Accurate Localization Research Articles

Accurate object localization facilitates automatic esophagus segmentation in deep learning

BackgroundCurrently, automatic esophagus segmentation remains a challenging task due to its small size, low contrast, and large shape variation. We aimed to improve the performance of esophagus segmentation in deep learning by applying a strategy that involves locating the object first and then performing the segmentation task.MethodsA total of 100 cases with thoracic computed tomography scans from two publicly available datasets were used in this study. A modified CenterNet, an object location network, was employed to locate the center of the esophagus for each slice. Subsequently, the 3D U-net and 2D U-net_coarse models were trained to segment the esophagus based on the predicted object center. A 2D U-net_fine model was trained based on the updated object center according to the 3D U-net model. The dice similarity coefficient and the 95% Hausdorff distance were used as quantitative evaluation indexes for the delineation performance. The characteristics of the automatically delineated esophageal contours by the 2D U-net and 3D U-net models were summarized. Additionally, the impact of the accuracy of object localization on the delineation performance was analyzed. Finally, the delineation performance in different segments of the esophagus was also summarized.ResultsThe mean dice coefficient of the 3D U-net, 2D U-net_coarse, and 2D U-net_fine models were 0.77, 0.81, and 0.82, respectively. The 95% Hausdorff distance for the above models was 6.55, 3.57, and 3.76, respectively. Compared with the 2D U-net, the 3D U-net has a lower incidence of delineating wrong objects and a higher incidence of missing objects. After using the fine object center, the average dice coefficient was improved by 5.5% in the cases with a dice coefficient less than 0.75, while that value was only 0.3% in the cases with a dice coefficient greater than 0.75. The dice coefficients were lower for the esophagus between the orifice of the inferior and the pulmonary bifurcation compared with the other regions.ConclusionThe 3D U-net model tended to delineate fewer incorrect objects but also miss more objects. Two-stage strategy with accurate object location could enhance the robustness of the segmentation model and significantly improve the esophageal delineation performance, especially for cases with poor delineation results.

Study on Fault Monitoring Technology of Photovoltaic Panel Based on Thermal Infrared and Optical Remote Sensing

Abstract. Rapid access to the operating status of Photovoltaic (PV) panels and troubleshooting can save management and maintenance costs for the development of PV power plants, which is important for PV power plant management and power generation capacity assurance. The use of remote sensing technology to identify the faults of photovoltaic panels has developed rapidly, however, current research usually relies only on a single optical data source to identify and count the area of PV panels in a PV electric field, although there are literature on PV panel fault detection, only the surface fault identification of PV panels is tested, while the internal faults (such as panel bad points or bad lines) cannot be identified because of the limitations of optical remote sensing. In this paper, a photovoltaic panel fault monitoring technology based on multi-source remote sensing is proposed. The optical and thermal infrared hybrid data combined with deep learning technology are used to achieve rapid and accurate fault identification and localization of PV panel arrays. It can not only automatically identify PV panels that are obscured by dust and foreign objects, but also locate PV panels that have bad dots or bad lines, which greatly improves the ability and effectiveness of remote sensing PV panel fault monitoring. The high-resolution unmanned air vehicle (UAV) optical image and thermal infrared image are applied in this experiment. The Mask RCNN algorithm is used to accurately locate and number the photovoltaic panel of the optical image. Then, the fault scene classification model is established for the multi-type fault characteristics of the optical image and thermal infrared image within the panel range, so as to identify five types of faults, such as dust cover, branch cover, bird droppings cover, internal bad points and bad lines of PV panel, which effectively solves the problem that the single optical remote sensing image cannot identify the internal component faults of the photovoltaic panel under normal conditions.

Analysis of Multiple Scattering Characteristics of Cable-Stayed Bridges with Multi-band SAR

Abstract. Accurate localization of multi-scattering features of cable-stayed bridges in multi-band Synthetic Aperture Radar (SAR) imagery is crucial for intelligent recognition of bridge targets within images, as well as for precise water level extraction. This study focuses on the Badong Yangtze River Bridge, utilizing Unmanned Aerial Vehicle (UAV) LiDAR data of the bridge, and analyzes the multi-scattering characteristics of different bridge structural targets based on Geometric Optics (GO) methods and the Range-Doppler principle. Furthermore, the study integrates LiDAR data of the bridge's cable-stays to examine their multi-scattering phenomena, finding that the undulations of the Yangtze River's surface waves significantly contribute to the pronounced double scattering features of the bridge's cable-stays. Additionally, statistical analysis of multi-source SAR data indicates that this phenomenon is not directly correlated with radar wavelength, implying no direct connection to surface roughness. Utilizing LiDAR point cloud data from the bridge's street lamps, this paper proposes a novel method for estimating water level elevation by identifying the center position of spots formed by double scattering from lamp posts. The results show that using TerraSAR ascending and descending orbit images, this method achieves a water level elevation accuracy of approximately 0.2 meters.

UHVDC transmission line diagnosis method for integrated community energy system based on wavelet analysis

With the large-scale development of renewable energy power, China has faced with the challenges of the reverse regional distribution of wind and solar resources and power load, as well as the intermittency and randomness of renewable energy power. Therefore, China is vigorously developing ultra-high voltage direct current (UHVDC) transmission technology to solve the problem of insufficient flexibility caused by the uncertainty of renewable energy and the fluctuation of multi-energy loads in integrated community energy systems. UHVDC plays an increasingly pivotal role in the west-east transmission system in China’s power system due to its high transmission capacity and long transmission distance. Once the fault occurs in the ultra-high voltage direct (UHVD) transmission line, quick and accurate fault location identification is of great significance. Hence, this paper proposes a UHVDC transmission line diagnosis method based on wavelet analysis for integrated community energy systems. Wavelet transform (WT) is used to decompose the transient signal on a multi-scale, and then power systems computer-aided design (PSCAD) software is utilized for simulation calculation to obtain the singular spectrum entropy of each layer and facilitate wavelet transformations for signal denoising with advanced tools such as MATLAB. The prediction results can distinguish outside the rectification side fault, within the rectification side fault, and outside the inverter fault with an accuracy of 100%. A large number of simulations demonstrate that combining singular spectrum entropy with support vector machines (SVM) has emerged as a robust technique for integrated community energy systems, suggesting its potential as a standard method in UHVDC transmission line diagnosis. This study is of significant reference for realizing the complementarity of multiple types of power supply and ensuring a reliable power supply.

A Dual-Cascade Activatable Near-Infrared Fluorescent Probe for Precise Intraoperative Imaging of Tumor.

Accurate intraoperative tumor delineation is critical to achieving successful surgical outcomes. However, conventional techniques typically suffer from poor specificity and low sensitivity and are time-consuming, which greatly affects intraoperative decision-making. Here, we report a cascade activatable near-infrared fluorescent (NIRF) probe IR780SS@CaP that can sequentially respond to tumor acidity and elevated glutathione levels for accurate intraoperative tumor localization. Compared with nonactivatable and single-factor activatable probes, IR780SS@CaP with a cascade strategy can minimize nonspecific activation and false positive signals in a complicated biological environment, affording a superior tumor-to-normal tissue ratio to facilitate the delineation of abdominal metastases. Small metastatic lesions that were less than 1 mm in diameter can be precisely identified by IR780SS@CaP and completely excised under NIRF imaging guidance. This study could benefit tumor diagnosis and image-guided tumor surgery by providing real-time information and reliable decision support, thus reducing the risk of both recurrence and complications to improve patient outcomes.

3D-printed Multifunctional Guide Plate for Fenestration and Screws Drill in Proximal Femoral Benign Tumor.

The accurate fenestration, screw implantation and assisting stabilizing-plate placement in surgery of benign tumors in the proximal femur needs be defined easily. The aim of this study was to investigate the value of 3D printed multifunctional guides plate (3D-MGP) based on computer aided design. Between January 2020 and June 2022, 17 patients (nine females and eight males) with benign proximal femoral tumor had lesion curettage and allograft combined with internal plate fixation using 3D-MGP. In this study, the patients had CT scans and a technician reconstructed the 3D images of tumor and the femur, a doctor designed the location and margin of the fenestration and screws, and integrated different functions into MGP for benign proximal femoral lesions, which assisted in precise localization, fenestration and screw drilling. Musculoskeletal Tumor Society (MSTS) scoring was used to evaluate lower extremity function. Bone healing and the screws location was assessed with the radiographs. All patients underwent successful surgery with complete resection of the tumor and internal fixation with using the 3D-MGP. The mean follow-up was 16.4 months. The operative time was 126.47 ± 18.44 min, intraoperative bleeding was 198.23 ± 67.94 mL, intraoperative fluoroscopy was 6.47 ± 0.62, postoperative drainage was 223.82 ± 119.51 mL, and MSTS score was 27.29 ± 1.31 points. There were no unplanned fenestration and improper screw fixation. The 3D-MGP enabled personalized and accurate location of tumor, fenestration, screw placement and assisted stabilizing-plate placement for the treatment of benign tumor of the proximal femur. This technique has the potential to shorten operative times, decrease intraoperative bleeding, and reduce radiation exposure to patients.

FGO-MFI: factor graph optimization-based multi-sensor fusion and integration for reliable localization

Reliable and precise information pertaining to the position, velocity, and attitude is essential for automated driving. This paper proposes FGO-MFI, a cost-effective and robust multi-sensor fusion and integration localization framework that utilizes factor graph optimization. Firstly, a tightly coupled Global Navigation Satellite Systems (GNSS)/on-board sensor fusion localization framework is established to estimate vehicle states, including position, velocity, and attitude. To address the large drift rate of the Inertial Measurement Unit (IMU), this study introduces a novel IMU/Dynamics pre-integration method based on the vehicle dynamics model. We establish a two-degree-of-freedom vehicle dynamics model utilizing measurements from the wheel speed sensor and steering wheel angle sensor. The IMU/Dynamics factor is devised through a close integration of the model output and IMU pre-integration, enabling the construction of precise odometry with low-cost on-board sensors. Then, to address the issue of the non-Gaussian distribution of GNSS pseudorange error, this paper employs a Gaussian Mixture Model (GMM) to characterize the pseudorange noise, which is then applied to further sensor fusion. Given the time-varying nature of pseudorange noise, the expectation maximization algorithm is utilized to estimate GMM parameters online, leveraging the pseudorange residuals within a sliding window. Comprehensive experiments, inclusive of challenging scenarios such as urban canyons, tunnels, and wooded areas, have been carried out. They affirm the superior performance of the proposed method. Experiments have shown that our method demonstrates reliable localization across different statuses of GNSS signal, exhibiting a 39.4% improvement in the root mean square error of position error when compared to the state-of-the-art. Additionally, this FGO-MFI is a general sensor data fusion framework and is able to incorporate diverse sensor measurements, for example, from cameras and LiDARs to provide more reliable and accurate localization information.

Waveform Complexity and Positioning Analysis of Acoustic Emission Events during the Compression Failure Process of a Rock Burst Prone Sample

The localization results of acoustic emission (AE) events can reflect the location and pattern of burst-prone rock failures. However, event localization heavily depends on the quality of the original waveform of the sensor. Therefore, this study analyzed the AE waveform of a rock sample under compression to evaluate its failure localization and quality. From the research results, it could be seen that the initial failure was relatively calm, with clear take-off points, which can be better used for accurate AE event positioning. However, the later failure was severe, causing the take-off points of most sensors to be very unclear, and positioning methods that rely on take-off points cannot be used for positioning, let alone simply using the positioning results of the built-in software. This research result reminds researchers who use AE signals for event localization to first examine the quality and status of the original waveform, providing a basis for obtaining accurate localization results, in order to further accurately study the subsequent failure patterns. The above facts indicate that the initial failure is small and scattered, while the later failure is large and concentrated, with certain fractal characteristics.

Ultrasonography of the normal donkey tarsus (equus asinus)

Tarsal joint illness is a frequent source of hind limb lameness due to the complex anatomy of the region and the presence of numerous bony and soft tissue structures. Proper lameness diagnosis aims to discover the structure provoking lameness. Ultrasonography documents valuable information of soft tissues and characterizes soft tissue injuries that have heretofore been difficult to obtain either noninvasively or via radiography. The objectives of the current study were to develop and describe a standardized ultrasonographic protocol for investigation of the tarsal region in donkeys. The donkey tarsal anatomy was investigated in 5 cadavers and the tarsi of 11 healthy lameness free adult donkeys were echographically investigated. The dorsal, plantar, lateral and medial aspects of the tarsal region were substantially evaluated at four anatomical landmarks in both the longitudinal and horizontal planes using a multi-frequency 5–12 MHz linear transducer. Sonoanatomy of the extensor and flexor tarsal tendons, collateral and plantar ligaments, and synovial pouches was delineated and described. Systematic echography of the tarsal region allowed accurate localization and thorough exploration of various soft tissues of clinical interest in the donkey tarsus. Sonograms provided in this study should serve as a reference database for tarsal ultrasonography in clinical circumstances.

NLOS Identification and Mitigation for Time-based Indoor Localization Systems: Survey and Future Research Directions

One hurdle to accurate indoor localization using time-based networks is the presence of Non-Line-Of-Sight (NLOS) and multipath signals, affecting the accuracy of ranging in indoor environments. NLOS identification and mitigation have been studied over the years and applied to different time-based networks, with most works considering NLOS links with WiFi and UWB channels. In this paper, we discuss the effects and challenges of NLOS conditions on indoor localization and present current state-of-the-art approaches to NLOS identification and mitigation in literature. We survey these approaches and classify them under different categories together with their merits and demerits. We further categorize approaches to tackle NLOS effects into single and hybrid measurement-based approaches in this work. Lessons learnt from the survey with future directions are also presented in this paper.

TDOA-based UWB indoor 1D localization via weighted sliding window filtering

Accurate localization of a person or object is a significant aspect in the field of industrial internet of things (IoT). Ultra-wideband (UWB) localization system, a promising technology, has been widely studied due to its high accuracy in 2D and 3D scenarios. Moreover, in narrow corridors and deep-well environments, the 1D localization is also an important application with the advantage of resource saving, however which attracts less attention. In this paper, we focus on the problem of 1D localization and propose a practical localization model with time difference of arrivals (TDOA). To solve the nonlinear 1D localization problem, we design an algorithm that integrates the dichotomous method and Newton's iterative method, which realizes an efficient solution. Meanwhile, in order to cope with the random errors in the real scenes, we introduce a sliding window algorithm to filter out the undesirable solutions. By observing the localization error map, we design a weighted sliding window algorithm to achieve further improvement of the localization accuracy. In addition, we derive the Cramér-Rao lower bound (CRLB) for 1D localization of TDOA. Simulations and practical experiments show that our proposed algorithm greatly benefits the localization accuracy of 1D TDOA and provides strong support for its practical application.

In-line imaging and recognition of flip chip fabrication defects by real-time photoacoustic remote sensing system

Microscopic defects in flip chips, originating from manufacturing, significantly affect performance and longevity. Post-fabrication sampling methods ensure product functionality but lack in-line defect monitoring to enhance chip yield and lifespan in real-time. This study introduces a photoacoustic remote sensing (PARS) system for in-line imaging and defect recognition during flip-chip fabrication. We first propose a real-time PARS imaging method based on continuous acquisition combined with parallel processing image reconstruction to achieve real-time imaging during the scanning of flip-chip samples, reducing reconstruction time from an average of approximately 1134 ms to 38 ms. Subsequently, we propose improved YOLOv7 with space-to-depth block (IYOLOv7-SPD), an enhanced deep learning defect recognition method, for accurate in-line recognition and localization of microscopic defects during the PARS real-time imaging process. The experimental results validate the viability of the proposed system for enhancing the lifespan and yield of flip-chip products in chip manufacturing facilities.

HST-MRF: Heterogeneous Swin Transformer With Multi-Receptive Field for Medical Image Segmentation.

The Transformer has been successfully used in medical image segmentation due to its excellent long-range modeling capabilities. However, patch segmentation is necessary when building a Transformer class model. This process ignores the tissue structure features within patch, resulting in the loss of shallow representation information. In this study, we propose a Heterogeneous Swin Transformer with Multi-Receptive Field (HST-MRF) model that fuses patch information from different receptive fields to solve the problem of loss of feature information caused by patch segmentation. The heterogeneous Swin Transformer (HST) is the core module, which achieves the interaction of multi-receptive field patch information through heterogeneous attention and passes it to the next stage for progressive learning, thus complementing the patch structure information. We also designed a two-stage fusion module, multimodal bilinear pooling (MBP), to assist HST in further fusing multi-receptive field information and combining low-level and high-level semantic information for accurate localization of lesion regions. In addition, we developed adaptive patch embedding (APE) and soft channel attention (SCA) modules to retain more valuable information when acquiring patch embedding and filtering channel features, respectively, thereby improving model segmentation quality. We evaluated HST-MRF on multiple datasets for polyp, skin lesion and breast ultrasound segmentation tasks. Experimental results show that our proposed method outperforms state-of-the-art models and can achieve superior performance. Furthermore, we verified the effectiveness of each module and the benefits of multi-receptive field segmentation in reducing the loss of structural information through ablation experiments and qualitative analysis.

Tracking and co-location of global point clouds for large-area indoor environments

Extended reality (XR) experiences are on the verge of becoming widely adopted in diverse application domains. An essential part of the technology is accurate tracking and localization of the headset to create an immersive experience. A subset of the applications require perfect co-location between the real and the virtual world, where virtual objects are aligned with real-world counterparts. Current headsets support co-location for small areas, but suffer from drift when scaling up to larger ones such as buildings or factories. This paper proposes tools and solutions for this challenge by splitting up the simultaneous localization and mapping (SLAM) into separate mapping and localization stages. In the pre-processing stage, a feature map is built for the entire tracking area. A global optimizer is applied to correct the deformations caused by drift, guided by a sparse set of ground truth markers in the point cloud of a laser scan. Optionally, further refinement is applied by matching features between the ground truth keyframe images and their rendered-out SLAM estimates of the point cloud. In the second, real-time stage, the rectified feature map is used to perform localization and sensor fusion between the global tracking and the headset. The results show that the approach achieves robust co-location between the virtual and the real 3D environment for large and complex tracking environments.

Neural processing in the primary auditory cortex following cholinergic lesions of the basal forebrain in ferrets

Cortical acetylcholine (ACh) release has been linked to various cognitive functions, including perceptual learning. We have previously shown that cortical cholinergic innervation is necessary for accurate sound localization in ferrets, as well as for their ability to adapt with training to altered spatial cues. To explore whether these behavioral deficits are associated with changes in the response properties of cortical neurons, we recorded neural activity in the primary auditory cortex (A1) of anesthetized ferrets in which cholinergic inputs had been reduced by making bilateral injections of the immunotoxin ME20.4-SAP in the nucleus basalis (NB) prior to training the animals. The pattern of spontaneous activity of A1 units recorded in the ferrets with cholinergic lesions (NB ACh–) was similar to that in controls, although the proportion of burst-type units was significantly lower. Depletion of ACh also resulted in more synchronous activity in A1. No changes in thresholds, frequency tuning or in the distribution of characteristic frequencies were found in these animals. When tested with normal acoustic inputs, the spatial sensitivity of A1 neurons in the NB ACh– ferrets and the distribution of their preferred interaural level differences also closely resembled those found in control animals, indicating that these properties had not been altered by sound localization training with one ear occluded. Simulating the animals’ previous experience with a virtual earplug in one ear reduced the contralateral preference of A1 units in both groups, but caused azimuth sensitivity to change in slightly different ways, which may reflect the modest adaptation observed in the NB ACh– group. These results show that while ACh is required for behavioral adaptation to altered spatial cues, it is not required for maintenance of the spectral and spatial response properties of A1 neurons.

Abstract PO3-22-10: 3D-CT mapping with a photo-translucent application using iPad delineates tumor lesions on patient breast skin in breast-conserving surgery: Preliminary results

Abstract Objective: The aim of this study was to evaluate the feasibility and efficacy of a 3D-CT mapping technique with a photo-translucent application using an iPad for tumor localization on the patient's breast skin in breast-conserving surgery (BCS). We also compared the accuracy of tumor localization between conventional methods and 3D-CT mapping. Methods: Between February and July 2023, 20 female patients with early-stage (0 to 2) breast cancer scheduled for BCS were enrolled in this study. Preoperative enhanced CT scans were performed, and a 3D-CT image was generated by incorporating the patient's skin and manually marking the contrast-enhanced lesion using SYNAPSE VINCENT software. The 3D-CT images, adjusted to individual rotation angles to match the horizontal skin position above the tumor, were imported into an iPad equipped with a photo-translucent application. Using a securely fixed iPad holder, the translucent image was superimposed on the patient's breast, and the enhanced lesion was marked on the skin using a blue pen. Conventional tumor marking was performed based on ultrasound imaging, CT, and MRI findings using a black pen. Photographs were taken to simulate BCS based on the 3D-CT blue marking. All patients underwent BCS with resection margins of 1.5-2 cm from the black marking, and intraoperative frozen section analysis was performed to assess the resection margins. Results: The majority of patients had hormone receptor-positive breast cancers, and neoadjuvant chemotherapy was administered in three patients (Table 1). The extent of tumor involvement assessed by pathological assessment and imaging findings showed the least discrepancy with MRI, with an average error of 4.2 mm. MRI also had the lowest rate of missed tumor detection compared to other imaging modalities. The initial positive margin rate was 20%, and the simulation conducted in this study did not alter that value. Furthermore, the simulation demonstrated more effective excision of the tumor's center compared to conventional surgical methods (Table 2). Conclusion: The use of 3D-CT mapping with a photo-translucent application on an iPad offers a promising approach for accurate tumor localization in BCS. This technique provides a simplified and efficient method, and it can also be applied to cases where lesions have completely disappeared after neoadjuvant chemotherapy. XX- XX XX- XX Citation Format: Gou Watanabe, Akihiko Suzuki. 3D-CT mapping with a photo-translucent application using iPad delineates tumor lesions on patient breast skin in breast-conserving surgery: Preliminary results [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-22-10.

A Section Location Method of Single-Phase Short-Circuit Faults for Distribution Networks Containing Distributed Generators Based on Fusion Fault Confidence of Short-Circuit Current Vectors

To ensure safe and stable operation, accurate fault localization within active distribution networks is required, and this has attracted much attention. Influenced by many factors such as the control strategy, control performance, initial state of the distributed generators, and distribution network topology, it is still difficult to reliably locate complex and variable single-phase short-circuit faults relying only on a single feature quantity, while localization methods incorporating intelligent algorithms are affected by the choice of a priori samples and the fact that the solution process is a black-box model. To address this challenge, in this work, an expression for the single-phase short-circuit current vector of a distribution network containing distributed generators is derived, and the differences in magnitude and phase angle of the short-circuit current vectors upstream and downstream of the fault point are analyzed. Based on measurement theory, a fault confidence distribution function that reacts to the relative size of the current magnitude difference and phase angle difference is established, and the fusion fault confidence of the short-circuit current vector is constructed with the help of evidence theory. Finally, a method of locating single-phase short-circuit faults in distribution networks that contain distributed generators is proposed. The simulation results show that the ratio of the fusion fault confidence of the short-circuit current vector between faulted and non-faulted sections under the influence of different distributed generator capacities, fault locations, and transition resistances differ significantly. The proposed single-phase short-circuit fault localization method is both adaptive and physically interpretable and has clear boundaries, sound sensitivity, and engineering practicability.

Multi-hop graph transformer network for 3D human pose estimation

Accurate 3D human pose estimation is a challenging task due to occlusion and depth ambiguity. In this paper, we introduce a multi-hop graph transformer network designed for 2D-to-3D human pose estimation in videos by leveraging the strengths of multi-head self-attention and multi-hop graph convolutional networks with disentangled neighborhoods to capture spatio-temporal dependencies and handle long-range interactions. The proposed network architecture consists of a graph attention block composed of stacked layers of multi-head self-attention and graph convolution with learnable adjacency matrix, and a multi-hop graph convolutional block comprised of multi-hop convolutional and dilated convolutional layers. The combination of multi-head self-attention and multi-hop graph convolutional layers enables the model to capture both local and global dependencies, while the integration of dilated convolutional layers enhances the model’s ability to handle spatial details required for accurate localization of the human body joints. Extensive experiments demonstrate the effectiveness and generalization ability of our model, achieving competitive performance on benchmark datasets.

Spot scanning imaging calibration method based on deviation model for wafer inspection

The spot scanning defect detection system has good detection performance for wafer defects. However, due to the existence of system assembly errors, the wafer defect inspection system could not scan following the planned trace, which resulted in distortion of the reconstructed image and had a greater impact on defect feature extraction and accurate defect location. The system deviation calibration and adjustment can further facilitate the accuracy of the system position and image reconstruction. To address this issue, the causes and shapes of imaging distortion were analyzed and simulated, and a simple and effective calibration method based on a scanning deviation model for the spot scanning defect detection system was proposed. First, a scanning deviation model was established following the characteristics of the spot scanning system, and different degrees of reconstructed image distortion were simulated and analyzed as the system deviation changes. Then, an 8-inch wafer with a grid was used as the calibrator and scanned using the developed spot scanning system. Serveral grid corners in the reconstructed image are selected and used as the calibration feature points. The constraint function is constructed using the equidistant features between adjacent grid points, and then the quasi-Newton method is used to optimize the parameters of the function to obtain the best deviation estimation of the spot scanning system. To prove the validity of the estimated deviation parameters, another patterned wafer was scanned and the distortion image was used for reconstruction. The experiment demonstrated that an accurate and distortion-free wafer image can be reconstructed using the calibrated deviation parameters, and the root mean square error of the measured distance between two points of interest was only 0.61 pixels, which proved the effectiveness of the proposed calibration method.

Contact localization from soft tactile array sensor using tactile image

Purpose This paper aims to estimate contact location from sparse and high-dimensional soft tactile array sensor data using the tactile image. The authors used three feature extraction methods: handcrafted features, convolutional features and autoencoder features. Subsequently, these features were mapped to contact locations through a contact location regression network. Finally, the network performance was evaluated using spherical fittings of three different radii to further determine the optimal feature extraction method. Design/methodology/approach This paper aims to estimate contact location from sparse and high-dimensional soft tactile array sensor data using the tactile image. Findings This research indicates that data collected by probes can be used for contact localization. Introducing a batch normalization layer after the feature extraction stage significantly enhances the model’s generalization performance. Through qualitative and quantitative analyses, the authors conclude that convolutional methods can more accurately estimate contact locations. Originality/value The paper provides both qualitative and quantitative analyses of the performance of three contact localization methods across different datasets. To address the challenge of obtaining accurate contact locations in quantitative analysis, an indirect measurement metric is proposed.