Dynamic Vision-Based Machinery Fault Diagnosis With Cross-Modality Feature Alignment

Intelligent machinery fault diagnosis methods have been popularly and successfully developed in the past decades,and the vibration acceleration data collected by contact accelerometers have been widely investigated.In many industrial scenarios,contactless sensors are more preferred.The event camera is an emerging bio-inspired technology for vision sensing,which asynchronously records per-pixel brightness change polarity with high temporal resolution and low latency.It offers a promising tool for contactless machine vibration sensing and fault diagnosis.However,the dynamic vision-based methods suffer from variations of practical factors such as camera position,machine operating condition,etc.Furthermore,as a new sensing technology,the labeled dynamic vision data are limited,which generally cannot cover a wide range of machine fault modes.Aiming at these challenges,a novel dynamic vision-based machinery fault diagnosis method is proposed in this paper.It is motivated to explore the abundant vibration acceleration data for enhancing the dynamic vision-based model performance.A crossmodality feature alignment method is thus proposed with deep adversarial neural networks to achieve fault diagnosis knowledge transfer.An event erasing method is further proposed for improving model robustness against variations.The proposed method can effectively identify unseen fault mode with dynamic vision data.Experiments on two rotating machine monitoring datasets are carried out for validations,and the results suggest the proposed method is promising for generalized contactless machinery fault diagnosis.

Association of age at diagnosis of diabetes with subsequent risk of age-related ocular diseases and vision acuity

BACKGROUND The importance of age on the development of ocular conditions has been reported by numerous studies.Diabetes may have different associations with different stages of ocular conditions,and the duration of diabetes may affect the development of diabetic eye disease.While there is a dose-response relationship between the age at diagnosis of diabetes and the risk of cardiovascular disease and mortality,whether the age at diagnosis of diabetes is associated with incident ocular conditions remains to be explored.It is unclear which types of diabetes are more predictive of ocular conditions.AIM To examine associations between the age of diabetes diagnosis and the incidence of cataract,glaucoma,age-related macular degeneration(AMD),and vision acuity.METHODS Our analysis was using the UK Biobank.The cohort included 8709 diabetic participants and 17418 controls for ocular condition analysis,and 6689 diabetic participants and 13378 controls for vision analysis.Ocular diseases were identified using inpatient records until January 2021.Vision acuity was assessed using a chart.RESULTS During a median follow-up of 11.0 years,3874,665,and 616 new cases of cataract,glaucoma,and AMD,respectively,were identified.A stronger association between diabetes and incident ocular conditions was observed where diabetes was diagnosed at a younger age.Individuals with type 2 diabetes(T2D)diagnosed at<45 years[HR(95%CI):2.71(1.49-4.93)],45-49 years[2.57(1.17-5.65)],50-54 years[1.85(1.13-3.04)],or 50-59 years of age[1.53(1.00-2.34)]had a higher risk of AMD independent of glycated haemoglobin.T2D diagnosed<45 years[HR(95%CI):2.18(1.71-2.79)],45-49 years[1.54(1.19-2.01)],50-54 years[1.60(1.31-1.96)],or 55-59 years of age[1.21(1.02-1.43)]was associated with an increased cataract risk.T2D diagnosed<45 years of age only was associated with an increased risk of glaucoma[HR(95%CI):1.76(1.00-3.12)].HRs(95%CIs)for AMD,cataract,and glaucoma associated with type 1 diabetes(T1D)were 4.12(1.99-8.53),2.95(2.17-4.02),and 2.40(1.09-5.31),respectively.In multivariable-adjusted analysis,individuals with T2D diagnosed<45 years of age[β95%CI:0.025(0.009,0.040)]had a larger increase in LogMAR.Theβ(95%CI)for LogMAR associated with T1D was 0.044(0.014,0.073).CONCLUSION The younger age at the diagnosis of diabetes is associated with a larger relative risk of incident ocular diseases and greater vision loss.

Hawk‐eye‐inspired perception algorithm of stereo vision for obtaining orchard 3D point cloud navigation map

The binocular stereo vision is the lowest cost sensor for obtaining 3D information.Considering the weakness of long‐distance measurement and stability,the improvement of accuracy and stability of stereo vision is urgently required for application of precision agriculture.To address the challenges of stereo vision long‐distance measurement and stable perception without hardware upgrade,inspired by hawk eyes,higher resolution perception and the adaptive HDR(High Dynamic Range)were introduced in this paper.Simulating the function from physiological structure of‘deep fovea’and‘shallow fovea’of hawk eye,the higher resolution reconstruction method in this paper was aimed at ac-curacy improving.Inspired by adjustment of pupils,the adaptive HDR method was proposed for high dynamic range optimisation and stable perception.In various light conditions,compared with default stereo vision,the accuracy of proposed algorithm was improved by 28.0%evaluated by error ratio,and the stability was improved by 26.56%by disparity accuracy.For fixed distance measurement,the maximum improvement was 78.6%by standard deviation.Based on the hawk‐eye‐inspired perception algorithm,the point cloud of orchard was improved both in quality and quantity.The hawk‐eye‐inspired perception algorithm contributed great advance in binocular 3D point cloud recon-struction in orchard navigation map.

Computer-aided diagnosis of retinopathy based on vision transformer

Age-related Macular Degeneration(AMD)and Diabetic Macular Edema(DME)are two com-mon retinal diseases for elder people that may ultimately cause irreversible blindness.Timely and accurate diagnosis is essential for the treatment of these diseases.In recent years,computer-aided diagnosis(CAD)has been deeply investigated and effectively used for rapid and early diagnosis.In this paper,we proposed a method of CAD using vision transformer to analyze optical co-herence tomography(OCT)images and to automatically discriminate AMD,DME,and normal eyes.A classification accuracy of 99.69%was achieved.After the model pruning,the recognition time reached 0.010 s and the classification accuracy did not drop.Compared with the Con-volutional Neural Network(CNN)image classification models(VGG16,Resnet50,Densenet121,and EfficientNet),vision transformer after pruning exhibited better recognition ability.Results show that vision transformer is an improved alternative to diagnose retinal diseases more accurately.

基于Vision Transformer的永磁同步电机故障智能诊断

针对电机运行过程中故障信号数据量少的问题,本文提出了一种基于Vision Transformer的永磁同步电机智能故障诊断方法。该方法首先通过格拉姆矩阵(Gram)、相对位置矩阵(RPM)方法将传感器获取的一维时序信号数据转换为二维图像数据,然后将矩阵图像数据作为ViT-B/16网络的输入进行故障诊断。经过实验验证,该方法能够对永磁同步电机正常、轴承故障、退磁故障等8种状态进行识别和分类,其中使用Gram矩阵图像作为该方法输入的准确率达到99.2%,使用RPM矩阵图像作为输入准确率达到99.6%,均高于AlexNet、VGG16、ResNet等卷积网络的故障分类准确度,证明该方法可有效提高永磁同步电机故障诊断的准确度。

基于短时傅立叶变换和改进Vision Transformer的滚动轴承故障诊断方法

针对传统故障诊断技术在精确与高效地诊断减速器滚动轴承故障信号方面所面临的挑战,文中提出了一种基于短时傅里叶变换与改进Vision Transformer模型的故障诊断新方法。此方法有效融合了短时傅里叶变换在处理非线性和非平稳信号上的优势以及Vision Transformer在图像分类任务上的卓越性能。通过短时傅里叶变换将一维的振动信号转化为包含时域和频域信息的二维图像数据,进而利用改进的Vision Transformer模型对这些图像数据进行处理,以实现对滚动轴承故障状态的精准诊断。在公开数据集上的实验结果验证了该方法的稳定性与高识别精度,展示了其在滚动轴承故障诊断领域的应用潜力。

Prevalence and risk factors for eye diseases, blindness, and low vision in Lhasa, Tibet

AIM: To determine the prevalence and risk factors for eye diseases, blindness, and low vision in Tibet, and to assist the development of eye disease prevention and treatment schemes.METHODS: We carried out a survey of eye diseases among a population living at high altitude. A total of 1 115 Tibetan permanent residents aged 40 years or older from the towns and villages of Qushui County, Lhasa Prefecture, Tibet Autonomous Region, participated in this study. All participants completed a detailed questio-nnaire, and underwent presenting and pinhole visual acuity tests,and a comprehensive ophthalmic examination.RESULTS: There were 187 blind eyes (8.43%), 231 eyes with low vision (10.41% ). The leading cause of visual impairment was cataract of 55.0% (101/187) blindness and of 50.2% (116/231) low vision, followed by fundus lesions of 22.9% blindness and 23.8% low vision, while only a low prevalence of glaucoma of 9.6% blindness and 1.7% low vision was observed. The analysis of 2 219 eyes showed that the most common external eye disease was pterygium (27.2%) in Tibet.CONCLUSION: The high prevalence of blindness and low vision in the Tibetan population at high altitude is a serious public health issue. There is a need to establish and maintain an appropriate effective eye care program in Tibet.

The enlightenment of artificial intelligence large-scale model on the research of intelligent eye diagnosis in traditional Chinese medicine

Eye diagnosis is a method for inspecting systemic diseases and syndromes by observing the eyes.With the development of intelligent diagnosis in traditional Chinese medicine(TCM);artificial intelligence(AI)can improve the accuracy and efficiency of eye diagnosis.However;the research on intelligent eye diagnosis still faces many challenges;including the lack of standardized and precisely labeled data;multi-modal information analysis;and artificial in-telligence models for syndrome differentiation.The widespread application of AI models in medicine provides new insights and opportunities for the research of eye diagnosis intelli-gence.This study elaborates on the three key technologies of AI models in the intelligent ap-plication of TCM eye diagnosis;and explores the implications for the research of eye diagno-sis intelligence.First;a database concerning eye diagnosis was established based on self-su-pervised learning so as to solve the issues related to the lack of standardized and precisely la-beled data.Next;the cross-modal understanding and generation of deep neural network models to address the problem of lacking multi-modal information analysis.Last;the build-ing of data-driven models for eye diagnosis to tackle the issue of the absence of syndrome dif-ferentiation models.In summary;research on intelligent eye diagnosis has great potential to be applied the surge of AI model applications.

Eye-Tracking Based Autism Spectrum Disorder Diagnosis Using Chaotic Butterfly Optimization with Deep Learning Model

Autism spectrum disorder(ASD)can be defined as a neurodevelopmental condition or illness that can disturb kids who have heterogeneous characteristics,like changes in behavior,social disabilities,and difficulty communicating with others.Eye tracking(ET)has become a useful method to detect ASD.One vital aspect of moral erudition is the aptitude to have common visual attention.The eye-tracking approach offers valuable data regarding the visual behavior of children for accurate and early detection.Eye-tracking data can offer insightful information about the behavior and thought processes of people with ASD,but it is important to be aware of its limitations and to combine it with other types of data and assessment techniques to increase the precision of ASD detection.It operates by scanning the paths of eyes for extracting a series of eye projection points on images for examining the behavior of children with autism.The purpose of this research is to use deep learning to identify autistic disorders based on eye tracking.The Chaotic Butterfly Optimization technique is used to identify this specific disturbance.Therefore,this study develops an ET-based Autism Spectrum Disorder Diagnosis using Chaotic Butterfly Optimization with Deep Learning(ETASD-CBODL)technique.The presented ETASDCBODL technique mainly focuses on the recognition of ASD via the ET and DL models.To accomplish this,the ETASD-CBODL technique exploits the U-Net segmentation technique to recognize interested AREASS.In addition,the ETASD-CBODL technique employs Inception v3 feature extraction with CBO algorithm-based hyperparameter optimization.Finally,the long-shorttermmemory(LSTM)model is exploited for the recognition and classification of ASD.To assess the performance of the ETASD-CBODL technique,a series of simulations were performed on datasets from the figure-shared data repository.The experimental values of accuracy(99.29%),precision(98.78%),sensitivity(99.29%)and specificity(99.29%)showed a better performance in the ETASD-CBODL technique over recent approaches.

Planar Motion Tracking with an Uncalibrated Robot/Vision System

This paper discusses the coordination process for a robot gripper to approach a movingobject with feedback from an uncalibrated visual system. The dynamic of the whole system, includingtarget's random motion and the gripper's tracking motion, is bounded to a 2-D working plane. Acamera, whose relations with the robot system and the 2-D working plane are unknown to the robotcontroller, is fixed aside to observe the object and gripper positions continually. Thus the movementsof the robot gripper can be decided on the positions of the object observed in each visual samplingmoment. The coordination of the vision and robot system is to be shown independently from therelations between the robot and the vision system, which should always be calibrated a prior forthe control of traditional robot/vision coordination system. Simulations are provided to show theproperty of the proposed method.

Automatic diagnosis of strawberry water stress status based on machine vision

Water stress status of plants is very important for irrigation scheduling.However,plant water stress status monitoring has become the bottleneck of irrigation scheduling.In this study,an automatic water stress status monitoring method for strawberry plant was proposed and realized using combined RGB and infrared image information.RGB image and infrared images were obtained using RGB digital camera and infrared thermal camera,which were placed in a fixed shell in parallel.In the first experimental stage,three kinds of water stress treatments were carried out on three groups of strawberry plants,and each group includes three repetitions.Single point plant temperature,dry surface temperature,wet surface temperature were measured.In the second experimental stage,the infrared and visible light images of the canopy leaves were obtained.Meanwhile,plant temperature,dry surface temperature,wet surface temperature,and stomatal conductance were measured not only for single point but also for plant area temperature measurement.Fusion information of infrared image and visible light image was analyzed using image processing technology,to calculate the average temperature of plant areas.Based on single point temperature,area temperature,dry surface temperature and wet surface temperature of the plant,single point crop water stress index(CWSI)and area CWSI were calculated.Through analysis of variance(ANOVA),the experimental results showed that CWSI measured for plants under different treatments,were significantly different.Through correlation analysis,the experimental results showed that,determination coefficient between area CWSI and the corresponding stomatal conductance of three strawberry groups were 0.8834,0.8730 and 0.8851,respectively,which were larger than that of single-point CWSI and stomatal conductance.The results showed that area CWSI is more suitable to be used as the criteria for automatic diagnosis of plants.

Recent advances in bioinspired vision systems with curved imaging structures

Limited by the planar imaging structure,the commercial camera needs to introduce additional optical elements to compensate for the curved focal plane to match the planar image sensor.This results in a complex and bulky structure.In contrast,biological eyes possess a simple and compact structure due to their curved imaging structure that can directly match with the curved focal plane.Inspired by the structures and functions of biological eyes,curved vision systems not only improve the image quality,but also offer a variety of advanced functions.Here,we review the recent advances in bioinspired vision systems with curved imaging structures.Specifically,we focus on their applications in implementing different functions of biological eyes,as well as the emerging curved neuromorphic imaging systems that incorporate bioinspired optical and neuromorphic processing technologies.In addition,the challenges and opportunities of bioinspired curved imaging systems are also discussed.

Original askiatic imaging used in Chinese medicine eye-feature diagnosis of visceral diseases

Eye-feature diagnosis is a time-homored met hod for studying many diseases in tradit ional Chinese medicine.There is a dlose relationship between eye feature and viscera,and eye feature is a reflect ion of viscer al health status.Commercially used ophthalmology diagnosis instr uments have disadvantages and cannot satisfy the requirements of eye feature diagnosis.In this paper,we proposed a novel askiatic imaging method that removes the interference of an ilumination source's reflection shadow and is free from image splicing.We developed a novel imaging system to implement this method,and some eye feature characteristics to analyze visceral diseases were obtained.

Review of eye diagnosis in traditional Chinese medicine:Modernization and future prospects

This article reviews the basic theories, methods, and clinical applications of eye diagnosis in traditional Chinese medicine(TCM). It introduces cutting-edge methods and applications and explains that the modernization of TCM eye diagnosis includes “equipment-assisted diagnosis” and “artificial intelligencebased diagnosis”. The article also notes that while there are many recent studies of the static attributes of eyes in modern TCM eye diagnosis, modern application research on the dynamic attributes of eyes in TCM diagnosis theory is relatively rare. We propose, therefore, that introducing advanced eye-movement detection technology into TCM clinical diagnosis could help to further modernize TCM eye diagnosis.

Optical study on the vision correction and supernormal vision based on the wave-front aberrations of human eye

In this paper we present the recent research results in the field of vision correction and supernormal vision according to the actual measurements of the wave-front aberrations and the corneal surface topography,the clinical detection of the visual function and the laser corneal refractive surgery,and the optimization of the optical system. These include the features of the aberrations of human eye with different pupil sizes,different fields of view and temporal accommodation,the influence of the polychromatic illumination of the visible wavelength on the supernormal vision,and the effect of the existing laser corneal refractive surgery on the wave-front ab-errations of the eye. It is shown that the wave-front aberration of human eye is of temporal variation and of synthesis with multi impact factors. To achieve super-normal vision,an optimum engineering data for the customized laser corneal sur-gery should be firstly acquired,which may involve the dynamic free-form optical surface. Although the myopia can be corrected by the laser in situ keratomileusis(LASIK) in a certain degree,it brings about negative effects under scotopic condi-tions.

Delayed diagnosis of prosopagnosia following a hemorrhagic stroke in an elderly man:A case report

BACKGROUND Acquired prosopagnosia is a rare condition characterized by the loss of familiarity with previously known faces and the inability to recognize new ones.It usually occurs after the onset of brain lesions such as in a stroke.The initial identification of prosopagnosia generally relies on a patient’s self-report,which can be challenging if it lacks an associated chief complaint.There were few cases of prosopagnosia presenting purely as eye symptoms in the previous literature confirmed by functional magnetic resonance imaging(MRI).CASE SUMMARY We present a case of delayed diagnosis of prosopagnosia after a right hemisphere stroke in an elderly man whose chief complaint was persistent and progressive"blurred vision"without facial recognition impairment.Ophthalmic tests revealed a homonymous left upper quadrantanopia,with normal visual acuity.He was found by accident to barely recognize familiar faces.The patient showed severe deficit in face recognition and perception tests,and mild memory loss in neuropsychological assessments.Further functional MRI revealed the visual recognition deficits were face-specific.After behavioral intervention,the patient started to rely on other cues to compensate for poor facial recognition.His prosopagnosia showed no obvious improvement eight months after the stroke,which had negative impact on his social network.CONCLUSION Our case demonstrates that the presentation of prosopagnosia can be atypical,and visual difficulties might be a clinical manifestation solely of prosopagnosia,which emphasizes the importance of routinely considering face recognition impairment among elderly patients with brain lesions.

基于EMDPWVD时频图像和改进ViT网络的滚动轴承智能故障诊断

滚动轴承是机械设备的关键零部件之一,其故障诊断对设备安全稳定运行至关重要。针对滚动轴承振动信号的非平稳特点,提出经验模态分解联合伪魏格纳分布(empirical mode decomposition&pseudo-Wigner-Ville distribution, EMDPWVD)时频图像联合改进Vision Transformer(ViT)网络模型的智能故障诊断新方法。首先针对实际信号研究短时傅里叶变换(short-time Fourier transform, STFT)、连续小波变换(continuous wavelet transform, CWT)和EMDPWVD三种时频分析方法,考虑STFT和CWT无法同时获得高的时间分辨率和频率分辨率,优选EMDPWVD作为智能故障诊断网络的时频图像构造方法。其次,以经典ViT作为故障诊断基础模型,将时频图像按照预定尺寸分块并线性映射为输入序列,通过自注意力机制整合图像全局信息,借助堆叠Transformer编码器完成网络传输,进而实现故障诊断。为进一步提高故障诊断准确率,将池化层作为ViT的预处理网络,获得改进的Pooling ViT(PiT)模型,实现时频图像的空间特征延展,提升模型对输入图像敏感度。结果表明,所提方法对滚动轴承不同故障类型均有高的诊断准确率,PiT较ViT的准确率提高4.40%,证明对ViT加入池化层能够实现滚动轴承故障诊断效果提升。

Machine learning models for stroke detection by observing the eye-movement features under five-color visual stimuli in traditional Chinese medicine

Objective:To develop a novel diagnostic modality to identify and diagnose stroke in daily life scenarios for improving the therapeutic effects and prognoses of patients.Methods:In this study,16 stroke patients and 24 age-matched healthy participants as controls were recruited for comparative analysis.Leveraging a portable eye-tracking device and integrating traditional Chinese medicine theory with modern color psychology principles,we recorded the eye movement signals and calculated eye movement features.Meanwhile,the stroke recognition models based on eye movement features were further trained by using random forest(RF),k-nearest neighbors(KNN),decision tree(DT),gradient boosting classifier(GBC),XGBoost,and CatBoost.Results:The stroke group and the healthy group showed significant differences in some eye movement features(P<.05).The models trained based on eye movement characteristics had good performances in recognizing stroke individuals,with accuracies ranging from 77.40%to 88.45%.Under the red stimulus,the eye movement model trained by RF became the best machine learning model with a recall of 84.65%,a precision of 86.48%,a F1 score of 85.47%.Among the six algorithms,RF and CatBoost performed better in classification.Conclusion:This study pioneers the application of traditional Chinese medicine's five-color stimuli to visual observation tasks.On the basis of the combined design,the eye-movement models can accurately identify stroke,and the developed high-performance models may be used in daily life scenarios.

基于CNN-VIT模型的非接触式机械密封多源故障状态识别方法

针对非接触式机械密封多源故障信号频率高、识别难和易受干扰的难题,搭建密封试验台,模拟4种典型故障工况和正常运行工况,布置声发射(AE)测试系统测取10000个特征样本,将卷积神经网络(CNN)结构与Vision Transformer模型相结合,并在CNN输出数据中引入Patch概念构建了一种改进模型,随后应用于密封故障状态识别,提出了一种新型的非接触式机械密封状态智能识别方法,并分析了改进模型各层网络决策对密封AE数据的影响。结果表明,AE技术能够高效监测非接触式机械密封多源故障状态信号,该方法对密封状态识别的准确率最高可达99.42%;改进模型更加关注稳定运行状态信号中频率为0~50 kHz和100~150 kHz的噪声频带数据,动环表面剥落状态信号中频率为100~150 kHz的噪声频带数据和270±40 kHz的摩擦AE频带数据。研究结果可为机械密封的智能化、高可靠、长寿命发展提供理论基础。

A novel minority sample fault diagnosis method based on multisource data enhancement

Effective fault diagnosis has a crucial impact on the safety and cost of complex manufacturing systems.However,the complex structure of the collected multisource data and scarcity of fault samples make it difficult to accurately identify multiple fault conditions.To address this challenge,this paper proposes a novel deep-learning model for multisource data augmentation and small sample fault diagnosis.The raw multisource data are first converted into two-dimensional images using the Gramian Angular Field,and a generator is built to transform random noise into images through transposed convolution operations.Then,two discriminators are constructed to evaluate the authenticity of input images and the fault diagnosis ability.The Vision Transformer network is built to diagnose faults and obtain the classification error for the discriminator.Furthermore,a global optimization strategy is designed to upgrade parameters in the model.The discriminators and generator compete with each other until Nash equilibrium is achieved.A real-world multistep forging machine is adopted to compare and validate the performance of different methods.The experimental results indicate that the proposed method has multisource data augmentation and minority sample fault diagnosis capabilities.Compared with other state-of-the-art models,the proposed approach has better fault diagnosis accuracy in various scenarios.