Segmentation Matching Research Articles

Specular highlight removal by federated generative adversarial network with attention mechanism

Specular highlight removal ensures the acquisition of high-quality images, which finds its important applications in stereo matching, text recognition and image segmentation. In order to prevent the leakage of images containing personal information, such as identification card (ID) photos, clients often train specular highlight removal models using local data resulting in a lack of precision and generalization of the trained model. To address this challenge, this paper introduces a new method to remove highlight in images using federated learning (FL) and attention generative adversarial network (AttGAN). Specifically, the former builds a global model in the central server and updates the global model by aggregating model parameters of clients. This process does not involve the transmission of image data, which enhances the privacy of clients; the later combining attention mechanisms and generative adversarial network aims to improve the quality of highlight removal by focusing on key image regions, resulting in more realistic and visually pleasing results. The proposed FL-AttGAN method is numerically evaluated, using SD1, SD2 amd RD datasets. The results show that the proposed FL-AttGAN outperforms existent methods.

Initial point positioning of weld seam and robot pose estimation based on binocular vision

Abstract In order to rapidly identify and locate the weld seam initial point in robotic automated welding, we established a binocular vision system and proposed a weld seam initial point localization algorithm named WIPL-Net. Built upon the Fully Convolutional One-Stage object detection network, WIPL-Net introduces a lightweight ResNext as its backbone network and incorporates channel attention and enhanced feature fusion mechanisms to enhance feature detection and extraction capabilities. Subsequently, WIPL-Net is utilized to obtain the weld seam’s initial point, and its three-dimensional coordinates are determined through trigonometric measurements. To further estimate the robot’s posture at the initial point, we performed sparse three-dimensional reconstruction of the local region centered on the weld seam initial point based on You Only Look At Coefficients of Tensors instance segmentation and feature point matching. Finally, we conducted comparative experiments on WIPL-Net and conducted weld seam initial point localization experiments in real welding scenarios. The results demonstrate that our proposed method achieves a positioning error of less than 1.2 mm for the weld seam’s initial point and a pose error of less than 10 degrees for the robot, meeting the requirement for real-time positioning of the weld seam’s initial point.

Research on intelligent auxiliary assembly technology based on deep learning

Background Auxiliary assembly refers to guiding and prompting the assembly process to help operators complete complex assembly operations. Due to the complex structure of products, the similar shape of parts and human factors, the misassembly and missing assembly of parts still occur in the process of product assembly, so it is of great significance to detect the assembly correctness of complex products. Methods Aiming at the problem that manual inspection is inefficient and depends heavily on the level of inspectors in the process of complex product assembly inspection, this paper proposes an assembly correctness detection method based on deep learning. Through the three steps of view transformation, semantic segmentation and template matching, the automatic judgment of assembly errors such as wrong assembly, missing assembly and redundancy is realized, and the method is verified by the computer motherboard. Results Taking the computer motherboard as the verification object to test the correctness of assembly, the experimental re sults show that the perspective adjustment of the image after homography transformation is very obvious. The evaluation index of the semantic segmentation network detection object is calculated, and each accuracy meets the requirements of assembly correctness detection. A visualization module is also used to visually display the results of assembly correctness detection based on template matching. Conclusions The assembly correctness detection method can provide a guarantee for the manual assembly process and reduce the error rate of assembly. The machine vision detection technology can be used for automatic detection of assembly quality to improve the efficiency and automation level of detection.

Estimation of Oil Palm Tree Quantity Using Oil Palm Ecognition and SAGA GIS Based on UAV Imagery (Case Study: Dumai, Riau)

Abstract Oil palm is a plantation commodity in Indonesia that drives the growth of the national economy. The growth of oil palm is rapidly expanding every year, thus requiring efficient land monitoring strategies. Tree counting is one technique used to monitor current conditions in vast plantation areas. Utilizing UAV technology in agricultural mapping is the best option as a tool for capturing high-resolution aerial photos. The objective of this research is to automatically count the number of oil palm trees by comparing two methods and algorithms. GEOBIA (Geography-Based Image Analysis) and Deep Learning classification are methods used for automatic object counting. In this study, high-resolution aerial photo data was utilized, making object detection much more feasible. The research applied the Template Matching and Watershed Segmentation algorithm models in a sample area of 32.5 hectares divided into two blocks: Unknown-2 and E-1. It can be concluded that both methods, with their respective algorithm models, are sufficiently relevant for use in the automatic counting process of oil palm trees. This is evidenced by the accuracy test error values, which are not more than 15%, indicating that the counted oil palm tree results can be used for further data analysis.

A Review on India Traffic Sign Detection Techniques

This abstract delves into the realm of traffic sign detection techniques tailored for India's diverse and dynamic traffic environment. It navigates through traditional methods like color segmentation and template matching, juxtaposing them with the contemporary prowess of deep learning, particularly convolutional neural networks (CNNs). The intricacies of Indian roads, encompassing varied signage designs, fluctuating lighting conditions, and complex infrastructural nuances, are scrutinized in the context of these detection mechanisms. The narrative extends to discuss the amalgamation of IoT devices, real-time processing frameworks, and vehicle-mounted cameras to forge more efficient detection systems. Furthermore, the review underscores the transformative impact of machine learning advancements, spotlighting transfer learning and ensemble techniques as instrumental in augmenting detection accuracy and scalability. This abstract encapsulates a comprehensive exploration of India's traffic sign detection landscape, offering insights into ongoing trends, persistent challenges, and promising avenues for future research and development.

Multi-scale network with attention mechanism for underwater image enhancement

Underwater images suffer from severe degradation due to the complicated environments. Though numerous approaches have been proposed, accurately correcting color bias meanwhile effectively enhancing contrast is still a difficult problem. In order to address this issue, a dedicated designed Multi-scale Network with Attention mechanism (MNA) is introduced in this work. Concretely, MNA contains four key characteristics: (a) setting more convolution layers in shallow flows, (b) letting connections from high-level to adjacent low-level stream progressively, (c) simplifying dual attention mechanism embedding it in conventional residual block, (d) exploiting channel attention module to fuse multi-scale information rather than conventional summation operation. Extensive experiments demonstrate that our MNA achieves better performance than some well-recognized technologies. Meanwhile, ablation study proves the effectiveness of each component in our MNA. In addition, extended applications demonstrate the improvement of our MNA in local feature points matching and image segmentation.

Two-stage correction prediction of wind power based on numerical weather prediction wind speed superposition correction and improved clustering

The unavoidable wind power prediction error poses a challenge for wind power to participate in day-ahead scheduling plan. This paper proposes a two-stage correction method considering NWP wind speed and power correction. Firstly, this paper fully considers the historical similarity of NWP wind speed, and adopts the improved clustering distance for segment matching and wind speed correction based on the extracted most relevant trend component, which enhances the utilization of NWP information and avoids the randomness and uncertainty of the intelligent correction method, and gram angular field (GAF) and stacked autoencoder (SAE) are used for feature extraction and conditional variational autoencoder (CVA) is used to generate specific samples to ensure the adequacy of the clustering process. Secondly, considering the historical similarity of error of predicted power, the power is further corrected based on the proposed weighted double-constraint to realize the secondary calibration. The Proposed method is applied to several wind farms in China to verify its effectiveness. The results show that the proposed method reduces the NRMSE by 3.82 % and NMAE by 3.40 % compared with direct prediction in the wind farms in western Inner Mongolia, which is important for promoting wind power consumption and maintaining the safe and stable operation of the power system.

Ozone exposure limits cardiorespiratory function during maximal cycling exercise in endurance athletes.

Ground-level ozone (O3) is a potent air pollutant well recognized to acutely induce adverse respiratory symptoms and impairments in pulmonary function. However, it is unclear how the hyperpnea of exercise may modulate these effects, and the subsequent consequences on exercise performance. We tested the hypothesis that pulmonary function and exercise capability would be diminished, and symptom development would be increased during peak real-world levels of O3 exposure compared with room air. Twenty aerobically trained participants [13 M, 7 F; maximal O2 uptake (V̇o2max), 64.1 ± 7.0 mL·kg-1·min-1] completed a three-visit double-blinded, randomized crossover trial. Following a screening visit, participants were exposed to 170 ppb O3 or room air (<10 ppb O3) on separate visits during exercise trials, consisting of a 25-min moderate-intensity warmup, 30-min heavy-intensity bout, and a subsequent time-to-exhaustion (TTE) performance test. No differences in O2 uptake or ventilation were observed during submaximal exercise between conditions. During the TTE test, we observed significantly lower end-exercise O2 uptake (-3.2 ± 4.3%, P = 0.004), minute ventilation (-3.2 ± 6.5%, P = 0.043), tidal volume (-3.6 ± 5.1%, P = 0.008), and a trend toward lower exercise duration in O3 compared with room air (-10.8 ± 26.5%, P = 0.092). As decreases in O2 uptake and alterations in respiratory pattern were also present at matched time segments between conditions, a limitation of oxygen transport seems likely during maximal exercise. A more comprehensive understanding of the direct mechanisms that limit oxygen transport during exercise in high-pollutant concentrations is key for mitigating performance changes.NEW & NOTEWORTHY We demonstrate that in highly trained endurance athletes, exposure to peak real-world levels of O3 air pollution (170 ppb) significantly diminishes O2 uptake along with corresponding changes in ventilation during maximal exercise. As no differences were observed during extended submaximal exercise, a combined effect of effective dose of pollution and exercise intensity on severity of responses seems likely.

Multiparametric prenatal imaging characterization of fetal brain edema in Chiari II malformation might help to select candidates for fetal surgery

ObjectiveTo identify brain edema in fetuses with Chiari II malformation using a multiparametric approach including structural T2-weighted, diffusion tensor imaging (DTI) metrics, and MRI-based radiomics.MethodsA single-center retrospective review of MRI scans obtained in fetuses with Chiari II was performed. Brain edema cases were radiologically identified using the following MR criteria: brain parenchymal T2 prolongation, blurring of lamination, and effacement of external CSF spaces. Fractional anisotropy (FA) values were calculated from regions of interest (ROI), including hemispheric parenchyma, internal capsule, and corticospinal tract, and compared group-wise. After 1:1 age matching and manual single-slice 2D segmentation of the fetal brain parenchyma using ITK-Snap, radiomics features were extracted using pyradiomics. Areas under the curve (AUCs) of the features regarding discriminating subgroups were calculated.ResultsNinety-one fetuses with Chiari II underwent a total of 101 MRI scans at a median gestational age of 24.4 weeks and were included. Fifty scans were visually classified as Chiari II with brain edema group and showed significantly reduced external CSF spaces compared to the nonedema group (9.8 vs. 18.3 mm, p < 0.001). FA values of all used ROIs were elevated in the edema group (p < 0.001 for all ROIs). The 10 most important radiomics features showed an AUC of 0.81 (95%CI: 0.71, 0.91) for discriminating between Chiari II fetuses with and without edema.ConclusionsBrain edema in fetuses with Chiari II is common and radiologically detectable on T2-weighted fetal MRI sequences, and DTI-based FA values and radiomics features provide further evidence of microstructure differences between subgroups with and without edema.Clinical relevance statementA more severe phenotype of fetuses with Chiari II malformation is characterized by prenatal brain edema and more postnatal clinical morbidity and disability. Fetal brain edema is a promising prenatal MR imaging biomarker candidate for optimizing the risk-benefit evaluation of selection for fetal surgery.Key PointsBrain edema of fetuses prenatally diagnosed with Chiari II malformation is a common, so far unknown, association.DTI metrics and radiomics confirm microstructural differences between the brains of Chiari II fetuses with and without edema.Fetal brain edema may explain worse motor outcomes in this Chiari II subgroup, who may substantially benefit from fetal surgery.Graphical

The One-Point-One-Line geometry for robust and efficient line segment correspondence

Three-dimensional (3D) lines are common elements in artificial scenes and serve as basic, yet essential features for structural 3D reconstruction. The crucial step of 3D line reconstruction, namely two-view line segment matching, still faces challenges in terms of both accuracy and efficiency improvements. Therefore, robust and efficient constraints are needed to establish valid line candidates. This paper introduces a novel geometry constraint called “one-point-one-line geometry” (OPOL) to enhance the precision of line matching and reduce computational complexity. OPOL offers two remarkable advantages: (1) It takes point orientations as the constraint, which is not only invariant to projective transformations, but also alleviates computational requirements. (2) It needs only one point match to construct the geometry constraint, thus both the grouping and validation are greatly reduced. Additionally, we incorporate the line sweep strategy into OPOL, leveraging depth and space constraints derived from existing 3D points to further enhance efficiency. Extensive experiments conducted on large-coverage and high-resolution images (as large as 10336 × 7788 pixels) demonstrated that OPOL matched lines within a second for an image pair. Both quantitative and qualitative results also demonstrated the superior accuracy and efficiency performance of OPOL. We integrated OPOL into multiple view line reconstruction frameworks, and the promising experimental results reveal the performance of OPOL for robust line reconstruction. The OPOL code is publicly available at https://github.com/JoeAlexxxxx/OPOL.

Accurate semantic segmentation of small-body craters for navigation

Feature extraction and matching play integral roles in autonomous vision navigation. As a typical morphological feature, craters can be used as navigation landmarks. However, the spin of small bodies introduces substantial variations in illumination and viewing angles within acquired images, thereby posing challenges to feature extraction and matching algorithms. Addressing these challenges, a deep learning-based approach emerges as pivotal in mitigating the limitations of existing navigation algorithms. This paper presents an algorithm designed to segment noteworthy craters and irregular regions on such small bodies, thereby furnishing crucial autonomous navigation data for missions involving uncharted terrains. By seamlessly integrating deep learning techniques with conventional computer vision-based feature description methods, the algorithm achieves matching segmentation results of noteworthy precision. The experimental results show that the algorithm has better results compared to the existing mainstream segmentation networks. And it can achieve the accurate matching of segmentation results.

Real-Time Cucumber Target Recognition in Greenhouse Environments Using Color Segmentation and Shape Matching

Accurate identification of fruits in greenhouse environments is an essential need for the precise functioning of agricultural robots. This study presents a solution to the problem of distinguishing cucumber fruits from their stems and leaves, which often have similar colors in their natural environment. The proposed algorithm for cucumber fruit identification relies on color segmentation and form matching. First, we get the boundary details from the acquired image of the cucumber sample. The edge information is described and reconstructed by utilizing a shape descriptor known as the Fourier descriptor in order to acquire a matching template image. Subsequently, we generate a multi-scale template by amalgamating computational and real-world data. The target image is subjected to color conditioning in order to enhance the segmenacktation of the target region inside the HSV color space. Then, the segmented target region is compared to the multi-scale template based on its shape. The method of color segmentation decreases the presence of unwanted information in the target image, hence improving the effectiveness of shape matching. An analysis was performed on a set of 200 cucumber photos that were obtained from the field. The findings indicate that the method presented in this study surpasses conventional recognition algorithms in terms of accuracy and efficiency, with a recognition rate of up to 86%. Moreover, the system has exceptional proficiency in identifying cucumber targets within greenhouses. This attribute renders it a great resource for offering technical assistance to agricultural robots that operate with accuracy.

Research on intelligent auxiliary assembly technology based on deep learning

Background Auxiliary assembly refers to guiding and prompting the assembly process to help operators complete complex assembly operations. Due to the complex structure of products, the similar shape of parts and human factors, the misassembly and missing assembly of parts still occur in the process of product assembly, so it is of great significance to detect the assembly correctness of complex products. Methods Aiming at the problem that manual inspection is inefficient and depends heavily on the level of inspectors in the process of complex product assembly inspection, this paper proposes an assembly correctness detection method based on deep learning. Through the three steps of view transformation, semantic segmentation and template matching, the automatic judgment of assembly errors such as wrong assembly, missing assembly and redundancy is realized, and the method is verified by the computer motherboard. Results Taking the computer motherboard as the verification object to test the correctness of assembly, the experimental re sults show that the perspective adjustment of the image after homography transformation is very obvious. The evaluation index of the semantic segmentation network detection object is calculated, and each accuracy meets the requirements of assembly correctness detection. A visualization module is also used to visually display the results of assembly correctness detection based on template matching. Conclusions The assembly correctness detection method can provide a guarantee for the manual assembly process and reduce the error rate of assembly. The machine vision detection technology can be used for automatic detection of assembly quality to improve the efficiency and automation level of detection.

S$^{3}$M-Net: Joint Learning of Semantic Segmentation and Stereo Matching for Autonomous Driving

S$^{3}$M-Net: Joint Learning of Semantic Segmentation and Stereo Matching for Autonomous Driving

Retracted: Music Choreography Algorithm Based on Feature Matching and Fragment Segmentation.

[This retracts the article DOI: 10.1155/2022/9274098.].

Graph neural network-based identification of ditch matching patterns across multi-scale geospatial data

Ditches are vital to water system data. To ease the task of matching multi-scale ditch data and enhance accuracy (Acc), it is essential to discern ditch data matching patterns. Despite its importance, limited research has been conducted on ditch matching patterns, and the primary reasons for variations in multi-scale geographical entities’ matching patterns remain underexplored. Here, we introduce a supervised graph neural network (GNN) method, targeting the direct analysis of 1:10,000 ditch characteristics to identify the matching patterns between 1:10,000 and 1:25,000 datasets. The ditch network is depicted as a graph structure, with nodes representing ditch segments and edges indicating their connections. Subsequently, each ditch segment’s geometric, semantic, and topological attributes are computed as node attributes, and their matching patterns with the 1:25,000 dataset are labeled as node annotations. Ditch segment matching patterns are then categorized using supervised learning. Experiments using Overijssel, Netherlands’ ditch data reveal that this method achieves a 97.3% classification Acc, outperforming other GNN methods by 25.6–26.3% and traditional machine learning methods by 33%. These findings underscore the efficacy and superiority of the proposed supervised GNN approach in pinpointing ditch matching patterns.

Convolutional neural network for substantiation of children's books, intelligent interactive communication and application analysis of voice question answering

At present, there are many children accompanying robots. A key indicator to measure their performance is convenient and efficient human–computer interaction. The implementation of voice question-answering technology needs to solve three problems: voice recognition, the establishment of a knowledge base, and answer matching. As the front-end data entry, speech recognition accuracy is directly related to the answer-matching effect of the back-end question-answering system. Based on convolutional neural network and intelligent communication technology, this paper analyzes the characteristics of children's reading robot and designs and develops intelligent voice interaction function on the Android platform. The focus of this paper is on Chinese word segmentation and keyword extraction and matching in the interaction function. Further, we designed a simple question-and-answer database for the children's reading robot. In this paper, convolutional neural network connection timing is used as the feature parameter. Through a large number of experiments, the structure of the CNN network is constantly adjusted to minimize the word error rate, optimize the model recognition rate, and better identify users' voice problems.

An ISAR Image Component Recognition Method Based on Semantic Segmentation and Mask Matching.

The inverse synthetic aperture radar (ISAR) image is a kind of target feature data acquired by radar for moving targets, which can reflect the shape, structure, and motion information of the target, and has attracted a great deal of attention from the radar automatic target recognition (RATR) community. The identification of ISAR image components in radar satellite identification missions has not been carried out in related research, and the relevant segmentation methods of optical images applied to the research of semantic segmentation of ISAR images do not achieve ideal segmentation results. To address this problem, this paper proposes an ISAR image part recognition method based on semantic segmentation and mask matching. Furthermore, a reliable automatic ISAR image component labeling method is designed, and the satellite target component labeling ISAR image samples are obtained accurately and efficiently, and the satellite target component labeling ISAR image data set is obtained. On this basis, an ISAR image component recognition method based on semantic segmentation and mask matching is proposed in this paper. U-Net and Siamese Network are designed to complete the ISAR image binary semantic segmentation and binary mask matching, respectively. The component label of the ISAR image is predicted by the mask matching results. Experiments based on satellite component labeling ISAR image datasets confirm that the proposed method is feasible and effective, and it has greater comparative advantages compared to other classical semantic segmentation networks.

VDVM: An automatic vertebrae detection and vertebral segment matching framework for C-arm X-ray image identification.

C-arm fluoroscopy, as an effective diagnosis and treatment method for spine surgery, can help doctors perform surgery procedures more precisely. In clinical surgery, the surgeon often determines the specific surgical location by comparing C-arm X-ray images with digital radiography (DR) images. However, this heavily relies on the doctor's experience. In this study, we design a framework for automatic vertebrae detection as well as vertebral segment matching (VDVM) for the identification of vertebrae in C-arm X-ray images. The proposed VDVM framework is mainly divided into two parts: vertebra detection and vertebra matching. In the first part, a data preprocessing method is used to improve the image quality of C-arm X-ray images and DR images. The YOLOv3 model is then used to detect the vertebrae, and the vertebral regions are extracted based on their position. In the second part, the Mobile-Unet model is first used to segment the vertebrae contour of the C-arm X-ray image and DR image based on vertebral regions respectively. The inclination angle of the contour is then calculated using the minimum bounding rectangle and corrected accordingly. Finally, a multi-vertebra strategy is applied to measure the visual information fidelity for the vertebral region, and the vertebrae are matched based on the measured results. We use 382 C-arm X-ray images and 203 full length X-ray images to train the vertebra detection model, and achieve a mAP of 0.87 in the test dataset of 31 C-arm X-ray images and 0.96 in the test dataset of 31 lumbar DR images. Finally, we achieve a vertebral segment matching accuracy of 0.733 on 31 C-arm X-ray images. A VDVM framework is proposed, which performs well for the detection of vertebrae and achieves good results in vertebral segment matching.

An Information-Theoretic Method to Automatic Shortcut Avoidance and Domain Generalization for Dense Prediction Tasks.

Deep convolutional neural networks for dense prediction tasks are commonly optimized using synthetic data, as generating pixel-wise annotations for real-world data is laborious. However, the synthetically trained models do not generalize well to real-world environments. This poor "synthetic to real" (S2R) generalization we address through the lens of shortcut learning. We demonstrate that the learning of feature representations in deep convolutional networks is heavily influenced by synthetic data artifacts (shortcut attributes). To mitigate this issue, we propose an Information-Theoretic Shortcut Avoidance(ITSA) approach to automatically restrict shortcut-related information from being encoded into the feature representations. Specifically, our proposed method minimizes the sensitivity of latent features to input variations: to regularize the learning of robust and shortcut-invariant features in synthetically trained models. To avoid the prohibitive computational cost of direct input sensitivity optimization, we propose a practical yet feasible algorithm to achieve robustness. Our results show that the proposed method can effectively improve S2R generalization in multiple distinct dense prediction tasks, such as stereo matching, optical flow, and semantic segmentation. Importantly, the proposed method enhances the robustness of the synthetically trained networks and outperforms their fine-tuned counterparts (on real data) for challenging out-of-domain applications.