Efficient Unsupervised Image Stitching Using Attention Mechanism with Deep Homography Estimation

Traditional feature-based image stitching techniques often encounter obstacles when dealing with images lackingunique attributes or suffering from quality degradation. The scarcity of annotated datasets in real-life scenesseverely undermines the reliability of supervised learning methods in image stitching. Furthermore, existing deeplearning architectures designed for image stitching are often too bulky to be deployed on mobile and peripheralcomputing devices. To address these challenges, this study proposes a novel unsupervised image stitching methodbased on the YOLOv8 (You Only Look Once version 8) framework that introduces deep homography networksand attentionmechanisms. Themethodology is partitioned into three distinct stages. The initial stage combines theattention mechanism with a pooling pyramid model to enhance the detection and recognition of compact objectsin images, the task of the deep homography networks module is to estimate the global homography of the inputimages consideringmultiple viewpoints. The second stage involves preliminary stitching of the masks generated inthe initial stage and further enhancement through weighted computation to eliminate common stitching artifacts.The final stage is characterized by adaptive reconstruction and careful refinement of the initial stitching results.Comprehensive experiments acrossmultiple datasets are executed tometiculously assess the proposed model. Ourmethod’s Peak Signal-to-Noise Ratio (PSNR) and Structure Similarity Index Measure (SSIM) improved by 10.6%and 6%. These experimental results confirm the efficacy and utility of the presented model in this paper.

Research on Stitching Algorithm Based on Tree Branch Image

Branch identification technology is a key technology to achieve automated pruning of fruit tree branches, and one of its technical bottlenecks lies in the stitching of branch images. To this end, we propose a set of branch image stitching technology algorithms. The algorithm is based on the grey-scale prime centroid method to determine the detection feature points, and uses the coordinate transformation matrix H of the corresponding points of the image to carry out the image geometric transformation, and realises the feature matching through sample comparison and classification methods. The experimental results show that the matched point images are more correct and less time-consuming.

Predicting the Stitch Density of Finished Fabrics Using Weft Blended Grey Knit Fabrics

Stitch density is one of the critical quality parameters of knit fabrics. This parameter is closely related to other physical quality parameters like fabric weight, fabric tightness factor, fiber types, blend ratio, yarn diameter and linear density, and fabric structure. Selecting stitch density (wales per inch, course per inch) is essential to getting the appropriate fabric weight and desired quality. Usually, no rules or assumptions exist to get the desired stitch density in the finished fabric stage. Fifteen types of blended knit fabrics were prepared to conduct the study. The varying percentages of cotton, polyester, and elastane are incorporated in the blends. Regression analysis and regression ANOVA tests were done to predict the stitch density of finished fabrics. A suitable regression equation is established to get the desired results. The study also found that the stitch density value in the finished stage fabric decreases by approximately 15% compared to the stitch density in the grey fabric stage. This study will help the fabric manufacturers get the finished fabric stitch density in advance by utilizing the grey fabric stitch density data set. The author expects this research to benefit the knitting and dyeing industry, new researchers, and advanced researchers.

Human tracking in camera network with non-overlapping FOVs

An adaptive human tracking method across spatially separated surveillance cameras with non-overlapping fields of views （FOVs） is proposed. The method relies on the two cues of the human appearance model and spatio-temporal information between cameras. For the human appearance model, an HSV color histogram is extracted from different human body parts （head, torso, and legs）, then a weighted algorithm is used to compute the similarity distance of two people. Finally, a similarity sorting algorithm with two thresholds is exploited to find the correspondence. The spatio- temporal information is established in the learning phase and is updated incrementally according to the latest correspondence. The experimental results prove that the proposed human tracking method is effective without requiring camera calibration and it becomes more accurate over time as new observations are accumulated.

A Gridless Router Based on Hierarchical PB Corner Stitching Structure

A multi layer gridless area router is reported.Based on corner stitching,this router adopts tile expansion to explore path for each net.A heuristic method that penalizes nodes deviating from the destination is devised to accelerate the algorithm.Besides,an enhanced interval tree is used to manage the intermediate data structure.In order to improve the completion rate of routing,a new gridless rip up and rerouting algorithm is proposed.The experimental results indicate that the completion rate is improved after the rip up and reroute process and the speed of this algorithm is satisfactory.

Computer aided stitching approach for dental restoration models

According to the bio-characteristics of the lower and upper cavity surfaces of dental restoration, a stitching approach is proposed based on a virtual zipper working mechanism and a minimization of the surface total curvature energy, which is used to resolve the stitching problems existing during computer-aided design for dental restorations. First, the two boundaries corresponding to the lower and upper surfaces are triangulated based on the zipper working mechanism to generate the initial stitching surface patch, of which the edges are distributed uniformly between the boundaries. Secondly, the initial stitching surface patch is subdivided and deformed to reconstruct an optimized surface patch according to the bio-characteristics of the teeth. The optimized surface patch is minimally distinguishable from the surrounding mesh in smoothness and density, and it can stitch the upper and lower cavity surfaces naturally. The experimental results show that the dental restorations obtained by the proposed method can satisfy both the shape aesthetic and the fitting accuracy, and meet the requirements of clinical oral medicine.

Adaptive topology learning of camera network across non-overlapping views

An adaptive topology learning approach is proposed to learn the topology of a practical camera network in an unsupervised way. The nodes are modeled by the Gaussian mixture model. The connectivity between nodes is judged by their cross-correlation function, which is also used to calculate their transition time distribution. The mutual information of the connected node pair is employed for transition probability calculation. A false link eliminating approach is proposed, along with a topology updating strategy to improve the learned topology. A real monitoring system with five disjoint cameras is built for experiments. Comparative results with traditional methods show that the proposed method is more accurate in topology learning and is more robust to environmental changes.

An Improved Hybrid Genetic Algorithm for Chemical Plant Layout Optimization with Novel Non-overlapping and Toxic Gas Dispersion Constraints

New approaches for facility distribution in chemical plants are proposed including an improved non-overlapping constraint based on projection relationships of facilities and a novel toxic gas dispersion constraint. In consideration of the large number of variables in the plant layout model, our new method can significantly reduce the number of variables with their own projection relationships. Also, as toxic gas dispersion is a usual incident in a chemical plant, a simple approach to describe the gas leakage is proposed, which can clearly represent the constraints of potential emission source and sitting facilities. For solving the plant layout model, an improved genetic algorithm (GA) based on infeasible solution fix technique is proposed, which improves the globe search ability of GA. The case study and experiment show that a better layout plan can be obtained with our method, and the safety factors such as gas dispersion and minimum distances can be well handled in the solution.

A morphology-stitching method to improve Landsat SLC-off images with stripes

Stripes are artifacts in satellite images caused by various factors such as hardware defects. In some cases, these artifacts are introduced by some mitigating algorithms like Landsat SLC-off （Scan Line Corrector） gap-filling methods of LLHM （Local Linear Histogram Matching） and AWLHM （Adaptive Window Linear Histogram Matching）, which leave stripes as a byproduct. To improve Landsat SLC-off images with stripes,we propose an algorithm involving some hypothetical stripe-crossing stitch lines using the mean pixel value of the stitch lines.

A novel non-overlapping approach to accurately represent 2D arbitrary particles for DEM modelling

Mechanical behaviors of granular materials are complicated and greatly influenced by the particle shape.Current,some composite approaches have been proposed for realistic particle shape modelling within discrete element method(DEM),while they cannot give a good representation to the shape and mass properties of a real particle.In this work,a novel algorithm is developed to model an arbitrary particle using a cluster of non-overlapping disks.The algorithm mainly consists of two components:boundary filling and domain filling.In the boundary filling,some disks are placed along the boundary for a precise representation of the particle shape,and some more disks are placed in the domain to give an approximation to the mass properties of the particle in the domain filling.Besides,a simple method is proposed to correct the mass properties of a cluster after domain filling and reduce the number of the disks in a cluster for lower computational load.Moreover,it is another great merit of the algorithm that a cluster generated by the algorithm can be used to simulate the particle breakage because of no overlaps between the disks in a cluster.Finally,several examples are used to show the robust performance of the algorithm.A current FORTRAN version of the algorithm is available by contacting the author.

A survey on image and video stitching

Image/video stitching is a technology for solving the field of view(FOV)limitation of images/videos.It stitches multiple overlapping images/videos to generate a wide-FOV image/video,and has been used in various fields such as sports broadcasting,video surveillance,street view,and entertainment.This survey reviews image/video stitching algorithms,with a particular focus on those developed in recent years.Image stitching first calculates the corresponding relationships between multiple overlapping images,deforms and aligns the matched images,and then blends the aligned images to generate a wide-FOV image.A seamless method is always adopted to eliminate such potential flaws as ghosting and blurring caused by parallax or objects moving across the overlapping regions.Video stitching is the further extension of image stitching.It usually stitches selected frames of original videos to generate a stitching template by performing image stitching algorithms,and the subsequent frames can then be stitched according to the template.Video stitching is more complicated with moving objects or violent camera movement,because these factors introduce jitter,shakiness,ghosting,and blurring.Foreground detection technique is usually combined into stitching to eliminate ghosting and blurring,while video stabilization algorithms are adopted to solve the jitter and shakiness.This paper further discusses panoramic stitching as a special-extension of image/video stitching.Panoramic stitching is currently the most widely used application in stitching.This survey reviews the latest image/video stitching methods,and introduces the fundamental principles/advantages/weaknesses of image/video stitching algorithms.Image/video stitching faces long-term challenges such as wide baseline,large parallax,and low-texture problem in the overlapping region.New technologies may present new opportunities to address these issues,such as deep learning-based semantic correspondence,and 3D image stitching.Finally,this survey discusses the challenges of image/video stitching and proposes potential solutions.

Efficient Image Stitching in the Presence of Dynamic Objects and Structure Misalignment

This paper presents a new method for simultaneously eliminating visual artifacts caused by moving objects and structure misalignment in image stitching. Given that the input images are roughly aligned, our approach is implemented in two stages. In the first stage, we discover motions between input images, and then extract their corresponding regions through a multi-seed based region growing algorithm. In the second stage, with prior information provided by the extracted regions, we perform a graph cut optimization in gradient-domain to determine which pixels to use from each image to achieve seamless stitching. Our method is simple to implement and effective. The experimental results illustrate that the proposed approach can produce comparable or superior results in comparison with state-of-the-art methods.

Underwater Terrain Image Stitching Based on Spatial Gradient Feature Block

At present,underwater terrain images are all strip-shaped small fragment images preprocessed by the side-scan sonar imaging system.However,the processed underwater terrain images have inconspicuous and few feature points.In order to better realize the stitching of underwater terrain images and solve the problems of slow traditional image stitching speed,we proposed an improved algorithm for underwater terrain image stitching based on spatial gradient feature block.First,the spatial gradient fuzzy C-Means algorithm is used to divide the underwater terrain image into feature blocks with the fusion of spatial gradient information.The accelerated-KAZE(AKAZE)algorithm is used to combine the feature block information to match the reference image and the target image.Then,the random sample consensus(RANSAC)is applied to optimize the matching results.Finally,image fusion is performed with the global homography and the optimal seam-line method to improve the accuracy of image overlay fusion.The experimental results show that the proposed method in this paper effectively divides images into feature blocks by combining spatial information and gradient information,which not only solves the problem of stitching failure of underwater terrain images due to unobvious features,and further reduces the sensitivity to noise,but also effectively reduces the iterative calculation in the feature point matching process of the traditional method,and improves the stitching speed.Ghosting and shape warping are significantly eliminated by re-optimizing the overlap of the image.

DYNAMICS OF THE LOOPER DRIVE MECHANISM ON A BLIND STITCHING MACHINE

The looper drive mechanism is a main moving part in the blind stitching machine, which is aspatial 5 bar RRRSR linkage. In this paper, a dynamic analysis of the looper drive mechanism is made by means of the ma-trix method. Two methods are adopted in the calculation of the shaking force and shaking moment, one isdone by the constraint reaction of the flame-connected kinematic parts; the other is the inertialforces of all moving links.

Boundary Element Method with Non-overlapping Domain Decomposition for Diffusion Equation

A boundary element method based on non-overlapping domain decomposition method to solve the time-dependent diffusion equations is presented. The time-dependent fundamental solution is used in the formulation of boundary integrals and the time integration process always restarts from the initial time condition. The process of replacing the interface values, which needs a summation of boundary integrals related to the boundary values at previous time steps can be treated in parallel iterative procedure. Numerical experiments demonstrate that the implementation of the present algorithm is efficient.

Parallel Implementation of the Non-Overlapping Template Matching Test Using CUDA

NIST(National Institute of Standards and Technology) statistical test recognized as the most authoritative is widely used in verifying the randomness of binary sequences. The Non-overlapping Template Matching Test as the 7 th test of the NIST Test Suit is remarkably time consuming and the slow performance is one of the major hurdles in the testing process. In this paper, we present an efficient bit-parallel matching algorithm and segmented scan-based strategy for execution on Graphics Processing Unit(GPU) using NVIDIA Compute Unified Device Architecture(CUDA). Experimental results show the significant performance improvement of the parallelized Non-overlapping Template Matching Test, the running speed is 483 times faster than the original NIST implementation without attenuating the test result accuracy.

Adaptive Human Tracking Across Non-overlapping Cameras in Depression Angles

To track human across non-overlapping cameras in depression angles for applications such as multi-airplane visual human tracking and urban multi-camera surveillance,an adaptive human tracking method is proposed,focusing on both feature representation and human tracking mechanism.Feature representation describes individual by using both improved local appearance descriptors and statistical geometric parameters.The improved feature descriptors can be extracted quickly and make the human feature more discriminative.Adaptive human tracking mechanism is based on feature representation and it arranges the human image blobs in field of view into matrix.Primary appearance models are created to include the maximum inter-camera appearance information captured from different visual angles.The persons appeared in camera are first filtered by statistical geometric parameters.Then the one among the filtered persons who has the maximum matching scale with the primary models is determined to be the target person.Subsequently,the image blobs of the target person are used to update and generate new primary appearance models for the next camera,thus being robust to visual angle changes.Experimental results prove the excellence of the feature representation and show the good generalization capability of tracking mechanism as well as its robustness to condition variables.

针刺/缝合多尺度联锁复合材料Ⅰ型层间力学行为

为明晰缝合工艺对针刺结构复合材料Ⅰ型层间力学性能的影响,以石英机织布和石英纱线为原料,设计制备了针刺/缝合多尺度联锁织物及复合材料,采用Micro-CT对多尺度联锁织物结构进行表征,进一步研究了多尺度联锁复合材料Ⅰ型层间力学行为,同时建立了Ⅰ型层间断裂行为有限元分析模型,阐明了多尺度联锁复合材料的层间强化机制。研究结果表明:相比针刺复合材料,多尺度联锁复合材料的层间载荷值最大提高46.61%,临界能量释放率最大提高55.55%;在缝合矩阵不变的情况下,单束缝合纱线从100 tex增大到200 tex,最大破坏载荷提高了12.91%,临界能量释放率提高了17.8%;随着缝合矩阵增大,总植入量从800 tex增大到1600 tex后,最大破坏载荷提高了22.8%,临界能量释放率提高了47.3%;此外,多尺度联锁复合材料有限元模型的Ⅰ型层间断裂模拟结果与实验结果相吻合,最大误差仅为3.1%,建立的有限元模型可较为准确地预测多尺度联锁复合材料Ⅰ型层间失效行为。

基于双反射镜的2D-DIC变形测量系统开发

二维数字图像相关(two-dimensional digital image correlation,2D-DIC)在测量过程中不可避免地会出现相机光轴与测量表面非垂直,由此产生的离面位移而将导致较大的测量误差,同时在视场受限的环境中难以通过单台相机完成大范围的变形测量。有鉴于此,该文开发了基于双反射镜的2D-DIC变形测量系统,使用双反射镜成像缓解离面运动对2D-DIC的影响,通过可移动相机实现小视场下的图像采集,提出基于频域移位的高精度图像拼接方法,并改进了融合函数,最终获得试样的高分辨率图像。单轴拉伸实验结果表明,轴向应变的平均相对误差相比传统2D-DIC方法降低12.82%,测量分辨率提高约34.92%,验证了测量系统的可行性和有效性。

基于PSD的分离式长导轨直线度测量方法研究

利用分段拼接测量方法能够将PSD激光准直测量系统测量范围扩大,相较于传统长导轨直线度测量方法,该方法可同时适用于连续型导轨和分离式超长导轨的直线度测量。首先,在(-5~+5) mm测量范围内,通过激光干涉仪分别测得激光准直测量系统的接收靶在水平和竖直方向上的位移,误差均优于±(1μm+1%H)。然后,在40 m范围内与激光准直测量系统的直线度测量精度进行对比,二者水平方向直线度误差相差0.06 mm,竖直方向直线度误差相差0.13 mm。最后,在70 m分离式超长导轨上,测得水平方向直线度为0.50 mm,竖直方向直线度为0.53 mm。该方法可迅速定位和调整直线度误差极值点位置,能够较为有效地解决分离式超长导轨的直线度装配调试问题。