sensitivity of 0.93; (2) the raw DICOM image is not necessary in testing. Highly compressed image like Jpeg can be used to allow a quick diagnosis; and (3) it discriminates the virus infection within 6 seconds and thus allows an online test with light cost. We also applied our model on 48 asymptomatic patients diagnosed with COVID-19. We found that: (1) the positive rate of RT-PCR assay is 63.5% (687/1082). (2) 45.8% (22/48) of the RT-PCR assay is negative for asymptomatic patients, yet the accuracy of CT scans is 95.8%. The online detection system is available: http://212.64.70.65/covid.Graphic

Effect of Stress-Sensitive Permeability and Porosity on Production Performance in Water-Soluble Gas Reservoirs

Abstract Water-soluble gas reservoirs have the characteristics of high temperature and high pressure (HTHP) and experience obvious pressure-sensitive effects during the production process. Therefore, the influences of formation water and dissolved natural gas in formation water on water-soluble gas reservoirs are different from conventional gas reservoirs. In view of this, this work first carried out a stress sensitivity test with irreducible water and variable internal pressure at high temperature for a water-soluble gas reservoir, showing that permeability loss ratio and effective stress have an exponential relationship, a result basically consistent with conventional tests. However, the stress sensitivity test result with irreducible water was greater than the stress sensitivity test result without irreducible water; porosity decreased slightly with increasing confining pressure, and the total decrease ratio was less than 5.2%, with an average of 3.01%. Second, a high-pressure, high-temperature, nuclear magnetic resonance (NMR) online detection system was introduced to detect the pore signal of core samples under different effective stress states, and pore compression and deformation characteristics were evaluated. Results show large pores to have been compressed slightly more than small pores, pores to be significantly compressed in the initial stage, and the greater the increase in effective stress, the more obvious the compression. Third, the occurrence and characteristic changes of irreducible water in the process of rock compression were detected by the NMR online system, indicating irreducible water to be difficult to migrate through compression in water-soluble gas reservoirs under slight compression of rock and pore structure and the occurrence and characteristics of irreducible water to have not changed significantly. Finally, by establishing a theoretical model of water-soluble gas reservoirs to simulate the water breakthrough of gas wells under stress sensitivity conditions, this work shows that when stress sensitivity exists, gas-well water breakthrough time is earlier and production is diminished.

A patient-independent classification system for onset detection of seizures.

Seizures are the most common brain dysfunction. Electroencephalography (EEG) is required for their detection and treatment initially. Studies show that if seizures are detected at their early stage, instant and effective treatment can be given to the patients. In this paper, an automated system for seizure onset detection is proposed. As the power spectrum of normal person's EEG and EEG of someone with epilepsy is plotted, powers present at different frequencies are found to be different for both. The proposed algorithm utilizes this frequency discrimination property of EEG with some statistical features to detect the seizure onset using simple linear classifier. The tests conducted on EEG data of 30 patients, obtained from the two different datasets, show the presence of all 183 seizures with mean latency of 0.9s and 1.02 false detections per hour. The main contribution of this study is the use of simple features and classifier in the field of seizures onset detection that reduces the computational complexity of the algorithm. Also, the classifier used is patient independent. This patient independency in the classification system would be helpful in the implementation of the proposed algorithm to develop an online detection system.

Design of an Online Detection System for COVID-19 based on Attention Mechanism

Corona Virus Disease 2019 (COVID-19) is an acute respiratory infection caused by the 2019 novel coronavirus infection that was discovered in some hospitals in Wuhan, China in December 2019. From a global perspective, COVID-19 epidemic is still in a pandemic period. Every country is taking timely preventive measures. Therefore, it is necessary to use Artificial Intelligence (AI) identification to assist radiologists in diagnosing COVID-19. This paper proposes a network model of dynamic self-attention machine for pneumonia images. The model adopts DPN92 and GCNet. It implements CR images and CT image classification of COVID-19. A COVID-19 detection system based on dynamic auto-attention machine is designed. The lung CT or CR images that meet the design requirements will be uploaded through the browser, and the background server will analyze and process the lung images, and finally output the detection category of the lung images.

High precision on-line liquid level detection system based on microwave penetration method

In order to meet the high precision and low cost of liquid level detection for bottled drinks in industrial production line, based on microwave transmission principle a liquid level detection method was designed on the basis of radar level meter. The rectangular microstrip antenna sent and received specific microwave signals. The signal was processed by the embedded ARM(Advanced RISC Machines) processor and digital signal processing method, and finally the data was transmitted. It was verified by experiments that the microwave penetration method can be used for liquid level detection, and the detection accuracy can reach ±1mm. Due to the use of microwave for measurement, the system has high stability and can be applied to the liquid level detection of bottled drinks on the production line.