Visual Detection Method Research Articles

A Visual Fault Detection Method for Induction Motors Based on a Zero-Sequence Current and an Improved Symmetrized Dot Pattern.

Motor faults, especially mechanical faults, reflect eminently faint characteristic amplitudes in the stator current. In order to solve the issue of the motor current lacking effective and direct signal representation, this paper introduces a visual fault detection method for an induction motor based on zero-sequence current and an improved symmetric dot matrix pattern. Empirical mode decomposition (EMD) is used to eliminate the power frequency in the zero-sequence current derived from the original signal. A local symmetrized dot pattern (LSDP) method is proposed to solve the adaptive problem of classical symmetric lattice patterns with outliers. The LSDP approach maps the zero-sequence current to the ultimate coordinate and obtains a more intuitive two-dimensional image representation than the time–frequency image. Kernel density estimation (KDE) is used to complete the information about the density distribution of the image further to enhance the visual difference between the normal and fault samples. This method mines fault features in the current signals, which avoids the need to deploy additional sensors to collect vibration signals. The test results show that the fault detection accuracy of the LSDP can reach 96.85%, indicating that two-dimensional image representation can be effectively applied to current-based motor fault detection.

Noninvasive and Individual-Centered Monitoring of Uric Acid for Precaution of Hyperuricemia via Optical Supramolecular Sensing.

Characterized by an excessively increased uric acid (UA) level in serum, hyperuricemia induces gout and also poses a great threat to renal and cardiovascular systems. It is urgent and meaningful to perform early warning by noninvasive diagnosis, thus conducing to blockage of disease aggravation. Here, guanidinocalix[5]arene (GC5A) is successfully identified from the self‐built macrocyclic library to specifically monitor UA from urine by the indicator displacement assay. UA is strongly bound to GC5A at micromolar‐level, while simultaneously excluding fluorescein (Fl) from the GC5A·Fl complex in the “switch‐on” mode. This method successfully differentiates patients with hyperuricemia from volunteers except for those with kidney dysfunction and targets a volunteer at high risk of hyperuricemia. In order to meet the trend from hospital‐centered to individual‐centered testing, visual detection of UA is studied through a smartphone equipped with a color‐scanning feature, whose adaptability and feasibility are demonstrated in sensing UA from authentic urine, leading to a promising method in family‐centered healthcare style. A high‐throughput and visual detection method is provided here for alarming hyperuricemic by noninvasive diagnosis.

Defect Detection for Mechanical Design Products with Faster R-CNN Network

The emergence of machine vision has promoted the automation of defect detection (DD) in the industrial field. Therefore, scholars at home and abroad have carried out a lot of research and exploration on the traditional visual DD method of mechanical design products. At the same time, this method has been widely used in the field of modern manufacturing due to its noncontact and fast detection speed. The traditional visual detection method is to use cameras, computers, and other equipment instead of people to detect the detected objects, although this method improves the production efficiency to a certain extent. However, this detection method is greatly affected by light, has a certain false detection rate, and has poor adaptability. The intelligent detection method based on deep learning developed on the basis of traditional vision is a further optimization of traditional visual detection methods. The rapid development of deep learning makes the advantages of visual DD more obvious.

Rolling circle amplification assisted dual signal amplification colorimetric biosensor for ultrasensitive detection of leukemia-derived exosomes

Tumor-derived exosomes as a liquid biopsy marker hold great promising for the accurate tumor diagnosis. However, the visual exosomes detection method with high sensitivity and convenience is still a challenge and highly valuable in the clinical application. Herein, we fabricated a colorimetric biosensor that could visually detect leukemia-derived exosomes, and dual amplified the signal based on the rolling circle amplification (RCA) assisted platform. To avoid the interference of components in serum, trigger-aptamer complex was firstly modified on the magnetic beads (MBs) to recognize exosomes, and then release trigger probe to initiate RCA reaction. Therefore, the trigger probe could produce long repeated sequences and hybridized with numerous signal probe. After that, GNPs-HRP (HRP modified gold nanoparticles) bound with signal probe and induced dramatic color change of TMB to achieve the sensing of exosomes. Benefiting from the RCA assisted dual signal amplification, the presented method showed a limit of detection as low as 100 particles/μL. Moreover, this method exhibited good specificity to distinguish healthy and leukemia patients, suggesting its great potential for clinical diagnosis.

A dual-signal fluorometric-colorimetric sensing platform and visual detection with a smartphone for the determination of β-galactosidase activity based on fluorescence silicon nanoparticles.

As one of primary biomarkers of ovarian cancer in early stages, β-galactosidase (β-Gal) is significant in the discovery and diagnosis of the disease. In this work, we constructed a multi-signal sensing platform based on silicon nanoparticles (Si NPs) for β-Gal activity determination. When β-Gal was introduced to the sensing system, 2-Nitrophenyl β-D-galactopyranoside (ONPG) could be converted to o-Nitrophenol (o-NP), which had a characteristic absorption peak at 416nm and the colorless solution turned yellow. The fluorescence emission of Si NPs at 450nm can be greatly quenched by o-NP as a consequence of inner filter effect (IFE). This dual-signal fluorometric and colorimetric determination approach could be utilized to detect β-Gal in the range of 2-120 U/L and 6-120 U/L. The limits of detection were 1.36 U/L and 1.07 U/L, respectively. This sensing platform could be successfully utilized to detect β-Gal in real samples. Additionally, a visual detection method was designed to achieve quantitative analysis of β-Gal with the assistance of the smartphone.

Research on Manufacturing Technology of New Energy Vehicle Based on NDT

For the detection of laser welding defects of new energy vehicles, the traditional detection methods are inefficient and have limited accuracy. The existing visual detection methods mainly use area array cameras and image processing algorithms for detection. The image acquisition efficiency of the area array camera is low. The traditional image processing methods are vulnerable to the strong reflection on the surface of new energy vehicles, and the recognition speed and accuracy are limited. Therefore, in this paper, the linear array high-resolution camera is used for two-dimensional image acquisition, and the deep learning method is used for the recognition of weld defects. Based on the single-stage target detection network, the YOLO network is optimized according to the particularity of weld defects. The weld data set and defect data set are prepared, and the data set is used to train the optimized YOLO model. The feasibility of the method is verified by experiments, which can meet the mainstream weld length detection standards in the automotive industry.

Rapid and visual detection of Mycoplasma synoviae by recombinase-aided amplification assay combined with a lateral flow dipstick

Mycoplasma synoviae (MS) is an important avian pathogen that has brought substantial economic losses to the global poultry industry. Fast and accurate diagnosis is one of the critical factors for the control of MS infection. This study established a simple, rapid and visual detection method for MS using a recombinase-aided amplification (RAA) combined with a lateral flow dipstick (LFD). The reaction temperature and time of the RAA-LFD assay were optimized after selecting the primers and probe, and the specificity and sensitivity rates were analyzed. The results showed that RAA could amplify the target gene in 20 min at a constant temperature of 38°C, and the amplification products could be visualized by LFD within 5 min. There was no cross-reaction with Mycoplasma gallisepticum (MG), Pasteurella multocida (P. multocida), Escherichia coli (E. coli), Newcastle disease virus (NDV), infectious bursal disease virus (IBDV), infectious bronchitis virus (IBV), and avian reovirus (ARV). Furthermore, the RAA-LFD assay exhibited high sensitivity with a detection limit of 10 copies/μL. A total of 128 clinical samples with suspected infection of MS were tested by RAA-LFD, PCR, and real-time fluorescence quantitative PCR (RFQ-PCR). The coincidence rate of the detection results was 95.3% between RAA-LFD and PCR, and 98.4% between RAA-LFD and RFQ-PCR. These results suggested that the RAA-LFD method established in the present study was easy to use and was associated with strong specificity and high sensitivity. This method was very suitable for the rapid detection of MS in clinical practice.

RPA-Cas12aDS: A visual and fast molecular diagnostics platform based on RPA-CRISPR-Cas12a method for infectious bursal disease virus detection

Infectious bursal disease (IBD), a major disease of birds, is caused by infectious bursal disease virus (IBDV). The disease can lead to immunosuppression, resulting in huge economic losses in the poultry industry. A specific, rapid, and simple detection method is important for the early diagnosis and prevention and control of IBDV. In this study, we established a naked-eye visual IBDV detection method, named “RPA-Cas12aDS”, by combining recombinase polymerase amplification (RPA) with CRISPR-Cas12a-based nucleic acid detection. The detection process can be accomplished in 50 min, and uncapping contamination can be avoided. The detection results can be observed under blue or UV light. We used the RPA-Cas12aDS method to detect IBDV in bursa of Fabricius tissue samples of chickens, and the results were consistent with those obtained using commercial RT-PCR kits. This method presents great potential for visual, rapid, and point-of-care molecular diagnostics of IBDV in poultry.

A visual method for threshold detection of sediment motion in a flume experiment without human interference

AbstractAn experimental investigation of the entrainment threshold of sediment grains conducted in a non‐tiltable flume is reported and an original processing method based on image correlation suitable for detection of sediment threshold of motion is presented. Unlike existing visual measurement methods, the proposed method is not affected by the individual judgement of the experimentalists, which may strongly impact the estimation of incipient motion. Three entrainment thresholds were exhibited. The method highlighted a first threshold for very low shear stress conditions that was interpreted as a nascent movement threshold. The other two thresholds of motion were related to distinct stages of surface mobility of the bed: one related to bedload threshold and the other to bedform threshold. Series of tests were first performed to study the sensitivity of the method to different parameters: angle of view of the camera, size of the image and acceleration rate of the linear ramp of flow velocity. A ratio of the image surface and the grain surface equal to 3000 was found to be the minimum size to perform reliable estimations of entrainment thresholds for all the sediments in the range of 0.328–2.648 mm tested. It was the only condition for the method to work. Once the method was validated, tests were performed with homogeneous sands and compared to results based on reference‐based and visual methods. The correlation method yields similar values for the bedload threshold to those obtained from reference‐based methods. Thus it could be an alternative to other methodologies by the simplicity of its implementation, precision, no requirement for human interference and direct correlation between threshold velocity and bed mobility without extrapolation.

Attention Deeplabv3 model and its application into gear pitting measurement

An efficient visual detection method is explored in this study to address the low accuracy and efficiency of manual detection for irregular gear pitting. The results of gear pitting detection are enhanced by embedding two attention modules into Deeplabv3 + to obtain an improved segmentation model called attention Deeplabv3. The attention mechanism of the proposed model endows the latter with an enhanced ability for feature representation of small and irregular objects and effectively improves the segmentation performance of Deeplabv3. The segmentation ability of attention Deeplabv3+ is verified by comparing its performance with those of other typical segmentation networks using two public datasets, namely, Cityscapes and Voc2012. The proposed model is subsequently applied to segment gear pitting and tooth surfaces simultaneously, and the pitting area ratio is calculated. Experimental results show that attention Deeplabv3 has higher segmentation performance and measurement accuracy compared with the existing classical models under the same computing speed. Thus, the proposed model is suitable for measuring various gear pittings.

Immunocapture Magnetic Beads Enhanced the LAMP-CRISPR/Cas12a Method for the Sensitive, Specific, and Visual Detection of Campylobacter jejuni.

Campylobacter jejuni is one of the most important causes of food-borne infectious disease, and poses challenges to food safety and public health. Establishing a rapid, accurate, sensitive, and simple detection method for C. jejuni enables early diagnosis, early intervention, and prevention of pathogen transmission. In this study, an immunocapture magnetic bead (ICB)-enhanced loop-mediated isothermal amplification (LAMP) CRISPR/Cas12a method (ICB-LAMP-CRISPR/Cas12a) was developed for the rapid and visual detection of C. jejuni. Using the ICB-LAMP-CRISPR/Cas12a method, C. jejuni was first captured by ICB, and the bacterial genomic DNA was then released by heating and used in the LAMP reaction. After the LAMP reaction, LAMP products were mixed and detected by the CRISPR/Cas12a cleavage mixture. This ICB-LAMP-CRISPR/Cas12a method could detect a minimum of 8 CFU/mL of C. jejuni within 70 min. Additionally, the method was performed in a closed tube in addition to ICB capture, which eliminates the need to separate preamplification and transfer of amplified products to avoid aerosol pollution. The ICB-LAMP-CRISPR/Cas12a method was further validated by testing 31 C. jejuni-positive fecal samples from different layer farms. This method is an all-in-one, simple, rapid, ultrasensitive, ultraspecific, visual detection method for instrument-free diagnosis of C. jejuni, and has wide application potential in future work.

Establishment and application of a rapid visual detection method for Listeria monocytogenes based on polymerase spiral reaction (PSR)

ABSTRACT Listeria monocytogenes is a common foodborne pathogen that presents in various food products, posing important threat to public health. The aim of this study was to establish a rapid and sensitive method with visualization to detect L. monocytogenes based on polymerase spiral reaction (PSR). Primers targeting conserved hlyA gene sequence of L. monocytogenes were designed based on bioinformatics analyses on the current available L. monocytogenes genomes. The isothermal amplification PSR can be completed under constant temperature (65ᵒC) within 60 min with high specificity and sensitivity. Twenty-five reference strains were used to evaluate the specificity of the developed reaction. The results showed that the sensitive of the reaction for L. monocytogenes in purified genomic DNA and artificially contaminated food samples were 41 pg/μL and 103 CFU/mL, respectively. It was 100-fold more sensitive than conventional PCR. In conclusion, this novel PSR method is rapid, cost-efficient, timesaving, and applicable on artificially contaminated food samples, providing broad prospects into the detection of foodborne microbes with the promising on-site inspection.

Needle-Probe Optical Coherence Tomography for Real-Time Visualization of Veress Peritoneal Needle Placement in a Porcine Model: A New Safety Concept for Pneumoperitoneum Establishment in Laparoscopic Surgery.

The safe establishment of pneumoperitoneum is a critical step in all laparoscopic surgeries. A closed pneumoperitoneum is usually obtained by inserting a Veress needle into the peritoneal cavity. However, there is no definite measure to visually confirm the position of the Veress needle tip inside the peritoneal cavity. This study aimed to describe a method of real-time visual detection of peritoneal placement of the Veress needle using an incorporated optical coherence tomography (OCT) probe in a porcine model. A 14-gauge Veress needle was incorporated with a miniature fiber probe to puncture the piglet’s abdominal wall into the peritoneal cavity. A total of 80 peritoneal punctures were attempted in four piglets. For each puncture, continuous two-dimensional OCT images of the abdominal wall were acquired for real-time visual detection of the needle placement into the peritoneal cavity. Characteristic OCT image patterns could be observed during the puncturing process, especially a deep V-shaped concave pattern before the peritoneum puncture, which was a crucial feature. A statistical difference in the OCT signal standard deviation value also indicated the differentiability of images between the peritoneum and extra-peritoneal tissue layers. A success rate of 97.5% could be achieved with the guidance of the OCT images. OCT images translate the blind closed technique of peritoneal access into a visualized procedure, thus improving peritoneal access safety.

Detecting Phytoplankton Cell Viability Using NIR Raman Spectroscopy and PCA.

Raman spectroscopy has long been suggested as a potentially fast and sensitive method to monitor phytoplankton abundance and composition in marine environments. However, the pitfalls of visible detection methods in pigment-rich biological material and the complexity of their spectra have hindered their application as reliable in situ detection methods. In this study we combine 1064 nm confocal Raman spectroscopy with multivariate statistical analysis techniques (principle component analysis and partial leas-squares discriminant analysis) to reliably measure differences in the cell viability of a diatom species (Chaetoceros muelleri) and two haptophyte species (Diacronema lutheri and Tisochrysis lutea) of phytoplankton. The low fluorescence background due to this combined approach of NIR Raman spectroscopy and multivariate data analysis allowed small changes in the overall spectral profiles to be reliably monitored, enabling the identification of the specific spectral features that could classify cells as viable or nonviable regardless of their species. The most significant differences upon cell death were shown by characteristic shifts in the carotenoid bands at 1527 and 1158 cm–1. The contributions from other biomolecules were less pronounced but revealed changes that could be identified using this combination of techniques.

ConRec: malware classification using convolutional recurrence

Today, the extensive reliae on technology has exposed us to a constant threat of sophisticated malware attacks. Various automated malware production techniques have evolved, some of which reuse specific segments of existing malware to produce new malware, making malware detection challenging. In this paper, we propose a Convolutional Recurrence based malware classification technique that exploits the visual recurrences in the grayscale images of the malware samples belonging to the same malware families. Firstly, we convert the malware samples into grayscale images to capture the structural similarities from the malware samples using a Convolutional Neural Network architecture. Then we perform data augmentation to counter the effects of high data imbalance and reduce the class bias, such that training on that dataset would generate a more generalized framework. The augmented dataset is then passed through a VGG16 based feature extractor to extract the visual outliers amongst the malware families. Now, the extracted features are processed by passing them through two stacked BiLSTM layers. The outputs generated by the BiLSTM layers and the VGG16 layer are then merged to perform the final classification of the malware sample into its malware family. The model’s performance is further improved by using proper hyperparameter tuning. We compare the performance of our algorithm against several baseline methods and some contemporary state-of-the-art methods for visual malware detection by utilizing two benchmarked datasets. The obtained experimental results reveal the utility and efficacy of our proposed malware family classification technique.

Quantum dot based molecularly imprinted polymer test strips for fluorescence detection of ferritin

A test strip for selective detection of ferritin was prepared via an integration of molecular imprinting technology and multicolor fluorescence elements. The synthesized nanoparticle of ferritin fluorescence molecularly imprinted polymers (FerFLMIPs) showed a fast kinetic feature and the data fitted a pseudo-second-order model. The selectivity experiment displayed that the FerFLMIPs had a much higher specificity for ferritin than that for the control proteins. In addition, the color of the FerFLMIPs changed gradually from green to red with the increment of ferritin concentration. Therefore, a manageable and rapid method for visual detection of ferritin was constructed by the combination of FerFLMIPs and test strips. This work thus provides a practical method for direct and minitype detection of targeted analytes.

Fluorescent Ionic Probe for Determination of Mechanical Properties of Healed Poly(ethylene-co-methacrylic acid) Ionomer Films.

In recent years, advanced materials with properties resembling biological systems, particularly artificial muscles, have received intense scrutiny. This is because the interesting conformational shape characteristics of such materials have benefited a variety of technologies, including textiles, 3D printing, and medical devices. Although a multitude of shape memory properties have been studied and developed in recent years, self-healing of these polymers after puncture or rupture has also become a major area of study. Most techniques for detection of such processes are mechanically based and require considerable hands-on monitoring. Thus, a rapid visual detection method for self-healing is highly desirable. Herein, we describe fluorescence studies for rapid detection of self-healing properties of a partially neutralized sodium ionomer poly(ethylene-co-methacrylic acid) (PEMA). In this study, two different fluorophores, parent non-ionic 4,6-dipyrenylpyrimidine and ionic 4,6-dipyrenylpyrimidinium iodide fluorophores, were evaluated as possible sensors of self-healing. Incorporation of these probes via solution blending and compatibility into a PEMA of these fluorophores were evaluated. Thermal characterizations using differential scanning calorimetry were also performed to elucidate physical characteristics of healed sites. Ratiometric fluorescence emission variations were explored within puncture-healed ionomer films and related to Young’s modulus properties with good linearity, indicating potential utility of this approach for monitoring elastic modulus properties after healing has occurred. Further statistical analyses of mechanical processes using quadratic discriminant analysis resulted in development of several highly accurate predictive models for determining time since damage healing.

Detection and discrimination of disease and insect stress of tea plants using hyperspectral imaging combined with wavelet analysis

Compared with the traditional visual detection method, hyperspectral imaging enables efficient and non-destructive plant monitoring. Besides, it has great potential in plant phenotyping in response to disease and insect infections. However, most previous studies on hyperspectral imaging have focused on detecting a single disease, which can rarely discriminate between multiple co-occurring diseases and insects. In this study, three tea plant stresses with similar symptoms, including the tea green leafhopper (Empoasca (Matsumurasca) onukii Matsuda), anthracnose (Gloeosporium theae-sinesis Miyake), and sunburn (disease-like stress), were evaluated. A multi-step approach was proposed based on hyperspectral imaging and continuous wavelet analysis (CWA) to discriminate the plant stresses. The process entailed: (1) Feature extraction for detection and discrimination of tea plant stresses based on CWA; (2) Detecting abnormal areas on tea leaves via the k-means clustering and support vector machine algorithms; (3) Construction of a model for identification and discrimination of the three tea plant stresses via the random forest algorithm. The results showed that CWA could effectively identify spectral features for distinguishing the three stresses. The overall accuracy (OA) of the proposed approach reached 90.26%-90.69%, with anthracnose having the highest OA (94.12%-94.28%), followed by tea green leafhopper (93.99%-94.20%), while sunburn damage was the least (82.50%-83.91%). Therefore, hyperspectral imaging is effective for plant phenotyping after diseases and insect infections.

Paper-based detection of Epstein-Barr virus using asymmetric polymerase chain reaction and gold silicon particles

A new paper-based lateral flow nucleic acid (LFNA) test platform was established in this study using asymmetric polymerase chain ceaction (A-PCR) for signal amplification. This new method allowed a visual detection of Epstein-Barr virus (EBV) nucleic acids with high specificity and low cost. In addition, as part of our strategy we employed a sandwich system of capture probe (CP)/gold nanoparticles (AuNPS) and silicon dioxide (SiO2) (AuNPS@SiO2) nanospheres/target DNA/avidin complexes as the sensing platform. Biotin-labeled target DNA was obtained by A-PCR and later introduced in the LF device. The CP/target DNA/AuNPS@SiO2 complexes were captured on the test zone by the specific reaction between biotin and avidin, and the remaining CP/AuNPS@SiO2 particles were captured on the quality control zone by the hybridization between CP and a quality control probe. Au@SiO2 accumulation in the test and quality control zones of the device enabled a visual detection of the specific target sequences. The method detection limit was 50 nM of the target DNA, which was lower than that of the LFNA biosensor (LFNAB) without PCR amplification and Au@SiO2 particles. In conclusion, the novel paper-based platform described here is a low cost, efficient and fast visual detection method that offers high sensitivity and other benefits compared to alternative methods in use.

High-Throughput Online Visual Detection Method of Cracked Preserved Eggs Based on Deep Learning

Cracked preserved eggs can easily decay, emit a peculiar smell, and cause cross-infection. The identification of cracked preserved eggs during production suffers from low efficiency and high cost. This paper proposes an online detection and identification method of cracked preserved eggs to address this issue. First, the images of preserved eggs are collected online. Then, each collected image is cut into a single image of the preserved egg, and the images of different surfaces of the same preserved egg are respectively spliced by the sequential splicing scheme and the matrix splicing scheme. Finally, the data sets obtained by the two stitching methods are exploited to establish a deep learning detection model. The experimental results indicate that the MobileNetV3_egg model, an improved version of the MobileNetV3_large model, achieves the best recognition ability for cracked preserved eggs by using the matrix splicing scheme. The accuracy reaches 96.3%, and the detection time for 300 images is only 4.267 s. The proposed method can meet the needs of actual production, and the application of this method will make the identification of cracked preserved eggs more automated and intelligent.