Image quality evaluation for FIB-SEM images.

Focused ion beam scanning electron microscopy (FIB-SEM) tomography is a serial sectioning technique where an FIB mills off slices from the material sample that is being analysed. After every slicing, an SEM image is taken showing the newly exposed layer of the sample. By combining all slices in a stack, a 3D image of the material is generated. However, specific artefacts caused by the imaging technique distort the images, hampering the morphological analysis of the structure. Typical quality problems in microscopy imaging are noise and lack of contrast or focus. Moreover, specific artefacts are caused by the FIB milling, namely, curtaining and charging artefacts. We propose quality indices for the evaluation of the quality of FIB-SEM data sets. The indices are validated on real and experimental data of different structures and materials.

Impact of blur on clinical and occupational colour vision test results.

PURPOSE: To evaluate whether colour vision normal (CVN) adults pass two Fletcher-Evans (CAM) lantern tests and to investigate the impact of imposed blur on Ishihara, CAM lantern and computerised colour discrimination test (colour assessment and diagnosis test [CAD] and Cambridge colour test [CCT]) results. METHODS: In a pilot experiment, 20 (16 CVN and 4 colour vision deficient [CVD]) participants with normal VA were tested with the CAM lantern. In the main experiment, the impact of imposed dioptric blur (up to +8.00 D) on visual acuity and the Ishihara test, CAM lantern, CAD and CCT was assessed for 15 CVN participants. RESULTS: CVN participants can fail the CAM lantern, with specificity of 81.25% (aviation mode) and 75% (clinical mode), despite following the test requirements of participants having at least 0.18 logMAR (6/9) in the better eye. With blur, test accuracy was affected. As expected, significant detrimental effects of blur on test results were found for logMAR VA and CAM lantern (aviation) with +1.00 D or higher. Ishihara, CAD and CCT results were not detrimentally affected until +8.00 D. Yellow-blue discrimination was more affected by blur for the CAD than the CCT, which was not explained by the different colour spaces used or vectors tested. CONCLUSION: False-positive findings on lantern colour vision tests with small apertures are likely to be increased in patients with blur due to uncorrected refractive error or ocular and visual pathway disease. Other colour vision tests with larger stimuli are more robust to blur.

Pregnancy and the eye: What do we need to watch out for? A review.

Pregnancy causes changes in all body systems, including the eye. The eye can undergo physiological and pathological changes in pregnancy. Some changes exacerbate pre-existing eye conditions while other conditions manifest for the first-time during pregnancy. Early recognition and management are essential to prevent sight threatening complications. In addition, some obstetric complications can be associated with ophthalmic signs. Prompt recognition of these eye findings may be life saving for both the mother and the foetus. The aim of this article is to present potential ocular complications in pregnancy and outline the appropriate management to preserve sight and maintain maternal and foetal safety. The safety of the use of common ophthalmological medications will also be discussed.

Temporal-Quality Ensemble Technique for Handling Image Blur in Packaging Defect Inspection.

Despite achieving numerous successes with surface defect inspection based on deep learning, the industry still faces challenges in conducting packaging defect inspections that include critical information such as ingredient lists. In particular, while previous achievements primarily focus on defect inspection in high-quality images, they do not consider defect inspection in low-quality images such as those containing image blur. To address this issue, we proposed a noble inference technique named temporal-quality ensemble (TQE), which combines temporal and quality weights. Temporal weighting assigns weights to input images by considering the timing in relation to the observed image. Quality weight prioritizes high-quality images to ensure the inference process emphasizes clear and reliable input images. These two weights improve both the accuracy and reliability of the inference process of low-quality images. In addition, to experimentally evaluate the general applicability of TQE, we adopt widely used convolutional neural networks (CNNs) such as ResNet-34, EfficientNet, ECAEfficientNet, GoogLeNet, and ShuffleNetV2 as the backbone network. In conclusion, considering cases where at least one low-quality image is included, TQE has an F1-score approximately 17.64% to 22.41% higher than using single CNN models and about 1.86% to 2.06% higher than an average voting ensemble.

ABDGAN: Arbitrary Time Blur Decomposition Using Critic-Guided TripleGAN.

Recent studies have proposed methods for extracting latent sharp frames from a single blurred image. However, these methods still suffer from limitations in restoring satisfactory images. In addition, most existing methods are limited to decomposing a blurred image into sharp frames with a fixed frame rate. To address these problems, we present an Arbitrary Time Blur Decomposition Triple Generative Adversarial Network (ABDGAN) that restores sharp frames with flexible frame rates. Our framework plays a min-max game consisting of a generator, a discriminator, and a time-code predictor. The generator serves as a time-conditional deblurring network, while the discriminator and the label predictor provide feedback to the generator on producing realistic and sharp image depending on given time code. To provide adequate feedback for the generator, we propose a critic-guided (CG) loss by collaboration of the discriminator and time-code predictor. We also propose a pairwise order-consistency (POC) loss to ensure that each pixel in a predicted image consistently corresponds to the same ground-truth frame. Extensive experiments show that our method outperforms previously reported methods in both qualitative and quantitative evaluations. Compared to the best competitor, the proposed ABDGAN improves PSNR, SSIM, and LPIPS on the GoPro test set by 16.67%, 9.16%, and 36.61%, respectively. For the B-Aist++ test set, our method shows improvements of 6.99%, 2.38%, and 17.05% in PSNR, SSIM, and LPIPS, respectively, compared to the best competitive method.

Case report of the evidence of a spontaneous Reverse Pulfrich effect in monovision after cataract surgery.

BACKGROUND: Cataracts affect the optics of the eye in terms of absorption, blur, and scattering. When cataracts are unilateral, they cause differences between the eyes that can produce visual discomfort and harm binocular vision. These interocular differences can also induce differences in the processing speed of the eyes that may cause a spontaneous Pulfrich effect, a visual illusion provoking important depth misperceptions. Interocular differences in light level, like those present in unilateral cataracts, can cause the Classic Pulfrich effect, and interocular differences in blur, like those present in monovision, a common correction for presbyopia, can cause the Reverse Pulfrich effect. The visual system may be able to adapt, or not, to the new optical condition, depending on the degree of the cataract and the magnitude of the monovision correction. CASE PRESENTATION: Here, we report a unique case of a 45-year-old patient that underwent unilateral cataract surgery resulting in a monovision correction of 2.5 diopters (D): left eye emmetropic after the surgery compensated with a monofocal intraocular lens and right eye myopic with a spherical equivalent of -2.50 D. This patient suffered severe symptoms in binocular vision, which can be explained by a spontaneous Pulfrich effect (a delay measured of 4.82 ms, that could be eliminated with a 0.19 optical density filter). After removing the monovision with clear lens extraction in the second eye, symptoms disappeared. We demonstrate that, at least in this patient, both Classic and Reverse Pulfrich effects coexist after unilateral cataract surgery and that can be readapted by reverting the interocular differences. Besides, we report that the adaptation/readaptation process to the Reverse Pulfrich effect happens in a timeframe of weeks, as opposed to the Classic Pulfrich effect, known to have timeframes of days. Additionally, we used the illusion measured in the laboratory to quantify the relevance of the spontaneous Pulfrich effect in different visual scenarios and tasks, using geometrical models and optic flow algorithms. CONCLUSIONS: Measuring the different versions of the Pulfrich effect might help to understand the visual discomfort reported by many patients after cataract surgery or with monovision and could guide compensation or intervention strategies.

Discovery, Quantitative Recurrence, and Inhibition of Motion-Blur Hysteresis Phenomenon in Visual Tracking Displacement Detection.

Motion blur is common in video tracking and detection, and severe motion blur can lead to failure in tracking and detection. In this work, a motion-blur hysteresis phenomenon (MBHP) was discovered, which has an impact on tracking and detection accuracy as well as image annotation. In order to accurately quantify MBHP, this paper proposes a motion-blur dataset construction method based on a motion-blur operator (MBO) generation method and self-similar object images, and designs APSF, a MBO generation method. The optimized sub-pixel estimation method of the point spread function (SPEPSF) is used to demonstrate the accuracy and robustness of the APSF method, showing the maximum error (ME) of APSF to be smaller than others (reduced by 86%, when motion-blur length > 20, motion-blur angle = 0), and the mean square error (MSE) of APSF to be smaller than others (reduced by 65.67% when motion-blur angle = 0). A fast image matching method based on a fast correlation response coefficient (FAST-PCC) and improved KCF were used with the motion-blur dataset to quantify MBHP. The results show that MBHP exists significantly when the motion blur changes and the error caused by MBHP is close to half of the difference of the motion-blur length between two consecutive frames. A general flow chart of visual tracking displacement detection with error compensation for MBHP was designed, and three methods for calculating compensation values were proposed: compensation values based on inter-frame displacement estimation error, SPEPSF, and no-reference image quality assessment (NR-IQA) indicators. Additionally, the implementation experiments showed that this error can be reduced by more than 96%.

Cascaded Degradation-Aware Blind Super-Resolution.

Image super-resolution (SR) usually synthesizes degraded low-resolution images with a predefined degradation model for training. Existing SR methods inevitably perform poorly when the true degradation does not follow the predefined degradation, especially in the case of the real world. To tackle this robustness issue, we propose a cascaded degradation-aware blind super-resolution network (CDASRN), which not only eliminates the influence of noise on blur kernel estimation but also can estimate the spatially varying blur kernel. With the addition of contrastive learning, our CDASRN can further distinguish the differences between local blur kernels, greatly improving its practicality. Experiments in various settings show that CDASRN outperforms state-of-the-art methods on both heavily degraded synthetic datasets and real-world datasets.

A 3.0 µm Pixels and 1.5 µm Pixels Combined Complementary Metal-Oxide Semiconductor Image Sensor for High Dynamic Range Vision beyond 106 dB.

We propose a new concept image sensor suitable for viewing and sensing applications. This is a report of a CMOS image sensor with a pixel architecture consisting of a 1.5 µm pixel with four-floating-diffusions-shared pixel structures and a 3.0 µm pixel with an in-pixel capacitor. These pixels are four small quadrate pixels and one big square pixel, also called quadrate-square pixels. They are arranged in a staggered pitch array. The 1.5 µm pixel pitch allows for a resolution high enough to recognize distant road signs. The 3 µm pixel with intra-pixel capacitance provides two types of signal outputs: a low-noise signal with high conversion efficiency and a highly saturated signal output, resulting in a high dynamic range (HDR). Two types of signals with long exposure times are read out from the vertical pixel, and four types of signals are read out from the horizontal pixel. In addition, two signals with short exposure times are read out again from the square pixel. A total of eight different signals are read out. This allows two rows to be read out simultaneously while reducing motion blur. This architecture achieves both an HDR of 106 dB and LED flicker mitigation (LFM), as well as being motion-artifact-free and motion-blur-less. As a result, moving subjects can be accurately recognized and detected with good color reproducibility in any lighting environment. This allows a single sensor to deliver the performance required for viewing and sensing applications.

Correction of motion artifact in CL based on MAFusNet.

Computed laminography (CL) is one of the best methods for nondestructive testing of plate-like objects. If the object and the detector move continually while the scanning is being done, the data acquisition efficiency of CL will be significantly increased. However, the projection images will contain motion artifact as a result. A multi-angle fusion network (MAFusNet) is presented in order to correct the motion artifact of CL projection images considering the properties of CL projection images. The multi-angle fusion module significantly increases the ability of MAFusNet to deblur by using data from nearby projection images, and the feature fusion module lessens information loss brought on by data flow between the encoders. In contrast to conventional deblurring networks, the MAFusNet network employs synthetic datasets for training and performed well on realistic data, proving the network's outstanding generalization. The multi-angle fusion-based network has a significant improvement in the correction effect of CL motion artifact through ablation study and comparison with existing classical deblurring networks, and the synthetic training dataset can also significantly lower the training cost, which can effectively improve the quality and efficiency of CL imaging in industrial nondestructive testing.

Critical factors in achieving fine-scale functional MRI: Removing sources of inadvertent spatial smoothing.

Ultra-high Field (≥7T) functional magnetic resonance imaging (UHF-fMRI) provides opportunities to resolve fine-scale features of functional architecture such as cerebral cortical columns and layers, in vivo. While the nominal resolution of modern fMRI acquisitions may appear to be sufficient to resolve these features, several common data preprocessing steps can introduce unwanted spatial blurring, especially those that require interpolation of the data. These resolution losses can impede the detection of the fine-scale features of interest. To examine quantitatively and systematically the sources of spatial resolution losses occurring during preprocessing, we used synthetic fMRI data and real fMRI data from the human visual cortex-the spatially interdigitated human V2 "thin" and "thick" stripes. The pattern of these cortical columns lies along the cortical surface and thus can be best appreciated using surface-based fMRI analysis. We used this as a testbed for evaluating strategies that can reduce spatial blurring of fMRI data. Our results show that resolution losses can be mitigated at multiple points in preprocessing pathway. We show that unwanted blur is introduced at each step of volume transformation and surface projection, and can be ameliorated by replacing multi-step transformations with equivalent single-step transformations. Surprisingly, the simple approaches of volume upsampling and of cortical mesh refinement also helped to reduce resolution losses caused by interpolation. Volume upsampling also serves to improve motion estimation accuracy, which helps to reduce blur. Moreover, we demonstrate that the level of spatial blurring is nonuniform over the brain-knowledge which is critical for interpreting data in high-resolution fMRI studies. Importantly, our study provides recommendations for reducing unwanted blurring during preprocessing as well as methods that enable quantitative comparisons between preprocessing strategies. These findings highlight several underappreciated sources of a spatial blur. Individually, the factors that contribute to spatial blur may appear to be minor, but in combination, the cumulative effects can hinder the interpretation of fine-scale fMRI and the detectability of these fine-scale features of functional architecture.

Long-term changes of choroidal blood flow velocity in Vogt-Koyanagi-Harada disease.

PURPOSE: To evaluate changes in choroidal blood flow in patients with Vogt-Koyanagi-Harada (VKH) disease after initiation of corticosteroid treatment. METHODS: Fourteen patients (10 men and 4 women) with acute VKH disease followed for 2 years were retrospectively reviewed; only right eyes were included in the analysis. Mean blur rate (MBR) in the macula was measured by laser speckle flowgraphy (LSFG) and central choroidal thickness (CCT) was measured by optical coherence tomography (OCT), both prior to treatment and over 2 years after initiation of corticosteroid treatment. RESULTS: Of 14 patients included in this study, 13 received initial treatment consisting of intravenous corticosteroid pulse therapy and one patient was treated using bilateral sub-Tenon injections of triamcinolone acetonide. Mean percentage change in MBR was significantly increased after initiation of treatment compared to pretreatment values (P < 0.001). Mean CCTs were significantly decreased after initiation of treatment, compared to pretreatment thicknesses (P < 0.001). There was no significant change in either MBR change or CCT at 1 month after initiation of treatment through 2 years of follow-up. The mean MBR percentage change was significantly higher in eyes with sunset glow fundus (SGF) compared to eyes without SGF at 1 year. CONCLUSION: With initiation of corticosteroid treatment in VKH disease patients, choroidal blood flow improved and was maintained for 2 years. However, the presence of SGF should be taken into consideration when interpreting MBR results in VKH disease patients.

Effects of glaucoma on detection and discrimination of image blur.

PURPOSE: Blur is one of the most commonly reported visual symptoms of glaucoma, but it is not directly measured by current clinical tests. We aimed to investigate the effects of glaucoma on detection and discrimination of image blur. METHODS: Participants were people with glaucoma, separated into two groups with (n = 15) or without (n = 17) central visual field defects measured by 10-2 perimetry, and an age-similar control group (n = 18). First, we measured contrast detection thresholds centrally using a 2-interval forced choice procedure. We then measured blur detection and discrimination thresholds for the same stimuli (reference blurs 0, 1 arcmin) using a 2-alternative forced choice procedure under two contrast conditions: 4× individual detection threshold for the low contrast condition; 95% contrast for the high contrast condition. The stimulus was a horizontal edge bisecting a hard-edged circle of 4.5° diameter. Data were analysed by linear mixed modelling. RESULTS: Contrast detection thresholds for the glaucoma group with central visual field defects were raised by 0.01 ± 0.004 (mean ± SE, Michelson units) (p = 0.002) and by 0.01 ± 0.004 (p = 0.03) relative to control and glaucoma without central visual field defect groups, respectively. Blur detection and discrimination thresholds were similar between groups, with small elevations in blur detection thresholds in the glaucoma groups not reaching statistical significance (detection p = 0.29, discrimination p = 0.91). The lower contrast level increased thresholds from the higher contrast level by 1.30 ± 0.10 arcmin (p < 0.001) and 1.05 ± 0.10 arcmin (p < 0.001) for blur detection and discrimination thresholds, respectively. CONCLUSIONS: Early-moderate glaucoma resulted in only minimal elevations of blur detection thresholds that did not reach statistical significance in this study. Despite the prevalence of blur as a visual symptom of glaucoma, psychophysical measurements of blur detection or discrimination may not be good candidates for development as clinical tests for glaucoma.

Low levels of refractive blur increase the risk of colour misperception of red train signals.

PURPOSE: Red signals signify danger in a range of situations, including train operations. Importantly, misperception of a red signal as yellow can have serious safety implications. This study investigated the effects of lens blur on incorrect colour perception of red signals, which has been implicated in previous train crashes. METHODS: Participants included 15 young (26.6 ± 4.6 years) and 15 older (55.8 ± 3.1 years) visually normal adults. Red and yellow wayside train signals were simulated for two brightness levels (dim, bright) using a custom-built projection system. The effect of blur (best-corrected refraction [No Blur], +0.25 DS, +0.50 DS, +0.75 DS, +1.00 DS, +1.25 DS) on the number of incorrect colour perception responses of the signals was recorded. The order of conditions was randomised between participants. RESULTS: For incorrect responses to the red signal, there were significant main effects of blur (p < 0.001) and signal brightness (p < 0.001) and a significant interaction between blur and brightness (p < 0.001). The effects of blur were greater for the dim compared to the bright signals, with significantly higher colour misperceptions for the dim signal for +0.50 DS blur and higher, compared with No Blur. Colour misperceptions of the yellow signals were low compared with that of the red signals, with only +1.25 DS blur resulting in a significantly higher number of incorrect responses than No Blur (p < 0.001). There were no effects of age for the red or yellow colour misperceptions (p > 0.19). CONCLUSIONS: Low levels of blur (+0.50 DS to +1.25 DS) resulted in a significant misperception of the red signals as orange-yellow, particularly for dim signals. The findings have implications for vision testing and refractive correction of train drivers to minimise the possibility of colour misperception of red train signals.

Motion Blur Kernel Rendering Using an Inertial Sensor: Interpreting the Mechanism of a Thermal Detector.

Various types of motion blur are frequently observed in the images captured by sensors based on thermal and photon detectors. The difference in mechanisms between thermal and photon detectors directly results in different patterns of motion blur. Motivated by this observation, we propose a novel method to synthesize blurry images from sharp images by analyzing the mechanisms of the thermal detector. Further, we propose a novel blur kernel rendering method, which combines our proposed motion blur model with the inertial sensor in the thermal image domain. The accuracy of the blur kernel rendering method is evaluated by the task of thermal image deblurring. We construct a synthetic blurry image dataset based on acquired thermal images using an infrared camera for evaluation. This dataset is the first blurry thermal image dataset with ground-truth images in the thermal image domain. Qualitative and quantitative experiments are extensively carried out on our dataset, which show that our proposed method outperforms state-of-the-art methods.

An Efficient Defocus Blur Segmentation Scheme Based on Hybrid LTP and PCNN.

The defocus or motion effect in images is one of the main reasons for the blurry regions in digital images. It can affect the image artifacts up to some extent. However, there is a need for automatic defocus segmentation to separate blurred and sharp regions to extract the information about defocus-blur objects in some specific areas, for example, scene enhancement and object detection or recognition in defocus-blur images. The existence of defocus-blur segmentation algorithms is less prominent in noise and also costly for designing metric maps of local clarity. In this research, the authors propose a novel and robust defocus-blur segmentation scheme consisting of a Local Ternary Pattern (LTP) measured alongside Pulse Coupled Neural Network (PCNN) technique. The proposed scheme segments the blur region from blurred fragments in the image scene to resolve the limitations mentioned above of the existing defocus segmentation methods. It is noticed that the extracted fusion of upper and lower patterns of proposed sharpness-measure yields more noticeable results in terms of regions and edges compared to referenced algorithms. Besides, the suggested parameters in the proposed descriptor can be flexible to modify for performing numerous settings. To test the proposed scheme's effectiveness, it is experimentally compared with eight referenced techniques along with a defocus-blur dataset of 1000 semi blurred images of numerous categories. The model adopted various evaluation metrics comprised of Precision, recall, and F1-Score, which improved the efficiency and accuracy of the proposed scheme. Moreover, the proposed scheme used some other flavors of evaluation parameters, e.g., Accuracy, Matthews Correlation-Coefficient (MCC), Dice-Similarity-Coefficient (DSC), and Specificity for ensuring provable evaluation results. Furthermore, the fuzzy-logic-based ranking approach of Evaluation Based on Distance from Average Solution (EDAS) module is also observed in the promising integrity analysis of the defocus blur segmentation and also in minimizing the time complexity.

The Way to Modern Shutter Speed Measurement Methods: A Historical Overview.

Exposure time is a fundamental parameter for the photographer when the photo is composed, and the exact length of the exposure may be an essential determinant of performance in certain camera-based applications, e.g., optical camera communication (OCC) systems. There can be several reasons to measure the shutter speed of a camera: shutter speed may be checked at the time of manufacturing; it may be necessary to recheck in case of an elder camera model; it may be necessary to be measured if its exact value is not provided by the manufacturer; or a precise measurement may be necessary for a demanding application. In this paper various methods for shutter speed measurement are reviewed, presenting and analyzing methods that are still relevant today either for manufacturers, service personnel, amateur photographers, or the developers of camera-based systems. Each presented method is illustrated by real measurement results and the performance properties of the methods are also presented.

An Image Augmentation Method Based on Limited Samples for Object Tracking Based on Mobile Platform.

This paper proposes an image augmentation model of limited samples on the mobile platform for object tracking. The augmentation method mainly aims at the detection failure caused by the small number of effective samples, jitter of tracking platform, and relative rotation between camera and object in the tracking process. Aiming at the object tracking problem, we first propose to use geometric projection transformation, multi-directional overlay blurring, and random background filling to improve the generalization ability of samples. Then, selecting suitable traditional augmentation methods as the supplements, an image augmentation model with an adjustable probability factor is provided to simulate various kinds of samples to help the detection model carry out more reliable training. Finally, combined with a spatial localization algorithm based on geometric constraints proposed by the author's previous work, a framework for object tracking with an image augmentation method is proposed. SSD, YOLOv3, YOLOv4, and YOLOx are adopted in the experiment of this paper as the detection models. And a large number of object recognition and object tracking experiments are carried out by combining with common data sets OTB50 and OTB100 as well as the OTMP data set proposed by us for mobile platform. The augmented module proposed in this paper is conducive for the detection model to improve the detection accuracy by at least 10%. Especially for objects with planar characteristics, the affine and projection transformation used in this paper can greatly improve the detection accuracy of the model. Based on the object tracking framework of our augmented model, the RMSE is estimated to be less than 4.21 cm in terms of the actual tracking of indoor objects.

Reconstructing a Deblurred 3D Structure in a Turbid Medium from a Single Blurred 2D Image-For Near-Infrared Transillumination Imaging of a Human Body.

To provide another modality for three-dimensional (3D) medical imaging, new techniques were developed to reconstruct a 3D structure in a turbid medium from a single blurred 2D image obtained using near-infrared transillumination imaging. One technique uses 1D information of a curvilinear absorber, or the intensity profile across the absorber image. Profiles in different conditions are calculated by convolution with the depth-dependent point spread function (PSF) of the transillumination image. In databanks, profiles are stored as lookup tables to connect the contrast and spread of the profile to the absorber depth. One-to-one correspondence from the contrast and spread to the absorber depth and thickness were newly found. Another technique uses 2D information of the transillumination image of a volumetric absorber. A blurred 2D image is deconvolved with the depth-dependent PSF, thereby producing many images with points of focus on different parts. The depth of the image part can be estimated by searching the deconvolved images for the image part in the best focus. To suppress difficulties of high-spatial-frequency noise, we applied a noise-robust focus stacking method. Experimentation verified the feasibility of the proposed techniques, and suggested their applicability to curvilinear and volumetric absorbers such as blood vessel networks and cancerous lesions in tissues.

Pantograph Detection Algorithm with Complex Background and External Disturbances.

As an important equipment for high-speed railway (HSR) to obtain electric power from outside, the state of the pantograph will directly affect the operation safety of HSR. In order to solve the problems that the current pantograph detection method is easily affected by the environment, cannot effectively deal with the interference of external scenes, has a low accuracy rate and can hardly meet the actual operation requirements of HSR, this study proposes a pantograph detection algorithm. The algorithm mainly includes three parts: the first is to use you only look once (YOLO) V4 to detect and locate the pantograph region in real-time; the second is the blur and dirt detection algorithm for the external interference directly affecting the high-speed camera (HSC), which leads to the pantograph not being detected; the last is the complex background detection algorithm for the external complex scene "overlapping" with the pantograph when imaging, which leads to the pantograph not being recognized effectively. The dirt and blur detection algorithm combined with blob detection and improved Brenner method can accurately evaluate the dirt or blur of HSC, and the complex background detection algorithm based on grayscale and vertical projection can greatly reduce the external scene interference during HSR operation. The algorithm proposed in this study was analyzed and studied on a large number of video samples of HSR operation, and the precision on three different test samples reached 99.92%, 99.90% and 99.98%, respectively. Experimental results show that the algorithm proposed in this study has strong environmental adaptability and can effectively overcome the effects of complex background and external interference on pantograph detection, and has high practical application value.