基于SIFT的甘肃庆阳刺绣针法检测增强算法研究

甘肃庆阳刺绣源于黄河流域,制品式样繁复种类多样,是陇东地区特色文化的重要载体。对庆阳刺绣的数字化保护可以作为中华文化传承和活化保护的有力手段。由于刺绣图案和针法具有较为复杂的边缘结构和重复率较高的纹理特征,因此对其进行区域检测是一项颇具挑战性的任务。文章选取5种庆阳刺绣中经典针法并建立用于学习训练的图像数据集,通过导向滤波搭建改进检测网络模型SIFT-E,对针法图案进行实时检测。与SIFT相比,改进后的方法识别匹配率更高,极大地提升了检测效率。

基于多尺度Scale-Unet的单样本图像翻译

随着生成对抗网络(GAN)的发展,基于单样本的无监督图像到图像翻译(UI2I)取得了重大进展。然而,以前方法无法捕获图像中的复杂纹理并保留原始内容信息。为解决这个问题,提出了一种基于尺度可变U-Net结构(Scale—Unet)的新型单样本图像翻译结构SUGAN。所提出的SUGAN使用Scale—Unet作为生成器,利用多尺度结构和渐进方法不断改进网络结构,以从粗到细地学习图像特征。同时,提出了尺度像素损失scale-pixel来更好地约束保留原始内容信息,防止信息丢失。实验表明,与SinGAN、TuiGAN、TSIT、StyTR2等公共数据集Summer■Winter、Horse■Zebra上的方法相比,该方法生成图像的SIFID值平均降低了30%。所提方法可更好地保留图像内容信息,同时生成详细逼真的高质量图像。

具有局部自适应阈值的SIFT快速图像匹配算法

针对传统SIFT匹配算法复杂、特征冗余点多、难以满足实时性等问题,本文提出了一种具有局部自适应阈值的SIFT快速图像匹配算法。首先,所提方法在SIFT算法的基础上,对构建的高斯金字塔进行了优化,通过减少金字塔层数来消除冗余特征点以提高检测效率,并根据图像局部对比度来自适应提取FAST算法中的阈值从而实现高质量的特征点检测,筛选出鲁棒性较强的特征点进行更准确的匹配;其次,采用高斯圆形窗口建立32维降维特征向量,提高算法运行效率;最后,根据匹配特征点对之间的几何一致性对特征点进行提纯,有效减少误匹配。实验结果表明,本文方法在匹配精度和运算效率方面的综合表现均优于SIFT算法及其他对比匹配算法,相比传统的SIFT算法,匹配精度提高了约10%,算法运行时间缩短了约49%。在图像发生尺度、旋转以及光照变化的情况下,正确匹配率在93%以上。

基于多特征融合和改进SIFT的目标跟踪算法

在进行跟踪任务时,当运动目标发生形变或旋转以及受到光照变化或背景干扰时,会发生偏移现象或丢失跟踪目标,从而导致跟踪精度降低。据此,提出基于多特征融合和改进SIFT的目标跟踪算法。在图像的高熵部分进行特征点的提取,并使用哈希算法将错误匹配的特征点剔除。同时对感知哈希和差异哈希进行改进,将改进后的图像哈希特征、颜色特征和SIFT特征进行融合并应用于跟踪算法。将算法在OTB-100数据集上进行实验,成功率达到了94.3%。

基于融合SIFT算法的光伏电站安监图像精准识别技术研究

随着我国“双碳”工作的不断推进,光伏电站的规模逐渐扩大。为解决光伏电站图像安防技术中存在的拍摄角度、光线影响大,识别准确率低的问题,提出了一种基于SIFT的光伏电站图像安防技术。通过SIFT算法提取光伏电站图像特征点,解决了光线、角度带来的识别误差问题;通过卷积神经网络方法准确的识别了光伏电站车辆、人员、动物、积雪等安防问题。该方法在某光伏电站进行应用,其光伏电站图像安防识别准确率平均为99.7%。所提方法能有效提高光伏电站图像安防识别准确率。

基于改进SIFT特征和极坐标的图像匹配定位研究

为解决图像配准过程中因图像旋转及缩放导致匹配误差较大的问题,提出了一种基于对数极坐标和改进的SIFT相结合的匹配算法。首先,通过对数极坐标变换方法进行图像粗匹配,实现对待匹配图像旋转及尺度的估计;其次,在粗匹配的基础上,利用信息熵原理提取子块的特征描述符;最后,利用欧氏距离进行匹配度量,完成图像配准,利用图像变换模型实现两幅图像中相同区域目标的匹配定位。实验结果表明,该算法能够实现两幅图像的高精度匹配与目标定位。

基于改进SIFT算法的图像识别研究

针对基于SIFT算法的图像识别中存在着识别准确度不高,匹配速度慢问题。提出了一种快速获取特征点和匹配特征点方法。首先基于图像边缘的特征点获取方法,对图像边缘进行扩展,只在扩展区域内判断特征点。然后在特征点描述符中增加特征点周围16个子区域归一化后的灰度差分值信息,增强特征点描述能力。通过KPCA降维方法对描述符降维,最后在特征点匹配阶段使用余弦相似度预处理噪声特征点,加快匹配速度。在实际测试中,相对于SIFT算法图像识别率提升了8%左右,识别速度提高了50%左右。

基于SIFT和INIR的地表未爆子弹药检测方法研究

采用基于SIFT(尺度不变特征变换)算法的可见光同源图像配准检测方法解决地表未爆子弹药检测问题,并通过试验验证了该方法的可行性,但也发现存在虚警误判的问题。在同源图像配准检测的基础上,提出了基于SIFT和INIR(图像邻域信息重建)的可见光-红外异源图像差异性检测方法,该方法有效解决了误判漏判的问题。对比试验表明,异源图像差异性检测方法的检测精度相较于同源图像配准检测有了较大提升,但也导致无人机载荷平台单次工作时间大幅减少,为下一步地表未爆子弹药目标的准确识别定位提供了有力的技术支撑。

基于SIFT特征点提取的ICP配准算法

为解决传统迭代最近点(ICP)算法对点云配准的起始点对选择不佳而导致配准时间长、效率低的问题,提出一种基于尺度不变特征变换(SIFT)特征点提取的ICP点云配准算法(ST-ICP)。首先使用SIFT算法进行原始点云与目标点云的SIFT特征点提取,根据提取特征点完成快速点特征直方图(FPFH)特征运算,通过采样一致性初始配准算法(SAC-IA)搜索对应点对、求解变换矩阵,再进一步运用ICP算法进行点云精细配准。实验结果表明:与ICP算法相比较,ST-ICP算法的配准误差在迭代次数为5次时减小了1.019 cm,迭代次数为10次时减小了0.443 cm;在配准误差达到10^(-2) cm级别时,ST-ICP算法所用时间比传统ICP算法减少了12.829 s。ST-ICP算法优化了对应点对的选择,提升了配准精度和配准效率。

一种基于改进SIFT的视频稳像方法

为提高计算效率并保持良好的稳像效果,本文提出一种基于改进SIFT的视频稳像方法。首先对SIFT进行改进,并命名为BO-SIFT(Binarized Octagonal SIFT)。该算法引入了同心八边形环特征描述子,通过降维和二值化对特征向量进行处理,然后使用汉明距离进行特征点匹配,有效缩短了描述和匹配时间。其次将BO-SIFT算法应用于视频稳像,提取视频帧的特征点进行匹配,并计算出帧与帧之间的运动偏移量,以此实现运动估计。最后采用卡尔曼滤波器对估计出的运动偏移量进行平滑处理,并利用仿射变换对视频帧进行逆向补偿,从而得到稳定的图像序列。实验结果表明:相较于原始的SIFT算法,BO-SIFT算法使稳像时间减少了56.404%;相较于现有的较好算法,BO-SIFT算法稳像后的视频具有更高的平均峰值信噪比。此外,本文算法在不同视频上进行稳像效果测试,也具有一定的可靠性和优越性。

基于特征点和SAR-SIFT算法的林权测绘地形图像几何配准

目前常规的测绘地形图像几何配准主要通过对单应性矩阵进行求解,从而实现图像配准,由于对待配准图像的灰度化处理程度较低,导致配准精度较差。对此,本文研究基于特征点和合成孔径雷达(SAR)-尺度不变特征变换(SIFT)算法的林权测绘地形图像几何配准方法。首先对SAR图像成像机理进行分析,在此基础上采用滤波算法对图像进行处理,并采用加权平均法对红、绿、蓝三原色(RGB)通道分量进行灰度化处理。然后结合SAR-SIFT算法,构建多层次的尺度空间,并结合海森矩阵对像素梯度值进行计算,生成特征描述子。最后通过对特征描述子的相似度进行计算,结合匹配阈值,实现图像特征点的匹配。实验结果表明:采用基于特征点和SAR-SIFT算法的林权测绘地形图像几何配准方法对图像进行几何配准时,算法的均方根误差值较小,具备较为理想的几何配准精度。

Rate-limiting factors in hydrate decomposition through depressurization across various scales:A mini-review

Natural gas hydrate is an energy resource for methane that has a carbon quantity twice more than all traditional fossil fuels combined.However,their practical application in the field has been limited due to the challenges of long-term preparation,high costs and associated risks.Experimental studies,on the other hand,offer a safe and cost-effective means of exploring the mechanisms of hydrate dissociation and optimizing exploitation conditions.Gas hydrate decomposition is a complicated process along with intrinsic kinetics,mass transfer and heat transfer,which are the influencing factors for hydrate decomposition rate.The identification of the rate-limiting factor for hydrate dissociation during depressurization varies with the scale of the reservoir,making it challenging to extrapolate findings from laboratory experiments to the actual exploitation.This review aims to summarize current knowledge of investigations on hydrate decomposition on the subject of the research scale(core scale,middle scale,large scale and field tests)and to analyze determining factors for decomposition rate,considering the various research scales and their associated influencing factors.

基于3D-SIFT与4PCS融合的大数据量点云快速配准方法

测量点云与模型点云的配准是视觉定位的关键。针对测量点云数据量大且与CAD模型点云重叠率低造成视觉定位精度差、算法效率低的问题,提出一种基于三维尺度不变特征变换(3D-SIFT)与4点快速鲁棒匹配算法(4PCS)融合的测量点云与模型点云配准方法。首先利用深度相机对零件进行点云提取并对提取到的测量点云进行降噪和滤波处理;接着利用3D-SIFT特征点提取算法对测量点云和CAD模型点云进行特征点提取;最后把提取的特征点作为4PCS算法的初始值进行2种点云数据的配准。与常用的4PCS算法、Super-4PCS算法相比,在算法仿真与实际应用实验结果表明,本文算法在保证配准精度的前提下将配准速度提高30%以上。

Scale-space effect and scale hybridization in image intelligent recognition of geological discontinuities on rock slopes

Geological discontinuity(GD)plays a pivotal role in determining the catastrophic mechanical failure of jointed rock masses.Accurate and efficient acquisition of GD networks is essential for characterizing and understanding the progressive damage mechanisms of slopes based on monitoring image data.Inspired by recent advances in computer vision,deep learning(DL)models have been widely utilized for image-based fracture identification.The multi-scale characteristics,image resolution and annotation quality of images will cause a scale-space effect(SSE)that makes features indistinguishable from noise,directly affecting the accuracy.However,this effect has not received adequate attention.Herein,we try to address this gap by collecting slope images at various proportional scales and constructing multi-scale datasets using image processing techniques.Next,we quantify the intensity of feature signals using metrics such as peak signal-to-noise ratio(PSNR)and structural similarity(SSIM).Combining these metrics with the scale-space theory,we investigate the influence of the SSE on the differentiation of multi-scale features and the accuracy of recognition.It is found that augmenting the image's detail capacity does not always yield benefits for vision-based recognition models.In light of these observations,we propose a scale hybridization approach based on the diffusion mechanism of scale-space representation.The results show that scale hybridization strengthens the tolerance of multi-scale feature recognition under complex environmental noise interference and significantly enhances the recognition accuracy of GD.It also facilitates the objective understanding,description and analysis of the rock behavior and stability of slopes from the perspective of image data.

Scale effect removal and range migration correction for hypersonic target coherent detection

The detection of hypersonic targets usually confronts range migration(RM)issue before coherent integration(CI).The traditional methods aiming at correcting RM to obtain CI mainly considers the narrow-band radar condition.However,with the increasing requirement of far-range detection,the time bandwidth product,which is corresponding to radar’s mean power,should be promoted in actual application.Thus,the echo signal generates the scale effect(SE)at large time bandwidth product situation,influencing the intra and inter pulse integration performance.To eliminate SE and correct RM,this paper proposes an effective algorithm,i.e.,scaled location rotation transform(ScLRT).The ScLRT can remove SE to obtain the matching pulse compression(PC)as well as correct RM to complete CI via the location rotation transform,being implemented by seeking the actual rotation angle.Compared to the traditional coherent detection algorithms,Sc LRT can address the SE problem to achieve better detection/estimation capabilities.At last,this paper gives several simulations to assess the viability of ScLRT.

Validity,Reliability,and Measurement Invariance of the Thai Smartphone Application-Based Addiction Scale and Bergen Social Media Addiction Scale

Background:In recent years,there has been increased research interest in both smartphone addiction and social media addiction as well as the development of psychometric instruments to assess these constructs.However,there is a lack of psychometric evaluation for instruments assessing smartphone addiction and social media addiction in Thailand.The present study evaluated the psychometric properties and gender measurement invariance of the Thai version of the Smartphone Application-Based Addiction Scale(SABAS)and Bergen Social Media Addiction Scale(BSMAS).Method:A total of 801 Thai university students participated in an online survey from January 2022 to July 2022 which included demographic information,SABAS,BSMAS,and the Internet Gaming Disorder Scale-Short Form(IGDS9-SF).Results:Confirmatory Factor Analyses(CFAs)found that both the SABAS and BSMAS had a one-factor structure.Findings demonstrated adequate psychometric properties of both instruments and also supported measurement invariance across genders.Moreover,scores on the SABAS and BSMAS were correlated with scores on the IGDS9-SF.Conclusion:The results indicated that the SABAS and BSMAS are useful psychometric instruments for assessing the risk of smartphone addiction and social media addiction among Thai young adults.

Scale offers the possibility of identifying adherence to lifestyle interventions in patients with non-alcoholic fatty liver disease

Nonalcoholic fatty liver disease(NAFLD)is the most common chronic liver disorder,and dietary and lifestyle interventions remain the mainstays of NAFLD therapy.Zeng et al established a prediction system to evaluate adherence to lifestyle interventions in patients with NAFLD and choose optimal management.Here,we discuss the application scenarios of the scale and the areas warranting further attention,aiming to provide a possible reference for clinical recommend-ations.

Performance Improvement of Kenaf/Glass Polymer Hybrid Composites by Effective Application of Fish Scale Powder as Filler:A Novel Approach

Modern technology for developing new items made from renewable resources is becoming more and more popular as a result of rising environmental concern.Recently,contemporary polymer composites have included the hybridization of natural fibers with synthetic ones,along with the inclusion of a variety of biowaste filler for developing sustainable goods.In this work,the kenaf/glass hybrid polyester composites are strengthened by the addition of fish scale(FS),which is taken from the fishs outermost layer of skin.Five different stacked-order laminates,such as KKKK,KGKG,GKKG,KGGK,and GGGG,are fabricated by using the hand lay-up method with four different weight concentrations of filler content:0%,5%,10%,and 15%.Mechanical possessions such as tensile,flexural,impact strength and micro-hardness have been evaluated through experimentation in accordance with ASTM standards.The experimental findings revealed that,the tensile strength and micro-hardness value of KGKG laminates with 15wt% of FS filler are found to be maximum of 118.72 MPa and 17.82 HV respectively which are 39.67%and 26.11%greater than that of KGKG laminates without FS filler.However,the flexural and impact strength of same laminates with 10 wt% FS filler exhibited a maximum value of 142.77 MPa and 62.08 kJ/m^(2).In order to corroborate its applicability for structural and building materials in open environment,the dimensional stability of the composite has been studied through moisture absorption test.The influences of FS filler loading on dimensional stability and resistance to moisture absorption capacity of laminates are also investigated.The experimental results reflected that the addition of FS-filler has significantly improved the dimensional stability of the laminates in moist environment by reducing the moisture absorption tendency.To further support the mode of failures,a fractography investigation of fractured surfaces was conducted.

Water as a Standard Substance of a Logarithmic Poison Scale

The lethal dose LD50 represents the most important experimental value for acute toxicity. The simple logarithmic calculation of -log10 LD50 = value leads to the possible poison power pLD. As with the pH or pK value, respectively, for acid or the scale of earthquake intensities the logarithm helps making large differences of orders of magnitude easier to understand since they are more comparable. The higher the pLD value, the higher is the power of poison. An increase of the pLD value by 1 stands for a tenfold increase in toxicity. The lethal acute dose for water, one of the most important and at the same time non-toxic substances of all, is about one tenth of the body weight. This leads to a possible pLD value for water of 1, an ideal starting value for a logarithmic poison scale.

Characterization of Small-Scale Farmers and Assessment of Their Access to Crop Production Information in Selected Counties of Kenya

Small-scale farming accounts for 78% of total agricultural production in Kenya and contributes to 23.5% of the country’s GDP. Their crop production activities are mostly rainfed subsistence with any surplus being sold to bring in some income. Timely decisions on farm practices such as farm preparation and planting are critical determinants of the seasonal outcomes. In Kenya, most small-scale farmers have no reliable source of information that would help them make timely and accurate decisions. County governments have extension officers who are mandated with giving farmers advisory services to farmers but they are not able to reach most farmers due to facilitation constraints. The mode and format of sharing information is also critical since it’s important to ensure that it’s timely, well-understood and usable. This study sought to assess access to geospatial derived and other crop production information by farmers in four selected counties of Kenya. Specific objectives were to determine the profile of small-scale farmers in terms of age, education and farm size;to determine the type of information that is made available to them by County and Sub-County extension officers including the format and mode of provision;and to determine if the information provided was useful in terms of accuracy, timeliness and adequacy. The results indicated that over 80% of the farmers were over 35 years of age and over 56% were male. Majority had attained primary education (34%) or secondary education (29%) and most farmers in all the counties grew maize (71%). Notably, fellow farmers were a source of information (71%) with the frequency of sharing information being mostly seasonal (37%) and when information was available (43%). Over 66% of interviewed farmers indicating that they faced challenges while using provided information. The results from the study are insightful and helpful in determining effective ways of providing farmers with useful information to ensure maximum benefits.