Efficient Quadtree based Fractal Image Coding Scheme in Wavelet Transform Domain *

This paper presents an efficient quadtree based fractal image coding scheme in wavelet transform domain based on the wavelet based theory of fractal image compression introduced by Davis. In the scheme, zerotrees of wavelet coefficients are used to reduce the number of domain blocks, which leads to lower bit cost required to represent the location information of fractal coding, and overall entropy constrained optimization is performed for the decision trees as well as for the sets of scalar quantizers and self quantizers of wavelet subtrees. Experiment results show that at the low bit rates, the proposed scheme gives about 1 dB improvement in PSNR over the reported results.

Numerical simulation of wind-driven circulation and pollutant transport in Taihu Lake based on a quadtree grid

In this study,a two-dimensional flow-pollutant coupled model was developed based on a quadtree grid.This model was established to allow the accurate simulation of wind-driven flow in a large-scale shallow lake with irregular natural boundaries when focusing on important smallscale localized flow features.The quadtree grid was created by domain decomposition.The governing equations were solved using the finite volume method,and the normal fluxes of mass,momentum,and pollutants across the interface between cells were computed by means of a Godunov-type Osher scheme.The model was employed to simulate wind-driven flow in a circular basin with non-uniform depth.The computed values were in agreement with analytical data.The results indicate that the quadtree grid has fine local resolution and high efficiency,and is convenient for local refinement.It is clear that the quadtree grid model is effective when applied to complex flow domains.Finally,the model was used to calculate the flow field and concentration field of Taihu Lake,demonstrating its ability to predict the flow and concentration fields in an actual water area with complex geometry.

Terrain Rendering LOD Algorithm Based on Improved Restrictive Quadtree Segmentation and Variation Coefficient of Elevation

Aiming to deal with the difficult issues of terrain data model simplification and crack disposal,the paper proposed an improved level of detail(LOD)terrain rendering algorithm,in which a variation coefficient of elevation is introduced to express the undulation of topography.Then the coefficient is used to construct a node evaluation function in the terrain data model simplification step.Furthermore,an edge reduction strategy is combined with the improved restrictive quadtree segmentation to handle the crack problem.The experiment results demonstrated that the proposed method can reduce the amount of rendering triangles and enhance the rendering speed on the premise of ensuring the rendering effect compared with a traditional LOD algorithm.

Spatial Topology Rule Checking Algorithm of Linear Entity Based on Quadtree

Spatial topology rule is the primary method to insure the consistency and validity of spatial topology relation in GIS software. Topology rule can be divided into three categories according to geographic entity’s geometric shape: point topology rule, line topology rule and polygon topology rule. At first, this paper summarizes the various linear geographic entities’ topological relations which have practical application, then designs a series of linear entity topology rules detailedly. Based on these rules, this paper proposes a topology rule checking algorithm using quadtree, which is designed on the basis of MAPGIS7.4 spatial data model. The algorithm has already been applied to MAPGIS platform and gained good effects.

Framed-Quadtree Path Planning for an Underwater Vehicle with the Task of Tracking a Moving Target

An autonomous underwater vehicle (AUV) must use an algorithm to plan its path to distant, mobile offshore objects. Because of the uneven distribution of obstacles in the real world, the efficiency of the algorithm decreases if the global environment is represented by regular grids with all of them at the highest resolution. The framed quadtree data structure is able to more efficiently represent the environment. When planning the path, the dynamic object is expressed instead as several static objects which are used by the path planner to update the path. By taking account of the characteristics of the framed quadtree, objects can be projected on the frame nodes to increase the precision of the path. Analysis and simulations showed the proposed planner could increase efficiency while improving the ability of the AUV to follow an object.

Image Coding Using Wavelet Transform and EntropyConstrained Vector Quantization with Quadtree Structure Vectors

This paper presents a new wavelet transform image coding method. On the basis of a hierarchical wavelet decomposition of images, entropy constrained vector quantization is employed to encode the wavelet coefficients at all the high frequency bands with

An Improved Method of Linear Encoding Representation-A Cell Linear Quadtree

The linear encoding of a quadtree is an efficient way to represent the quadtree. In this paper, an improved linear quadtree, a cell linear quadtree, is proposed, in which its redundant storage is eliminated and the concept of a cell is introduced. Therefore, it has higher storage efficiency than a conventional linear quadtree.

基于改进Quadtree的移动对象索引策略

针对传统Quadtree索引移动对象存在的问题，提出多级阈值方法降低Quadtree的深度，采用数据缓存技术减少索引更新频度，增加对象标识的直接索引提高移动对象更新与查询性能，引进时间、速度、位置不确定性变量，以支持移动对象当前及未来信息的查询管理。改进后的Quadtree成为高效索引移动对象当前及未来位置信息的新方法。

基于要素稠度的电子海图矢量瓦片组合构建方法

为进一步提升基于WebGIS技术的电子海图在前端的加载效率,对海图矢量瓦片的划分方法进行深入研究。在考虑要素空间分布稠度的基础上,提出一种海图瓦片的组合格网划分方法。在四叉树划分过程中引入矢量瓦片数据阈值将其改造成非平衡四叉树,作为瓦片划分的第1阶段,在瓦片划分至单瓦片最大数据量小于数据阈值后,通过融合变异系数的二叉树对瓦片进行第2阶段的划分。将组合格网法、原始均匀格网法和改进均匀格网法生成的矢量瓦片数据与试验结果进行对比。研究结果表明,在瓦片第3~第9层级范围内,相对于2种均匀格网法,基于要素稠度的组合格网法生成的矢量瓦片在同一层级的标准差更小,矢量要素在瓦片上的分布更加均衡,在同一视窗范围内,组合格网法生成的瓦片加载时间更短。

基于四叉树扇形层值聚类的无人船障碍物检测

为了实现无人船自主导航过程中对障碍物的精确检测,提出了一种基于四叉树扇形层值聚类的无人船障碍物检测方法。首先基于四叉树扇形划分进行障碍物点云数据的检索,并剔除扇形象限内不可信数据;然后利用所获得的四叉树层值来求取全局密度距离,进而获得层值阈值,以此来对不规则多线形障碍物特征进行检测;最后通过建立数据点之间的空间拓扑关系来求取参考距离,并以参考距离为基准对障碍物点云数据进行聚类判定,提高聚类分割准确性。多线形障碍物特征识别性能测试及水面无人船障碍物检测实验结果表明,相较于其他密度聚类算法,在正检率、误检率和漏检率性能指标方面,多线形障碍物特征识别性能测试中,所提算法分别平均下降了9.86%、5.04%、3.10%,水面无人船障碍物检测中,所提算法分别平均下降了10.50%、6.97%、2.95%。

基于SBFEM的剪切型裂纹动态开裂模拟

剪切型断裂是岩土工程中常见的破坏模式,了解剪切破坏机理并准确预测剪切型裂纹的萌生、扩展过程对保障工程结构的安全性与稳定性具有重要意义.文章建立了基于比例边界有限元法(scaled boundary finite element methods,SBFEM)和非局部宏-微观损伤模型的剪切型裂纹动态开裂模拟方法,定义了基于偏应变概念的物质点对的正伸长量,可作为预测剪切型裂纹扩展行为的动态开裂准则,一点的损伤定义为该点影响域范围内连接的物质键损伤的加权平均值,而物质键的损伤则与基于偏应变概念的物质点对的正伸长量相关联,并引入能量退化函数建立结构域几何拓扑损伤与能量损失之间的关系,将拓扑损伤与应力应变联系起来,通过能量退化函数修正了SBFEM的刚度系数矩阵,得到了子域在损伤状态下的刚度矩阵,推导了考虑结构损伤的SBFEM动力控制方程,采用Newmark隐式算法对控制方程进行时间离散.最后,通过3个典型算例验证了建议的模型可较好地模拟剪切型断裂问题,能够很好地捕捉剪切型裂纹的扩展路径,并得到较为准确的载荷-位移曲线.

基于颜色直方图的电路板表面缺陷检测

为了提高废旧电路板的回收再利用率,针对回收电路板常见表面缺陷,在均匀分块基础上,提出了四叉树分裂颜色直方图缺陷检测方法。该方法能快速定位电路板表面缺陷,并通过支持向量机(SVM)实现缺陷分类,进而为电路板的二次利用提供质量保障。重点分析了分块大小与判断阈值对缺陷定位结果的影响,在保证检测精度的同时,检测速度相比均匀分块方法得到明显提升。与Faster-RCNN网络方法进行对比,结果表明该方法定位效果好,分类准确率平均达81%。

工业CT图像直接生成四边形网格的方法研究

针对具有内腔结构工件任意指定截面或局部的应力-应变分析问题,提出了一种将工业CT图像直接生成四边形网格有限元模型的方法。首先使用四叉树对工业CT图像进行初始网格划分;然后生成边界网格;最后进行网格优化。在生成边界网格过程中,结合二次误差函数与贝塞尔曲线,提出了准确定位网格边界点的方法;在四叉树邻域搜索算法基础上,提出了边界单元连接模板。为了提高边界网格质量,提出了优化模板进行优化。对生成网格进行质量评价,结果表明:网格边长比均小于2,倾斜度均小于0.6,满足实际工程要求。通过对汽车轮毂和阀体CT图像生成的四边形网格进行应力-应变分析,数值仿真结果验证了提出方法的正确性。

改进ORB提取匹配算法的SLAM应用研究

由于传统的ORB特征点提取匹配方法在图像纹理信息不丰富或者光照变化剧烈时极易产生特征点丢失、分布不均等问题,不利于SLAM系统的定位与建图。为此本文提出了一套较为鲁棒、精度较高的提取匹配算法。首先基于ORB特征点对其提取算法进行改进,计算自适应阈值并基于网格模型提取特征点,可提高特征点提取的鲁棒性并使其分布均匀。此外还提出了G-R图像匹配算法,基于网格特征计算邻域支持估计量来区分正误匹配点,再结合引入评价函数的RANSAC算法进一步剔除误匹配点,相比ORB-SLAM2原始匹配算法提高匹配精度9.36%,并减少时间消耗约13.6%。最后将本文提出的特征点提取匹配算法加入到ORB-SLAM2算法框架,经数据集与实际场景验证本文方法能有效提高ORB-SLAM2系统定位精度36.6%以上,使系统更具鲁棒性。

基于四叉树法和PROSAC算法改进的视觉SLAM技术

为解决在同时定位与地图构建(simultaneous localization and mapping,SLAM)的前端进行特征点匹配时,随机抽样一致法(random sample consensus,RANSAC)存在的迭代次数高、实时性较差、鲁棒性不稳定等问题,提出一种基于四叉树法和渐进一致采样法(progressive sample consensus,PROSAC)算法融合改进的图像匹配算法。实现四叉树法+PROSAC算法的误匹配剔除算法,在EuRoC数据集上对改进后的ORB-SLAM2算法进行实验。结果表明:相比于ORB-SLAM2系统,该算法在Vicon Room 103数据集上总体绝对轨迹误差平均值减小了39.28%,总体相对位姿误差减小了35.45%,具有更高的建图精度。

昏暗环境下自适应ORB-SLAM2算法研究

本文针对ORB-SLAM2算法在黑暗环境或纹理较少的环境下提取特征点少,从而导致SLAM系统定位精度不高、匹配对数较少,进而导致系统崩溃的问题,提出了一种基于自适应阈值的特征点提取算法与改进的四叉树均匀化策略。首先基于图像的亮度进行基于自适应阈值的FAST特征点提取,之后通过改进的四叉树均匀化策略对图像的特征点进行剔除与补偿,完成特征点选取。实验结果表明,与原算法相比,改进后的特征点提取算法在黑暗环境与纹理较少的环境下,匹配对的数量提升了17.6%,SLAM轨迹精度提升了49.8%,有效的提升了SLAM系统的鲁棒性和精度。

结合差分隐私与安全聚集的联邦空间数据发布方法

针对联邦空间数据的数据孤岛问题、空间数据索引问题以及发布联邦空间数据存在的隐私问题,提出基于动态四分树的联邦空间数据发布(FSP)方法。首先,在FSP方法的每轮迭代中,服务端把四分树副本共享给该轮中每个客户端,每个客户端利用四分树副本编码自身位置数据,利用Polya分布产生离散噪声在本地扰动编码结果;其次,结合容错学习(LWE)生成本地掩码对噪声结果进行加密;再次,安全聚集端结合该轮迭代中每个客户端的报告值,执行安全聚集与消除掩码操作,然后把聚集结果发送给服务端;最后,服务端结合收集的编码向量与噪声方差自底向上地动态修剪四分树结构。在Beijing、Checkin、NYC和Landmark 4个空间数据集上的实验结果表明,FSP方法在保证客户端隐私的同时,与已有的较好的联邦空间数据发布方法AHH(Adaptive Hierarchical Histograms)相比,在隐私预算为1.8时,FSP的均方误差(MSE)分别降低了3.80%、2.96%、7.51%和14.13%。可见使用FSP方法进行联邦空间数据发布的精度优于同类方法。

一种基于编码单元快速划分的VVC帧内编码方法

相比于高效视频编码(high efficiency video coding,HEVC)标准,新一代编码标准多功能视频编码(versatile video coding,VVC)引入了很多新的技术,其中包括四叉树(quadtree,QT)和多类型树(multi-type tree,MTT)划分,MTT划分由HEVC中的QT划分延伸而来。新划分方法提高了压缩效率,但导致编码时间急剧增加。为了降低编码复杂度,提出了一种结合深度学习方法和MTT方向早期判决的快速帧内编码算法。首先使用轻量级的卷积神经网络(convolutional neural network,CNN)对QT和部分MTT进行预测划分,其余MTT则采用提前预测MTT划分方向的方法作进一步的优化。实验结果表明,所提方法能够大幅降低编码复杂度,相比于原始编码器的编码时间减少了74.3%,且只有3.3%的码率损失,性能优于对比的方法。

基于改进四叉树和比例边界有限元法的自适应设计域拓扑优化方法

针对大型结构拓扑优化计算成本高和固体各向同性材料惩罚模型(SIMP)在优化后结构边界处求解精度低的问题,提出了一种基于SIMP法的自适应设计域(ADD)拓扑优化方法。将改进四叉树法应用在拓扑优化的过程中,通过自动划分不同等级的网格单元来减少网格数量、减轻计算负担并提高边界处求解精度;采用比例边界有限元法(SBFEM)实时计算划分后结构的有限元信息,解决了不同等级网格间悬挂节点的问题。所提方法可在初始网格相对较少的情况下得到更加精确的结果,大幅度地降低了计算成本。数值算例结果表明,所提方法在最终结构边界处精度相同的情况下,计算时间最快可缩短为原来的1/100,可以为后续降低大型结构拓扑优化的计算成本提供参考。

基于多尺度引导滤波的实时视频去雾算法

为解决当前去雾算法中存在的透射率估计不准确、天空区域颜色恢复较差、运行速度较慢等问题,本文提出了一种基于多尺度引导滤波的实时视频去雾算法。首先,采用自动白平衡算法对有雾图像进行颜色校正,颜色校正后的图像与暗原色置信因子作为引导滤波,随后使用金字塔采样技术获取缩小后的图像,接着使用四叉树算法来估算透射率和大气光强度,并且不断地迭代上采样和引导图像滤波,从而有效防止了信息的损失,最后达到最佳的传输效果。此外,将单图像去雾算法扩展到了实时视频去雾,通过使传输值在时间上一致来减少去雾视频中的闪烁伪影。实验结果表明,该算法的运行速率较快,去雾效果明显,在视频去雾中闪烁伪影较少。相比于屏蔽泊松方程去雾算法,本文算法的平均梯度提高了67.08%,相比于暗通道先验逐帧视频去雾算法,本算法的时间加快了81.32%,符合实时视频运行稳定、处理快速的要求。