基于PSO的Wang Tiles纹理合成

如何在纹理样图中选择组成Wang Tiles的图像块决定着纹理的合成质量。基于PSO的Wang Tiles纹理合成通过粒子群优化算法在纹理样图中快速搜索边界差异最小的图像块,并用选取的图像块构建Wang Tiles,最后用Wang Tiles纹理合成算法合成纹理。实验表明,该算法合成的纹理具有较少的接缝,比随机选择图像块具有更好的合成效果。     

Piecewise generation of surfaces of curved ducts of varying cross-sectional area using patch shifting or overlap

Using the results of previous numerical experiments, it is shown how patch shifting or overlap can be used to produce a continuous surface for curved ducts of varying elliptical or circular cross-sectional area. It is shown that if two overlapping Bézier patches are bounded in the v-parametric direction, say, by the same two u-parametric curves, the derivatives of both patches can, indeed be equal over the overlapping portions of the patches. It is then deduced that the mathematical result may hold only over a limited area of the overlapping patches on which sections of the patches can be joined to produce a continuous duct surface.     

Seamless patches for GPU-based terrain rendering

In this paper we present a novel approach for interactive rendering of large terrain datasets. Our approach is based on subdividing a terrain into rectangular patches at different resolutions. Each patch is represented by four triangular tiles that are selected form different resolutions, and four strips which are used to stitch the four tiles in a seamless manner. Such a scheme maintains resolution changes within patches through the stitching strips, and not across patches. At runtime, these patches are used to construct a level-of-detail representation of the input terrain based on view-parameters. A selected level of detail only includes the layout of the patches and their boundary edges resolutions. The layout includes the location and dimension of each patch. Within the graphics hardware, the GPU generates the meshes of the patches by using scaled instances of cached tiles and assigns elevation for each vertex from cached textures. Since adjacent rectangular patches agree on the resolution of the common edges, the resulted mesh does not include cracks or degenerate triangles. Our algorithm manages to achieve quality images at high frame rates while providing seamless transition between different levels of detail.     

团雾激光散射特性及其在智能交通监测中的应用

针对团雾会导致智能交通的安全性受到极大影响,通过分析团雾中水凝物的散射特性,定量给出了高浓度水凝物散射造成的激光能量在散射过程中将出现较大幅度的衰减,并给出了一种基于水凝物激光散射特性的团雾监测系统架构.     

An empirical study of supplementary patches in open source projects

Developers occasionally make more than one patch to fix a bug. The related patches sometimes are intentionally separated, but unintended omission errors require supplementary patches. Several change recommendation systems have been suggested based on clone analysis, structural dependency, and historical change coupling in order to reduce or prevent incomplete patches. However, very few studies have examined the reason that incomplete patches occur and how real-world omission errors could be reduced. This paper systematically studies a group of bugs that were fixed more than once in open source projects in order to understand the characteristics of incomplete patches. Our study on Eclipse JDT core, Eclipse SWT, Mozilla, and Equinox p2 showed that a significant portion of the resolved bugs require more than one attempt to fix. Compared to single-fix bugs, the multi-fix bugs did not have a lower quality of bug reports, but more attribute changes (i.e., cc’ed developers or title) were made to the multi-fix bugs than to the single-fix bugs. Multi-fix bugs are more likely to have high severity levels than single-fix bugs. Hence, more developers have participated in discussions about multi-fix bugs compared to single-fix bugs. Multi-fix bugs take more time to resolve than single-fix bugs do. Incomplete patches are longer and more scattered, and they are related to more files than regular patches are. Our manual inspection showed that the causes of incomplete patches were diverse, including missed porting updates, incorrect handling of conditional statements, and incomplete refactoring. Our investigation showed that only 7 % to 17 % of supplementary patches had content similar to their initial patches, which implies that supplementary patch locations cannot be predicted by code clone analysis alone. Furthermore, 16 % to 46 % of supplementary patches were beyond the scope of the immediate structural dependency of their initial patch locations. Historical co-change patterns also showed low precision in predicting supplementary patch locations. Code clones, structural dependencies, and historical co-change analyses predicted different supplementary patch locations, and there was little overlap between them. Combining these analyses did not cover all supplementary patch locations. The present study investigates the characteristics of incomplete patches and multi-fix bugs, which have not been systematically examined in previous research. We reveal that predicting supplementary patch is a difficult problem that existing change recommendation approaches could not cover. New type of approaches should be developed and validated on a supplementary patch data set, which developers failed to make the complete patches at once in practice.     

Reference consistent reconstruction of 3D cloth surface

We propose a multiview method for reconstructing a folded cloth surface on which regularly-textured color patches are printed. These patches provide not only easy pixel-correspondence between multiviews but also the following two new functions. (1) Error recovery: errors in 3D surface reconstruction (e.g. errors in occlusion boundaries and shaded regions) can be recovered based on the spatio-temporal consistency of the patches. (2) Single-view hole filling: patches that are visible only from a single view can be extrapolated from the reconstructed ones based on the regularity of the patches. Using these functions for improving 3D reconstruction also produces the patch configuration on the reconstructed surface, showing how the cloth is deformed from its reference shape. Experimental results demonstrate the above improvements and the accurate patch configurations produced by our method.     

基于图像块分类的图像超分辨率重建

针对当前图像超分辨率重建算法中存在的字典单一而导致重建图像质量不佳的问题，提出一种将图像块分类与图像卡通纹理分解相结合的单幅图像超分辨率重建算法。首先，将图像分块，并将图像块分为边缘类、纹理类和平滑类三类，其中纹理类用形态成分分析（MCA）算法分解为卡通部分和纹理部分；然后，对边缘类、卡通部分和纹理部分分别训练高低分辨率字典；最后，求解稀疏系数并与高分辨率字典重建图像块。仿真结果显示，与基于稀疏表示的超分辨率重建（SCSR）算法和单幅图像超分辨率重建（SISR）算法相比，所提算法的峰值信噪比（PNSR）值分别提高了0.26 dB和0.14 dB，表明该算法的重建效果更好，重建图像纹理细节更丰富。     

Distribution Update of Deformable Patches for Texture Synthesis on the Free Surface of Fluids

We propose an approach for temporally coherent patch‐based texture synthesis on the free surface of fluids. Our approach is applied as a post‐process, using the surface and velocity field from any fluid simulator. We apply the texture from the exemplar through multiple local mesh patches fitted to the surface and mapped to the exemplar. Our patches are constructed from the fluid free surface by taking a subsection of the free surface mesh. As such, they are initially very well adapted to the fluid's surface, and can later deform according to the free surface velocity field, allowing a greater ability to represent surface motion than rigid or 2D grid‐based patches. From one frame to the next, the patch centers and surrounding patch vertices are advected according to the velocity field. We seek to maintain a Poisson disk distribution of patches, and following advection, the Poisson disk criterion determines where to add new patches and which patches should e flagged for removal. The removal considers the local number of patches: in regions containing too many patches, we accelerate the temporal removal. This reduces the number of patches while still meeting the Poisson disk criterion. Reducing areas with too many patches speeds up the computation and avoids patch‐blending artifacts. The final step of our approach creates the overall texture in an atlas where each texel is computed from the patches using a contrast‐preserving blending function. Our tests show that the approach works well on free surfaces undergoing significant deformation and topological changes. Furthermore, we show that our approach provides good results for many fluid simulation scenarios, and with many texture exemplars. We also confirm that the optical flow from the resulting texture matches the fluid velocity field. Overall, our approach compares favorably against recent work in this area.     

Triangulation of 3D surfaces

A simple generator of graded triangular meshes on spatial surfaces is introduced in this paper. The algorithm is based on the approximation of the surface by tensor product polynomial patches which are uniquely mappable on a planar parametric space. Each of the patches is triangulated separately in its parametric space, using modified advancing front technique allowing for generation of pre-stretched elements, and the obtained triangulation is mapped back onto the original surface. Large effort has been devoted to the treatment of singularities arising on surfaces approximated by degenerated patches.     

Goal‐based Caustics

We propose a novel system for designing and manufacturing surfaces that produce desired caustic images when illuminated by a light source. Our system is based on a nonnegative image decomposition using a set of possibly overlapping anisotropic Gaussian kernels. We utilize this decomposition to construct an array of continuous surface patches, each of which focuses light onto one of the Gaussian kernels, either through refraction or reflection. We show how to derive the shape of each continuous patch and arrange them by performing a discrete assignment of patches to kernels in the desired caustic. Our decomposition provides for high fidelity reconstruction of natural images using a small collection of patches. We demonstrate our approach on a wide variety of caustic images by manufacturing physical surfaces with a small number of patches.     

Jointly Optimized Regressors for Image Super‐resolution

Learning regressors from low‐resolution patches to high‐resolution patches has shown promising results for image super‐resolution. We observe that some regressors are better at dealing with certain cases, and others with different cases. In this paper, we jointly learn a collection of regressors, which collectively yield the smallest super‐resolving error for all training data. After training, each training sample is associated with a label to indicate its ‘best’ regressor, the one yielding the smallest error. During testing, our method bases on the concept of ‘adaptive selection’ to select the most appropriate regressor for each input patch. We assume that similar patches can be super‐resolved by the same regressor and use a fast, approximate kNN approach to transfer the labels of training patches to test patches. The method is conceptually simple and computationally efficient, yet very effective. Experiments on four datasets show that our method outperforms competing methods.     

Deep reinforcement learning-based patch selection for illuminant estimation

Previous deep learning based approaches to illuminant estimation either resized the raw image to lower resolution or randomly cropped image patches for the deep learning model. However, such practices would inevitably lead to information loss or the selection of noisy patches that would affect estimation accuracy. In this paper, we regard patch selection in neural network based illuminant estimation as a controlling problem of selecting image patches that could help remove noisy patches and improve estimation accuracy. To achieve this, we construct a selection network (SeNet) to learn a patch selection policy. Based on data statistics and the learning progression state of the deep illuminant estimation network (DeNet), the SeNet decides which training patches should be input to the DeNet, which in turn gives feedback to the SeNet for it to update its selection policy. To achieve such interactive and intelligent learning, we utilize a reinforcement learning approach termed policy gradient to optimize the SeNet. We show that the proposed learning strategy can enhance the illuminant estimation accuracy, speed up the convergence and improve the stability of the training process of DeNet. We evaluate our method on two public datasets and demonstrate our method outperforms state-of-the-art approaches.     

A New Framework for Multiscale Saliency Detection Based on Image Patches

In this paper, we propose a new multiscale saliency detection algorithm based on image patches. To measure saliency of pixels in a given image, we segment the image into patches by a fixed scale and then use principal component analysis to reduce the dimensions which are noises with respect to the saliency calculation. The dissimilarities between a patch and other patches, which indicate the patch’s saliency, are computed based on the dissimilarity of colors and the spatial distance. Finally, we implement our algorithm through multiple scales that further decrease the saliency of background. Our method is compared with other saliency detection approaches on two public image datasets. Experimental results show that our method outperforms the state-of-the-art methods on predicting human fixations and salient object segmentation.     

Patch-based Segmentation of Latent Fingerprint Images Using Convolutional Neural Network

Latent fingerprint segmentation involves marking out all the foreground regions accurately in a latent fingerprint image, but due to poor quality images and complex background, segmentation of latent fingerprint images is one of the most difficult tasks in automatic latent fingerprint recognition systems. In this article, we propose a patch-based technique for segmentation of latent fingerprint images, which uses Convolutional Neural Network (CNN) to classify patches. CNN has recently shown impressive performance in the field of pattern recognition, classification, and object detection, which inspired us to use CNN for this complex task. We trained the CNN model using SGD to classify image patches into fingerprint and non-fingerprint classes followed by proposed false patch removal technique, which uses “majority of neighbors” to remove the isolated and miss-classified patches. Finally, based on the final class of patches, an ROI is constructed to mark out the foreground from the background of latent fingerprint images. We tested our model on IIIT-D latent fingerprint database and the experimental results show improvements in the overall accuracy compared to existing methods.     

基于熵及随机擦除的针对目标检测物理攻击的防御

物理攻击通过在图像中添加受扰动的对抗块使得基于深度神经网络(DNNs)的应用失效,对DNNs的安全性带来严重的挑战。针对物理攻击方法生成的对抗块与真实图像块之间的信息分布不同的特点,本文提出了能有效避免现有物理攻击的防御算法。该算法由基于熵的检测组件(Entropy-basedDetectionComponent,EDC)和随机擦除组件(RandomErasingComponent,REC)两部分组成。EDC组件采用熵值度量检测对抗块并对其灰度替换。该方法不仅能显著降低对抗块对模型推理的影响,而且不依赖大规模的训练数据。REC模块改进了深度学习通用训练范式。利用该方法训练得到的深度学习模型,在不改变现有网络结构的前提下,不仅能有效防御现有物理攻击,而且能显著提升图像分析效果。上述两个组件都具有较强的可转移性且不需要额外的训练数据,它们的有机结合构成了本文的防御策略。实验表明,本文提出的算法不仅能有效的防御针对目标检测的物理攻击(在Pascal VOC 2007上的平均精度(m AP)由31.3%提升到64.0%及在Inria数据集上由19.0%提升到41.0%),并且证明算法具有较好的...     

基于图像块匹配的肾小球基底膜自动分割

为了实现肾小球基底膜的自动分割,提出了一种基于图像块匹配策略的图像自动分割方法。首先,针对肾小球基底膜的特点,将块匹配算法的搜索范围从一幅参考图像扩展到多幅参考图像,并采用了一种改进的搜索方式提高匹配效率;然后,开始搜索最优的图像匹配块;最后,提取最优匹配块对应的标记匹配块进行加权,重组为肾小球基底膜的初始分割结果,经形态学处理后即可得到基底膜最终的分割结果。在采集到的肾小球透射电（TEM）图像上进行测试,算法分割结果与病理专家手动分割结果之间的Jaccard相似系数最低为83%,最高为95%。实验结果表明所提方法能取得精度较高的分割结果。     

Global multipath Monte Carlo algorithms for radiosity

We present a new algorithm for radiosity, which is linear both in time and storage with the number of patches in a given scene; that is, the number of patches is increased by dividing each existing patch into smaller patches. The new algorithm is based on Integral Geometry and on the well-known algorithms for Monte Carlo particle transport. It considers the scene embedded in a field of lines, each representing the exchange between the pairs of points that result from the intersections of the line with the scene. For a given pair of points, this exchange is bidirectional. Lines are taken randomly. Both graphical and numerical results are given for the proposed algorithm, which is compared with a local Monte Carlo particle transport algorithm.     

Discovering discriminative patches for free-hand sketch analysis

Since the ancient times, free-hand sketch has been widely used as an effective and convenient intermediate means to express human thoughts and highly diverse objects in reality. In recent years, a great quantity of researchers realized the significance of sketch and gradually focused on sketch-related problems, such as sketch-based image retrieval and recognition. Despite so many achievements, very few works concentrate on exploring the intrinsic factors which potentially influence the vivid degree of sketch. In this paper, we propose a weak supervised approach to discover the most discriminative patches for different categories of sketches, which perhaps grasp the key to a good free-hand sketch. In the beginning, we randomly extract tens of thousands of patches at multiple scales. After that, pyramid histogram of oriented gradient is calculated to represent these patches as an effective and uniform feature representation. To find the most discriminative patches for each class of sketches, we design an iterative detection process which combines cluster merging and discriminative ranking. The experimental results on the TU-Berlin sketch benchmark dataset demonstrate the effectiveness of the proposed method, as compared to other available approaches. Moreover, a reasonable analysis and discussion about good and bad sketches is provided based on the visual results.