Circular Markers Research Articles

Facial and mandibular landmark tracking with habitual head posture estimation using linear and fiducial markers.

This study compared the accuracy of facial landmark measurements using deep learning-based fiducial marker (FM) and arbitrary width reference (AWR) approaches. It quantitatively analysed mandibular hard and soft tissue lateral excursions and head tilting from consumer camera footage of 37 participants. A custom deep learning system recognised facial landmarks for measuring head tilt and mandibular lateral excursions. Circular fiducial markers (FM) and inter-zygion measurements (AWR) were validated against physical measurements using electrognathography and electronic rulers. Results showed notable differences in lower and mid-face estimations for both FM and AWR compared to physical measurements. The study also demonstrated the comparability of both approaches in assessing lateral movement, though fiducial markers exhibited variability in mid-face and lower face parameter assessments. Regardless of the technique applied, hard tissue movement was typically seen to be 30% less than soft tissue among the participants. Additionally, a significant number of participants consistently displayed a 5 to 10° head tilt.

Fast and Omnidirectional Relative Position Estimation With Circular Markers for UAV Swarm

Fast and Omnidirectional Relative Position Estimation With Circular Markers for UAV Swarm

Efficient 6-DoF camera pose tracking with circular edges

Camera pose tracking attracts much interest from both academic and industrial communities, of which the methods based on planar markers are easy to be implemented. However, most existing methods need to identify multiple points in the marker images for matching to space points. Then, PnP methods are used to compute the camera poses. If cameras move fast or are far away from the markers, the matching is easy to generate errors. To address these problems, we design circular markers and represent 6D camera pose analytically as concise forms from each marker by projective invariance. Afterwards, the pose is further optimized by a cost function based on a polar-n-direction geometric distance. The proposed method is from imaged circular edges, which makes camera pose tracking more robust to noise, blur and distance from camera to marker than existing methods. Extensive experimental results show that the proposed 6-DoF camera pose tracking method outperforms state-of-the-art methods in terms of noise, blur, and distance from camera to marker. Simultaneously, it efficiently runs at about 100 FPS on a consumer computer.

3D observation instrument for pathology tissues

Pathology telemedicine uses digital data of pathological samples to acquire technical support from sophisticated remote experts, aiding cancer diagnosis in less developed regions. In this paper, we propose a color 3D observation system composed of three cameras and a digital projector, achieving complete range 3D imaging of pathology tissues. A new phase mapping method for the projector calibration is proposed by locating the geometric centroid of the checkboard’s white squares, avoiding the center positioning error of traditional circular markers introduced by perspective distortion. A denoising method successively using amplitude modulation judge, phase monotonicity check, and phase abruptness constraint is proposed, which facilitates obtaining clean 3D point clouds.

Forensic age estimation from human blood using age-related microRNAs and circular RNAs markers.

Aging is a complicated process characterized by progressive and extensive changes in physiological homeostasis at the organismal, tissue, and cellular levels. In modern society, age estimation is essential in a large variety of legal rights and duties. Accumulating evidence suggests roles for microRNAs (miRNAs) and circular RNAs (circRNAs) in regulating numerous processes during aging. Here, we performed circRNA sequencing in two age groups and analyzed microarray data of 171 healthy subjects (17-104years old) downloaded from Gene Expression Omnibus (GEO) and ArrayExpress databases with integrated bioinformatics methods. A total of 1,403 circular RNAs were differentially expressed between young and old groups, and 141 circular RNAs were expressed exclusively in elderly samples while 10 circular RNAs were expressed only in young subjects. Based on their expression pattern in these two groups, the circular RNAs were categorized into three classes: age-related expression between young and old, age-limited expression-young only, and age-limited expression-old only. Top five expressed circular RNAs among three classes and a total of 18 differentially expressed microRNAs screened from online databases were selected to validate using RT-qPCR tests. An independent set of 200 blood samples (20-80years old) was used to develop age prediction models based on 15 age-related noncoding RNAs (11 microRNAs and 4 circular RNAs). Different machine learning algorithms for age prediction were applied, including regression tree, bagging, support vector regression (SVR), random forest regression (RFR), and XGBoost. Among them, random forest regression model performed best in both training set (mean absolute error = 3.68years, r = 0.96) and testing set (MAE = 6.840years, r = 0.77). Models using one single type of predictors, circular RNAs-only or microRNAs-only, result in bigger errors. Smaller prediction errors were shown in males than females when constructing models according to different-sex separately. Putative microRNA targets (430 genes) were enriched in the cellular senescence pathway and cell homeostasis and cell differentiation regulation, indirectly indicating that the microRNAs screened in our study were correlated with development and aging. This study demonstrates that the noncoding RNA aging clock has potential in predicting chronological age and will be an available biological marker in routine forensic investigation to predict the age of biological samples.

Early detection of colorectal cancer based on circular DNA and common clinical detection indicators.

BACKGROUNDColorectal cancer (CRC) is the third most common cancer worldwide, and it is the second leading cause of death from cancer in the world, accounting for approximately 9% of all cancer deaths. Early detection of CRC is urgently needed in clinical practice.AIMTo build a multi-parameter diagnostic model for early detection of CRC. METHODSTotal 59 colorectal polyps (CRP) groups, and 101 CRC patients (38 early-stage CRC and 63 advanced CRC) for model establishment. In addition, 30 CRP groups, and 62 CRC patients (30 early-stage CRC and 32 advanced CRC) were separately included to validate the model. 51 commonly used clinical detection indicators and the 4 extrachromosomal circular DNA markers NDUFB7, CAMK1D, PIK3CD and PSEN2 that we screened earlier. Four multi-parameter joint analysis methods: binary logistic regression analysis, discriminant analysis, classification tree and neural network to establish a multi-parameter joint diagnosis model. RESULTSNeural network included carcinoembryonic antigen (CEA), ischemia-modified albumin (IMA), sialic acid (SA), PIK3CD and lipoprotein a (LPa) was chosen as the optimal multi-parameter combined auxiliary diagnosis model to distinguish CRP and CRC group, when it differentiated 59 CRP and 101 CRC, its overall accuracy was 90.8%, its area under the curve (AUC) was 0.959 (0.934, 0.985), and the sensitivity and specificity were 91.5% and 82.2%, respectively. After validation, when distinguishing based on 30 CRP and 62 CRC patients, the AUC was 0.965 (0.930-1.000), and its sensitivity and specificity were 66.1% and 70.0%. When distinguishing based on 30 CRP and 32 early-stage CRC patients, the AUC was 0.960 (0.916-1.000), with a sensitivity and specificity of 87.5% and 90.0%, distinguishing based on 30 CRP and 30 advanced CRC patients, the AUC was 0.970 (0.936-1.000), with a sensitivity and specificity of 96.7% and 86.7%.CONCLUSIONWe built a multi-parameter neural network diagnostic model included CEA, IMA, SA, PIK3CD and LPa for early detection of CRC, compared to the conventional CEA, it showed significant improvement.

Vision-Based Detection of Bolt Loosening Using YOLOv5.

Bolted connections have been widely applied in engineering structures, loosening will happen when bolted connections are subjected to continuous cyclic load, and a significant rotation between the nut and the bolt can be observed. Combining deep learning with machine vision, a bolt loosening detection method based on the fifth version of You Only Look Once (YOLOv5) is proposed, and the rotation of the nut is identified to detect the bolt loosening. Two different circular markers are added to the bolt and the nut separately, and then YOLOv5 is used to identify the circular markers, and the rotation angle of the nut against the bolt is calculated according to the center coordinate of each predicted box. A bolted connection structure is adopted to illustrate the effectiveness of the method. First, 200 images containing bolts and circular markers are collected to make the dataset, which is divided into a training set, verification set and test set. Second, YOLOv5 is used to train the model; the precision rate and recall rate are respectively 99.8% and 100%. Finally, the robustness of the proposed method in different shooting environments is verified by changing the shooting distance, shooting angle and light condition. When using this method to detect the bolt loosening angle, the minimum identifiable angle is 1°, and the maximum detection error is 5.91% when the camera is tilted 45°. The experimental results show that the proposed method can detect the loosening angle of the bolted connection with high accuracy; especially, the tiny angle of bolt loosening can be identified. Even under some difficult shooting conditions, the method still works. The early stage of bolt loosening can be detected by measuring the rotation angle of the nut against the bolt.

Position-adjusted subset tracking analysis coupled with a double circle marker

A novel digital image displacement measurement system using a position-adjusted subset tracking analysis (PASTA) method that tracks subsets in the x- and y-directions coupled with double circular markers is proposed in this study. The PASTA method realizes the subpixel resolution in x- and y-directions based on differences in the luminance of subsets. In this method, a subset is defined as a part of the diameter of the inner circular marker (size of the subset = 1 × d or d × 1), which includes the two-colored edges of the double circular marker. The circular marker makes it easy to track the location of the subset in the x- or y-direction. This enables the calculation of the displacement in the x- and y-directions with high resolution and reduces errors. In numerical simulations, images with displacements of -2–2 pixels in one- or two-dimensions were prepared, and the measurement resolution, measurement range, and validity were verified. By focusing on the pixel rows or columns of the diameter of the circular marker and determining the subset position, we confirmed that we can determine the luminance change affected only by the displacement in the x- or y-direction. We experimentally verified the validity of the PASTA method and evaluated the robustness of the measurement system. The mean absolute error was less than approximately 15 μm within the measurement range and approximately 40 μm outside the measurement range. Within the measurement range, the relationship between the change in the luminance value and the displacement is linear, and stable measurement is possible. The PASTA method can measure a displacement of 50 × 10−3 mm with sufficient tracking.

Artificial landmarks and areas’ accuracy gradient: a three-dimensional analysis of trueness and precision of IOS scans on the total edentulous maxilla

To assessing the areas’ accuracy gradient of the intraoral scanner (IOS) scans on a completely edentulous maxilla and the effectiveness of artificial landmarks. A completely edentulous maxillary typodont (MT) was scanned with an industrial metrological machine to obtain a reference scan in .stl format. Ten scans were made on MT with an IOS, subsequently superimposed to detect the areas’ accuracy gradient on MT scans. Artificial landmarks of 1-mm diameter were placed in the less accurate areas following two methods. In the first, flow-composite semispherical markers were used. In the second, circular markers were drawn with a dermographic pen. Three experimental groups were made (n=10): “No-Markers” for the control group (without markers), “Embossed-markers” for flow-composite markers, and “Flat-Markers” for ink-drawn ones. The scans were imported into a dedicated software, then trueness and precision were evaluated in µm. In addition to descriptive statistics (95% CI), the Games-Howell and Kruskal-Wallis tests were run to analyze the difference among groups (α=.05). Mean values for trueness were No-Markers=48.8 (39.2-58.3); Embossed-Markers=39.2 (37.5-40.8); Flat-Markers=60.5 (47.7-73.4), with statistically significant differences for Embossed-Markers/Flat-Markers (P=.011). Mean values for precision were No-Markers=46.7 (29.7-63.7); Embossed-Markers=41.4 (34.7-48); Flat-Markers=99.8 (69.3-130.3), with significant differences for Embossed-Markers/Flat-Markers (P=0.008) and No-Markers/Flat-Markers (P=0.005). The most inaccurate areas were tuberosities, palate, and ridges’ flattened areas. To increase the IOS scans accuracy on the completely edentulous maxilla, it is suggested to place embossed flow-composite markers specifically in the areas of greatest inaccuracy, but to avoid using flat markers.

Bird protection treatments reduce bird-window collision risk at low-rise buildings within a Pacific coastal protected area.

BackgroundIn North America, up to one billion birds are estimated to die annually due to collisions with glass. The transparent and reflective properties of glass present the illusion of a clear flight passage or continuous habitat. Approaches to reducing collision risk involve installing visual cues on glass that enable birds to perceive glass as a solid hazard at a sufficient distance to avoid it.MethodsWe monitored for bird-window collisions between 2013 and 2018 to measure response to bird protection window treatments at two low-rise buildings at the Alaksen National Wildlife Area in Delta, British Columbia, Canada. After 2 years of collision monitoring in an untreated state, we retrofitted one building with Feather Friendly® circular adhesive markers applied in a grid pattern across all windows, enabling a field-based assessment of the relative reduction in collisions in the 2 years of monitoring following treatment. An adjacent building that had been constructed with a bird protective UV-treated glass called ORNILUX® Mikado, was monitored throughout the two study periods. Carcass persistence trials were conducted to evaluate the likelihood that carcasses were missed due to carcass removal between scheduled searches.Results and ConclusionsAfter accounting for differences in area of glass between the two buildings, year, and observer effects, our best-fit model for explaining collision risk included the building’s treatment group, when compared to models that included building and season only. We found that the Feather Friendly® markers reduced collision risk at the retrofitted building by 95%. Collision incidence was also lower at the two monitored façades of the building with ORNILUX® glass compared to the building with untreated glass. Although more research is needed on the effectiveness of bird-protection products across a range of conditions, our results highlight the benefit of these products for reducing avian mortality due to collisions with glass.

Perspective transformation based-initial value estimation for the speckle control points matching in an out-of-focus camera calibration using a synthetic speckle pattern.

Despite camera calibration methods using regular planar chessboard or circular marker array calibration targets having been widely used, the control point extraction accuracy is low if the image is defocused or if the noise level is high. Due to the noise robustness of digital image correlation (DIC) in speckle image matching, random speckle pattern is a better choice for camera calibration than chessboard or circular markers, if the imaging quality is low. The foremost process of this method is to conduct speckle control points matching DIC, where the initial value must be estimated close to the true value. It is challenging to provide accurate initial values for DIC if the difference of physical pixel scale is large between the reference image and the target image or if the target image is out-of-focus. To solve this problem, this work presents an efficient initial value estimation method for speckle control points matching using DIC, based on perspective transformation. Firstly, the four pairs of corners of the speckle regions in the reference image and target image are detected. Secondly, the target image is transformed to a new image that has the considerable size of pixel scale with the reference image, then four neighborhood points of the control point in the reference image and the corresponding points in the transformed new image are matched coarsely by fixed subset searching. Lastly, the matched points in the transformed target image are transformed back to the origin target image by the inverse perspective transformation matrix, then the initial value for DIC can be estimated by the matched four pairs of neighborhood points. Experiment results confirm the higher calibration accuracy delivered by the proposed method, rather than that of the chessboard or the circular marker array. Measurement precision is higher than the speckle pattern calibration method that uses SIFT-based initial value estimation.

The Effect of the Upper and Lower Electrochemical Treatment Potentials on the Electrode Activity Towards Vanadium Redox Couples

Vanadium flow batteries (VFBs) are a promising energy storage technology. They have long cycle lives and are essentially immune to cross contamination through their ion-exchange membranes1. Battery energy efficiency depends on voltage and coulombic efficiency. Sluggish electrode kinetics reduces voltage efficiency by increasing overpotential at the electrodes and thereby can result in an increase in the fraction of current associated with unwanted side reactions at the electrodes, reducing columbic efficiency. It follows that study of the VFB electrode kinetics is very important.The kinetics of the redox couples at the positive and negative electrodes have been reported in flow batteries and three-electrode cell experiments for a range of carbon materials. The reports show that the kinetics of these reactions are not fully understood and discrepancies exist regarding which half-cell redox couple has slower kinetics. Many researchers2-4 have concluded that the redox couple has slower kinetics but some4-6 have concluded that the redox couple has slower kinetics.Carbon electrodes are very sensitive to changes in their environment which can subsequently alter the kinetics of redox couples.10 Thus, comparing the reported kinetics in the literature is complicated. We have previously reported7-9for a range of carbon materials – including carbon fibres, glassy carbon, reticulated vitreous carbon (RVC) and carbon paper – that the kinetics of are enhanced by cathodic treatment of the electrode and inhibited by anodic treatment. In contrast, the electrode kinetics of are enhanced by anodic treatment of the electrode and inhibited by cathodic treatment. Therefore, it is more reasonable when comparing the electrode kinetics towards different redox couples to compare electrodes that have been treated in the same manner.We showed previously7-9 that the optimised kinetics at the positive electrode is faster than at the negative electrode. In this presentation we further investigate the effect of the electrochemical treatment on the electrode kinetics by controlling the upper and lower treatment potentials experienced by the electrode. We will present the resulting dependency of electrode kinetics on the range of the electrochemical treatment window experienced by the electrode and will show, as can be seen in Fig. 1, that determining which redox couple has the fastest kinetics at an electrode is highly dependent on the range of potentials experienced by the electrode. Fig.1: The activity of VIV-VV (circular markers) and VII-VIII (triangular markers) electrodes plotted as a function of the anodic (red markers) and cathodic (blue markers) treatment potentials for two different potential windows/ranges of maximum upper and minimum lower potential. It is clearly shown that the optimised kinetics of VII-VIII exceed those of VIV-VV for a window of upper and lower potentials of –1.2 V and +1.2 V, respectively (graph on left), while the optimised kinetics of VIV-VV exceed those of VII-VIII for a window of upper and lower potentials of –1.2 V and +1.5 V, respectively (graph on right).

Retracted] High‐Precision 3D Reconstruction of Cooperative Markers under Motion Blur

The application of artificial intelligence and deep learning in the fields of wireless communication, image and speech recognition, and 3D reconstruction has successfully solved some difficult modeling problems. This paper focuses on the high‐precision 3D reconstruction of the motion‐blurred cooperative markers, including the Chinese character coded targets (CCTs) and the noncoded circular markers. A simulation‐based motion‐blurred image generation model is constructed to provide sufficient samples for training the convolutional neural network to identify and match the motion‐blurred CCTs on the moving object. The blurred noncoded marker matching is performed through homography. The 3D reconstruction of the markers is realized via the optimization of the spatial moving path within the exposure period. The midpoint of the moving path of the markers is taken as the final reconstruction result. The experimental results show that the 3D reconstruction accuracy of the markers with a certain motion blur effect is about 0.08 mm.

Robust circular marker localization under non-uniform illuminations based on homomorphic filtering

Circular markers are widely used in image-based measurement fields, e.g. vision system calibration, displacement measurement, and vibration monitoring. Generally, the marker positioning error inevitably leads to inaccurate results. As one of the key factors, the non-uniform illumination seriously affects the localization precision of circular markers. In order to reduce the influences of uneven lighting, a homomorphic filtering-gray weighted centroid method (HF-GWCM) is presented in this work. Firstly, the basic principles of both HF and its combination with GWCM are introduced. Subsequently, some simulation examples are conducted to analyze the localization accuracy of concentric circle markers under non-uniform illumination situations, which comprise the influences of HF parameters and different localization methods. Finally, a single camera imaging system with multiple light sources is employed to perform some tests for the effectiveness validation of HF-GWCM. Both positioning accuracy and stability of HF-GWCM are analyzed under different uneven illumination situations, including lighting movement, multiple light sources, and shaded region. Results demonstrate that HF-GWCM owns the ability to resist the interference of non-uniform illuminations, and it is also an effective way to improve the localization accuracy of circular markers.

Visual Positioning of Distant Wall-Climbing Robots Using Convolutional Neural Networks

Detection of visual markers, such as circular markers or quick response codes, is a commonly used approach to the positioning of wall-climbing robots. However, when the camera is far from the wall-climbing robot (e.g., 20 m), these markers become extremely blurred and difficult to detect. In this paper, a convolutional neural network-based positioning scheme comprised of a global bounding box detector and local wheel detector is proposed. The light-weight local wheel detector can quickly and accurately detect the four wheel points of a distant wall-climbing robot, and the detected wheel points can be used for calculating its position and direction angle. Our wheel detector has a single-frame processing time of 72.2 ms on a CPU and 7.1 ms on a GPU, where the latter meets the real-time positioning requirements of the wall-climbing robot. We also developed an efficient cost function for wheel matching between video frames. Simulation results and multiple test videos confirmed that the proposed cost function can match wheels between video frames perfectly. The high performance of this positioning system indicates that it may be used in a variety of industrial applications.

An extrinsic calibration method for multiple RGB-D cameras in a limited field of view

A 3D scanning system involving multiple RGB-D cameras has the potential to accelerate the reconstruction of an object and improve the measurement accuracy because it can capture an object in a comprehensive way. However, the extrinsic calibration of a multi-RGB-D-camera system is a fundamental and challenging problem, especially in a limited field of view. In this work, a system with four SR300 cameras on a platform with a size of approximately 0.64 m2 and an extrinsic calibration method assisted by a tower calibration pattern with circular markers in a limited field of view are proposed. The Hough transform algorithm is used to identify the centres of the circular markers in a colour image, and then the 3D coordinates are extracted by employing the alignment relationship between the colour image and the depth image. An improved adaptive cuckoo search (IACS) algorithm and unit quaternion are proposed to optimize the extrinsic parameters. The effectiveness of the IACS is verified by a comparison with the singular value decomposition (SVD), standard cuckoo search (CS) and two CS variant algorithms. In addition, the accuracy of the proposed extrinsic calibration method is verified by reconstructing and measuring a human foot and a cube box. The mean errors from the standard values of the length and width of the reconstructed foot mode are 0.18 cm and 0.13 cm, and the diameter of the yellow sphere are 0.17 cm, respectively. All of the experimental results show that the proposed extrinsic calibration method has a high calibration accuracy and good potential application prospects.

Machine-vision-guided rotation axis alignment for an optomechanical derotator

An image derotator is a device that can compensate the image blur caused by the rotation of an object. However, a key problem when using a derotator is to ensure the coincident alignment between the rotation axes of the derotator and measured object. An approach of machine-vision-guided rotation axis alignment is proposed in this paper. Circular markers and a laser dot are deployed to indicate the rotation axes of the derotator and object, respectively. A binocular stereo sensor is established to estimate the poses of both axes. With the misalignment information between the two axes updated in real-time, axis alignment can be achieved through manual adjustment. The validity and precision of the proposed approach are demonstrated through experiments.

Development of a positioning system using UAV-based computer vision for an airboat navigation in paddy field

Currently, a few agricultural corporations are developing and producing a kind of radio-controlled agricultural airboat which could be used to weed and fertilize instead of human in the paddy fields. The authors improved the airboat realizing autonomous navigation by a global-positioning-system (GPS) compass in the previous research. To solve the limitation of GPS-based navigation in bad environment (e.g., areas with tall trees, bad weather conditions, etc.), a new positioning system was proposed by using a computer vision system mounted on an unmanned aerial vehicle (UAV) to provide position information for an agricultural airboat engaged in autonomous fertilizing and herbicide application in this study. The computer vision system included a mini-computer, a color camera with a wide-angle lens, and a Bluetooth transmitter. The UAV was operated to hover high enough to overlook the paddy field by the computer vision system. The colored concentric circular markers were made to put at the corners of the paddy field to facilitate the paddy field recognition using Fitzgibbon ellipse fitting algorithm. Utilize the geographic position by GPS and image position by computer vision system of each markers to obtain the perspective transformation matrix for transforming the top view image of the paddy field. At the created paddy-field-based coordinate, the UAV detects the airboat position by using the white color feature and central symmetrical contour feature of the airboat and transmits the position information to airboat in real time by Bluetooth for airboat navigation. The results of navigation experiment showed that the RMS lateral errors were 0.17 m, 0.10 m and 0.11 m in 3 predefined paths, respectively. This accuracy level was better than that of differential-GPS (DGPS). It is acceptable for providing positioning service for the airboat autonomous navigation in paddy fields. Furthermore, this UAV-airboat solution was a beneficial attempt which can not only prevent the instability of GPS-based airboat navigation in bad environment, but also avoid the herbicide spray drift problem that occurs in UAV-spraying in the air.

Radiography calibration marker detection using Hough transformation

In this paper we analyse the possibility of simple detection of circular radiography markers. To detect the marker, we utilised the Hough transform. Two approaches were analysed: with detecting image edges and without image edge detection where pixel gradient was used in the Hough voting process, i.e. to increase the accumulator values. Approaches were evaluated on 13 clinical radiography images. It was shown that the approach that detects image edges spatially matches the reference circles only 0.22 % less than manual annotation values, whereas the approach that uses just the gradient magnitudes spatially matches the reference circles 3.2 % less than manual annotations.

Optical positioning technology of an assisted puncture robot based on binocular vision

AbstractPercutaneous image‐guided interventions are increasing in number in clinical practice because they are minimally invasive. Needle positioning placement is crucial and highly dependent on the physician's skills and experience, it is often the longest part of the intervention. Medical robotics and computer‐assisted surgery are hotspots in the field of robotics and medicine, changing the essence of traditional surgery using a combination of robotic, image processing, and computer technologies. The present paper aimed to study the auxiliary puncture procedure using a robot based on optical positioning technology that can be used to mark points in puncturing operation. Binocular camera is used for image acquisition, and Zhang's calibration method is used to establish the binocular camera model. In addition, the circular markers are identified by the least square method detection circle, and the coordinate information of the markers in three‐dimensional space is solved by using the visual depth information of binocular phases. This paper studies the verification of the three‐dimensional bone model of the human body, which lays a foundation for the application of the assistant puncture robot.