Visual Detection Method Research Articles

Research on Monocular-Vision-Based Finger-Joint-Angle-Measurement System

The quantitative measurement of finger-joint range of motion plays an important role in assessing the level of hand disability and intervening in the treatment of patients. An industrial monocular-vision-based knuckle-joint-activity-measurement system is proposed with short measurement time and the simultaneous measurement of multiple joints. In terms of hardware, the system can adjust the light-irradiation angle and the light-irradiation intensity of the marker by actively adjusting the height of the light source to enhance the difference between the marker and the background and reduce the difficulty of segmenting the target marker and the background. In terms of algorithms, a combination of multiple-vision algorithms is used to compare the image-threshold segmentation and Hough outer- and inner linear detection as the knuckle-activity-range detection method of the system. To verify the accuracy of the visual-detection method, nine healthy volunteers were recruited for experimental validation, and the experimental results showed that the average angular deviation in the flexion/extension of the knuckle was 0.43° at the minimum and 0.59° at the maximum, and the average angular deviation in the adduction/abduction of the knuckle was 0.30° at the minimum and 0.81° at the maximum, which were all less than 1°. In the multi-angle velocimetry experiment, the time taken by the system was much less than that taken by the conventional method.

Rapid colorimetric and ratiometric fluorescence method for on-site detection and visualization of phosgene by amino-functionalized carbon dot-based portable droplet system

Phosgene is a very dangerous but important industrial intermediate and is a significant threat to public health. There is, therefore, an urgent need to develop a rapid, visual, and on-site detection method for phosgene. Sensitive and selective phosgene responsive carbon dots (CDs) were synthesized by employing 2-nitro-4-aminodiphenylamine as precursors. The amine group of the CDs can interact selectively with phosgene to effect acylation reactions, which cause color changes from yellow to amaranth and remarkable changes in fluorescence from bright yellow to orange-red due to modification of the CDs’ intramolecular charge transfer (ICT) process. The CDs exhibited fast response (<2 s) for detection of phosgene with an ultralow detection limit (0.63 nM). Furthermore, a portable CD/polystyrene droplet system was developed for rapid and on-site visualization of phosgene. Specifically, the droplet system not only can realize rapid on-spot visual inspection of gaseous phosgene in the liquid phase with the naked eye, but also could make long-term visualization in the solid phase via a polystyrene membrane after solvent volatilization. This study has expanded the scope for application of CDs in sensing field, and provides a new and effective strategy for fabrication excellent platform for real-time detection of phosgene leakage in industrial and public safety applications.

Visual detection and differentiation of porcine epidemic diarrhea virus wild-type strains and attenuated vaccine strains using CRISPR/Cas13a-based lateral flow strip.

Porcine epidemic diarrhea virus (PEDV) is an enteric coronavirus that causes acute watery diarrhea and vomiting in unweaned piglets. Infections result in high mortality and serious economic losses to the swine industry. PEDV attenuated vaccine does not completely protect against all mutant wild-type strains, and PEDV infection can periodically occur. A sensitive, accurate, and simple detection method for PEDV is needed to reduce the occurrence of the disease. In this study, the CRISPR/Cas13a system was combined with recombinase aided amplification to develop a rapid diagnostic method to distinguish PEDV wild-type strains from attenuated vaccine strains. The method is based on isothermal detection at 37°C. The results are used for visual readout. The assay had high sensitivity and specificity, with a detection limit of 101 copies/μL for the gene of interest, and no cross-reactivity with other pathogens. The Cas13a detection worked well with clinical samples. This visual, sensitive, and specific nucleic acid detection method based on CRISPR/Cas13a should be a powerful tool for detecting PEDV.

Easy detection of Prorocentrum donghaiense by polymerase chain reaction-nucleic acid chromatography strip.

In recent years, Prorocentrum donghaiense, as a dominant species, has ranked first in terms of cumulative number and area of algal blooms in the East China Sea. In this study, the D1-D2 region of the large ribosomal subunit of P. donghaiense was used as the target gene, and specific primers DH-FP/DH-RP were designed according to the species-specific region of the target gene. An easy, sensitive and visual detection method refered to as polymerase chain reaction-nucleic acid chromatography strip (PCR-NACS) was established for P. donghaiense. The optimized parameters of the PCR amplification system are as follows: primer concentration, 0.15μM; annealing temperature, 62°C; and Mg2+ concentration, 1.5mM. The specificity test showed that PCR-NACS was exlusively specific for the detection of the target algae. The sensitivity test show that the lowest detection limit (LDL) of PCR-NACS was 2.7 × 10-2ng·μL-1 for genomic DNA and 3.58 × 102 copies·μL-1 for plasmid DNA, respectively. The tests using both genomic DNA and plasmid DNA as templates showed that the sensitivity of PCR-NACS was 10 times higher than that of ordinary PCR. The stability test showed that the interfering algal species did not affect the detection of the target algae by PCR-NACS. In addition, the test with simulated natural samples containing target algae showed that the LDL of PCR-NACS could reach 1.27 × 101 cells·mL-1. In summary, the PCR-NACS established in this study may provide a new method for easy identification of P. donghaiense in natural water samples.

Improved YOLOv3 detection method for PCB plug-in solder joint defects based on ordered probability density weighting and attention mechanism

Printed Circuit Board (PCB) is the heart component of electronic products, and its defect detection is the basic requirement of PCB quality control in the production process. Traditional visual detection methods need artificial design features, so their detection accuracy is poor, and the rate of missed and false detection is high. To solve the above problems, this paper proposes an improved YOLOv3 (You Only Look Once) detection method for PCB plug-in solder spot defects based on the combination of the ordered probability density weighting and the attention mechanism. First, in order to obtain a higher priority priori box, the ordered probability density weighting (OWA) method is used to optimize the multiple sets of a priori boxes generated by K-means. Then, to get more effective feature information, the Squeeze-and-Excitation mechanism (SE) is added to the backbone network. In the feature detection network, the Convolutional Block Attention Module (CBAM) attention mechanism is joined, at the same time in the inspection network output layer three layer feature are fusions. Finally, in order to accelerate the convergence speed of model and improve the accuracy of the model, the network loss function was improved by using the generalized joint generalized intersection over union (GIoU), and the COCO data model was applied to PCB solder spot defect training by transfer learning method. After testing, the average detection accuracy of improved network is improved from 84.35% to 96.69%, and the improved network has better convergence than the original network. The study shows that the improved method based on YOLOv3 is more suitable for industrial application of PCB plug-in solder spot defect detection.

Visual Detection and Image Processing of Parking Space Based on Deep Learning.

The automatic parking system based on vision is greatly affected by uneven lighting, which is difficult to make an accurate judgment on parking spaces in the case of complex image information. To solve this problem, this paper proposes a parking space visual detection and image processing method based on deep learning. Firstly, a 360-degree panoramic system was designed to photograph the vehicle environment. The image has been processed to obtain a panoramic aerial view, which was input as the original image of the parking space detection system. Secondly, the Faster R-CNN (Region-Convolutional Neural Network) parking detection model was established based on deep learning. It was aimed to detect and extract the parking space from the input image. Thirdly, the problems of uneven illumination and complex background were solved effectively by removing the background light from the image. Finally, a parking space extraction method based on the connected region has been designed, which further simplified the parking space extraction and image processing. The experiment results show that the mAP (mean Average Precision) value of the Faster R-CNN model using 101-Floor ResNet as the feature extraction network is 89.30%, which is 2.28% higher than that of the Faster R-CNN model using 50-Floor ResNet as the feature extraction network. The model built in this paper can detect most parking spaces well. The position of the output target box is accurate. In some test scenarios, the confidence of parking space recognition can even reach 100%. In summary, the proposed method can realize the effective identification and accurate positioning of parking spaces.

Real-Time Monocular Joint Perception Network for Autonomous Driving

Comprehensive and accurate perception of the real 3D world is the basis of autonomous driving. However, many perceptual methods focus on a single task or object type, and the accuracy of existing multi-task or multi-object methods is difficult to balance against their real-time performance. This paper presents a unified framework for concurrent dynamic multi-object joint perception, which introduces a real-time monocular joint perception network termed MJPNet. In MJPNet relative weightings are automatically learned by a series of developed network branches. By training an end-to-end deep convolutional neural network on a shared feature encoder and many proposed decoding sub-branches, the information of the 2D category and 3D position/pose/size of an object are reconstructed both simultaneously and accurately. Moreover, the effective information among subtasks is transferred by multi-stream learning, guaranteeing the accuracy of each task. Compared to various state-of-the-arts, comprehensive evaluations on the benchmark of challenging image sequences demonstrate the superior performance of our 2D detection and 3D reconstruction of depth, lateral distance, orientation, and heading angle. Moreover, on the KITTI test set, the real-time runtime (up to 15 fps) of MJPNet significantly outran the public state-of-the-art visual detection methods. Accompanying video: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://youtu.be/Z-goToOlI94</uri> .

Nano-gold-enhanced LAMP method for qualitative visual detection of Salmonella in milk

Since Salmonella can cause foodborne disease and public health safety issues and requires a robust, rapid, on-site detection method, a novel visual qualitative method with nano-gold-enhanced loop-mediated isothermal amplification (LAMP) reaction was established for detecting Salmonella in an integrated tube. During the experiment, nano-gold were used to enhance LAMP amplification, improving amplification efficiency and shortening the reaction time to within 30min. Visual qualitative detection is achieved via negative staining, involving the addition of CuSO4 to the final products of the LAMP reaction. Ring-like white accumulation occurs in the absence of Salmonella targets but not when they are present. After completing the LAMP reaction, the integration tube was shaken gently for 1min to observe the liquid phase system changes, realizing the closed tube detection of Salmonella. The process resolved the challenge presented by cross-contamination, false positives, and nonspecific amplification during the LAMP reaction. This method was used to detect Salmonella in milk, further highlighting its prospects in the field of rapid food safety detection.

Individual North Atlantic right whales identified from space

AbstractThe population of the critically endangered North Atlantic right whale, Eubalaena glacialis, numbers approximately 336 individuals, and continues to decline. Current development and implementation of protection measures and monitoring of right whale presence relies on visual surveys from vessel and aerial platforms, and passive acoustic monitoring, which contribute to location data. Here we demonstrate that satellite imagery can be used to detect and confirm the North Atlantic right whales at a species level using newly available imaging methods, providing another tool to inform conservation efforts. Using optical satellite imagery with 15 cm resolution, 25 right whales were observed in Cape Cod Bay on April 24, 2021. Species confirmation was possible due to clearly visible callosity patterns indicative of their species within this range. Although the variations in callosity patterns commonly used to identify individuals were too small to be resolved at this image resolution, one whale with large distinctive markings visible on his body was identified at an individual level from satellite imagery alone. Although visual and acoustic survey methods can be combined for monitoring this species in critical habitats, satellite‐based monitoring can be especially helpful to understand whale presence outside the areas monitored with existing visual and acoustic detection methods.

Smartphone recognition-based immune microparticles for rapid on-site visual data-sharing detection of Newcastle disease virus

Since the last century, animal viruses have posed great threats to the health of humans and the farming industry. Therefore, virus control is of great urgency, and regular, timely, and accurate detection is essential to it. Here, we designed a rapid on-site visual data-sharing detection method for the Newcastle disease virus with smartphone recognition-based immune microparticles. The detection method we developed includes three major modules: preparation of virus detection vectors, sample detection, and smartphone image analysis with data upload. First, the hydrogel microparticles containing active carboxyl were manufactured, which coated nucleocapsid protein of NDV. Then, HRP enzyme-labeled anti-NP nanobody was used to compete with the NDV antibody in the serum for color reaction. Then the rough detection results were visible to the human eyes according to the different shades of color of the hydrogel microparticles. Next, the smartphone application was used to analyze the image to determine the accurate detection results according to the gray value of the hydrogel microparticles. Meanwhile, the result was automatically uploaded to the homemade cloud system. The total detection time was less than 50 min, even without trained personnel, which is shorter than conventional detection methods. According to experimental results, this detection method has high sensitivity and accuracy. And especially, it uploads the detection information via a cloud platform to realize data sharing, which plays an early warning function. We anticipate that this rapid on-site visual data-sharing detection method can promote the development of virus self-checking at home.

Super-assembly of integrated gold magnetic assay with loop-mediated isothermal amplification for point-of-care testing.

With the increasing global threat of various diseases and infections, it is essential to develop a fast, low-cost, and easy-to-use point-of-care testing (POCT) system for inspections at all levels of medical institutions and self-examination at home. In this work, gold magnetic nanoparticles (GMNPs) are used as the key material, and a rapid visual detection method is designed through integrating loop-mediated isothermal amplification (LAMP) and lateral flow assay (LFA) biosensor for detecting a variety of analytes which includes whole blood, buccal swabs, and DNA. It is worth to note that the proposed method does not need DNA extraction. Furthermore, uracil DNA glycosylase (UDG) is employed to eliminate carrier contamination for preventing false positive results. The whole detection process can be finished within 25 min. The accuracy of detection is measured by assessing the polymorphisms of the methylenetetrahydrofolate reductase (MTHFR) C677T. The detection limit of the newly developed extraction-free detection system for MTHFR C677T is 0.16 ng/μL. A preliminary clinical study of the proposed method is carried out by analyzing 600 clinical samples (including 200 whole blood samples, 100 buccal swabs, and 300 genomic DNA samples). The results indicate that the proposed method is 100% consistent with the sequencing results which provides a new choice for POCT and shows a broad application prospect in all levels of medical clinics and at home.Electronic Supplementary MaterialSupplementary material (details for MTHFR C677T primer sequences, the cell count results of samples at different dilution ratios, genotyping results and frequency samples, a Hardy—Weinberg equilibrium test, the sensitivity of the system, detection results of multiple samples, and optimization of the system) is available in the online version of this article at 10.1007/s12274-022-4692-9.

Multitarget Tracking Algorithm in Intelligent Analysis of Football Movement Training Stance

In recent years, with the continuous development of computer technology, deep learning has been widely applied to computer vision tasks and has achieved great success in areas such as visual detection and tracking. On this basis, making deep learning techniques truly accessible to people becomes the next objective. Target detection and tracking in football gesture training is a quite challenging task with great practical and commercial value. In traditional football training methods, target trajectories are often extracted by means of a recording chip carried by the player. However, the cost of this method is high and it is difficult to replicate in amateur stadiums. Some studies have also used only cameras to process targets in football videos. However, due to the similarity in appearance and frequent occlusion of targets in football videos, these methods often only segment targets such as players and balls in the image but do not allow them to be tracked. Target tracking techniques are of great importance in football training and are the basis for tasks such as player training analysis and match strategy development. In recent years, many excellent algorithms have emerged in the field of target tracking, mainly in the categories of correlation filtering and deep learning, but none of them are able to achieve high accuracy in player tracking for football training videos. After all, the problem of locating clips of interest to athletes from a full-length video is a pressing one. Traditional machine learning-based approaches to sports event detection have poor accuracy and are limited in the types of events they can detect. These traditional methods often rely on auxiliary information such as audio commentary and relevant text, which are less stable than video. In recent years, deep learning-based methods have made great progress in the detection of single-player video events and actions, but less so in the detection of sports video events. As a result, there are few sports video datasets that can be used for deep learning training. Based on research in computer vision and deep learning, this paper designs a multitarget tracking system for football training. To be specific, this algorithm uses multiple cameras for image acquisition in the stadium in order to accurately track multiple targets in the stadium over time. Furthermore, the framework for a single camera multitarget tracking approach has been designed based on deep learning-based visual detection methods and correlation filter-based tracking methods. This framework focuses on using data correlation algorithms to fuse the results of detectors and trackers so that multiple targets can be tracked accurately in a single camera. To sum up, this research allows for robust and real-time long-term accurate tracking of targets in football training videos through multitarget tracking algorithms and the intercorrection of multiple camera systems.

Power Transmission Line Broken Strand Repair Robot and Visual Control Method

This paper proposes a design of machine vision based broken strand repair robot on power transmission line. Through the analysis of the task and process of repairing the broken strands, the design scheme of the broken strand repair robot mechanical structure based on the clamp block repair method is proposed. Aiming at the disadvantages of traditional visual detection methods under the influence of complex object background and strong light, a visual inspection method based on deep learning is proposed to identify the conductor, vibration damper and broken strand, and automatically locate the bifurcation point of the broken strand. Using visual measurement and control technology, control the operation manipulator to straighten, reset, reinforce and repair the forked aluminum wire. Developed a robot experimental prototype, the experimental results verify the feasibility of the robot structure design and the effectiveness of the visual control algorithm.

Mechanoresponsive luminogen (MRL)-based real-time and visible detection method for biaxial fatigue crack propagation

Fatigue cracks under complex stress state is crucial for safety assessment. Owing to the characteristics of visualization, rapid response and nondestructive measurement, mechanoresponsive luminogen (MRL) based methods have a great potential for stress sensing and damage detection. In this study, fatigue crack propagation in aluminum alloys under biaxial loadings were detected using a pure organic mechanochromic luminescent material. Both fatigue crack propagation behavior and luminescence emission were investigated. Real-time fatigue crack length and crack propagation path can be successfully detected using the fluorescence distribution. The MRL-based methods may open up new opportunities for nondestructive evaluation and structural damage evaluation.

A Visual Assay of a Loop-Mediated Isothermal Amplification Based Vertical Immunoassay for SARS-CoV-2 RNA Detection.

SARS-CoV-2 is a novel coronavirus that has caused a global pandemic. To date, 504,907,616 people have been infected and developed coronavirus disease 2019 (COVID-19). A rapid and simple diagnostic method is needed to control this pandemic. In this study, a visual nucleic acid detection method combining reverse transcription loop-mediated isothermal amplification and a vertical flow visualization strip (RT-LAMP-VF) was successfully established and could detect 20 copies/μl of SARS-CoV-2 RNA transcript within 50 min at 61°C. This assay had no cross-reactivity with a variety of coronaviruses, including human coronavirus OC43, 229E, HKU1, NL63, severe acute respiratory syndrome-related coronavirus (SARSr-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and bat coronavirus HKU4, exhibiting very high levels of diagnostic sensitivity and specificity. Most strikingly, this method can be used for detecting multiple SARS-CoV-2 variants, including the Wuhan-Hu-1 strain, Delta, and Omicron variants. Compared with the RT-qPCR method recommended by the World Health Organization (WHO), RT-LAMP-VF does not require special equipment and is easy to perform. As a result, it is more suitable for rapid screening of suspected SARS-CoV-2 samples in the field and local laboratories.

Development of a loop-mediated isothermal amplification (LAMP) assay for rapid visual detection of porcine circovirus type 2 (PCV2) and its application.

A cost effective, simple and rapid method is critical for detection of porcine circovirus type 2 (PCV2) infection in pigs. The present study reports the development and evaluation a loop-mediated isothermal amplification (LAMP) assay for rapid visual detection of PCV2 in pigs. The time and temperature conditions for amplification of PCV2 genes were optimized to be 30 min at 67 °C. The developed assay was 10 fold more sensitive than conventional PCR with analytical sensitivity of 5 pg and 50 pg, respectively. The developed LAMP assay had a sensitivity of 100%, specificity of 85.45% and overall accuracy of 89.70%. This is perhaps the most rapid of all LAMP reports for PCV2 detection available globally. The assay did not cross-react with porcine parvovirus or classical swine fever virus. DNA sequencing was done to ensure accuracy of LAMP assay results. The assay was assembled into a kit of 20 reactions and validated in different laboratories in India. The developed LAMP assay was proved to be a specific, sensitive and rapid method for visual detection of PCV2 which does not require costly equipments.

Visual Detection of Vibrio parahaemolyticus using Combined CRISPR/Cas12a and Recombinase Polymerase Amplification

ObjectiveTo establish an ultra-sensitive, ultra-fast, visible detection method for Vibrio parahaemolyticus (VP). MethodsWe established a new method for detecting the tdh and trh genes of VP using clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 12a (CRISPR/Cas12a) combined with recombinase polymerase amplification and visual detection (CRISPR/Cas12a-VD). ResultsCRISPR/Cas12a-VD accurately detected target DNA at concentrations as low as 10-18 M (single molecule detection) within 30 min without cross-reactivity against other bacteria. When detecting pure cultures of VP, the consistency of results reached 100% compared with real-time PCR. The method accurately analysed pure cultures and spiked shrimp samples at concentrations as low as 102 CFU/g. ConclusionThe novel CRISPR/Cas12a-VD method for detecting VP performed better than traditional detection methods, such as real-time PCR, and has great potential for preventing the spread of pathogens.

Visual LAMP Method for Detection of White Spot Syndrome Virus in Shrimp and Crayfish

White spot syndrome virus (WSSV) has caused large economic losses to the aquaculture industry, and thus far there are no efficient therapeutic treatments available against this lethal virus. Therefore, rapid detection of white spot syndrome virus (WSSV) in shrimp is crucial. This study was conducted to establish a rapid visual detection method for WSSV in shrimp and on the basis of loop-mediated isothermal amplification (LAMP). The color change of the amplification products was observed using a fluorescence amplification curve (with the SYTO-9 fluorescent dye) and LAMP visualization (with calcein), and the method`s specificity, sensitivity, and reproducibility were analyzed. It showed that this method can specifically detect WSSV, with a detection sensitivity of 1 fg/μL and has good reproducibility and reliability. This method was used to analyze cultured shrimp and crayfish samples; among 329 suspected clinical samples, the positive diagnostic rate of WSSV was 27.96%. The infection activity of WSSV was the highest at temperatures of 20–26°C. The LAMP-based method for rapid visual detection of WSSV does not need complex experimental instruments, and an equipment with a stable heat source is sufficient for the reaction. It is easy to use and can be used at shrimp breeding sites.

Dual-response ratiometric fluorescence based ligand-functionalized CsPbBr3 perovskite quantum dots for sensitive detection of trace water in edible oils

The sensitive and accurate detection of trace water in edible oils is of significance for the protection of their quality and safety. Herein, a kind of novel dual-emission CsPbBr3 perovskite quantum dots modified with a functional ligand dimethyl aminoterephthalate (CsPbBr3 @DMT-NH2 QDs) was synthesized at low temperature. The QDs display highly-sensitive fluorescence turn-on/off dual response and distinct fluorescence color change from green to blue in the presence of water. Thus, both the ratiometric fluorescence sensing and the visual ratiometric chromaticity detection method were established for water assay with the ultralow limit of detection of 0.006% (v/v) and 0.01% (v/v), respectively, and they can realize the accurate detection (recovery of 93.0%–108.0%, relative error ≤ 5.33%) of trace water in edible oils. Furthermore, it is found that the high polarity levels of protic solvents disintegrate the QDs, and the relatively-low polarity levels are conducive to the aggregation of the QDs, which ultimately lead to green fluorescence quenching. Meanwhile, the deprotonation and polarity of water promote the excited-state intramolecular proton transfer and intramolecular charge transfer of DMT-NH2, so enhancing its blue fluorescence. To our knowledge, this is the first report on dual-response ratiometric fluorescence perovskite QDs for ultrasensitive water detection in edible oils.

RBF-Based 3D Visual Detection Method for Chinese Martial Art Wrong Movements

The accuracy of action detection is limited by the extracted action, and there are problems of high processing complexity and low efficiency. Therefore, a three-dimensional visual detection method of martial art wrong action based on RBF is proposed. After noise reduction and weighting processing of martial art action video images, a martial art action 3D visual transformation model is established. According to the 3D visual model, C3D features are used to represent martial art actions. The video is segmented using sparse coding to determine the detection range. RBF neural network model is established, and the combination of the above 3D visual model and network parameters is obtained by sample training to detect martial art wrong actions. The test method of the experimental results shows the detection of the research under the condition of different degrees of precision, an average of at least 5%, and the method of detection of high efficiency and stability.