估计小波系数方差邻域大小的研究

利用小波变换去噪时小波系数方差的估计对去噪结果影响很大。自然图像小波分解后得到的系数在不同的分辨率中差异很大,所以利用邻域估计中心点方差时,不同分辨率应有不同大小的邻域。首先对在邻域中利用极大似然准则估计中心点方差进行分析,再结合自然图像小波分解后的系数在不同分辨率子带中,根据平稳性和重要性选择邻域的大小。最后进行去噪实验,并取得正交小波分解下理想的去噪性能。     

基于复合顺序形态变换的车牌边缘检测

边缘检测是图像分割的重要环节,边缘提取的好坏直接影响了整个图像处理的效果。数学形态学用具有一定形态结构的“结构元素”去度量图像的形态,以解决图像处理和分析问题。文章通过对多种边缘检测方法的分析比较,结合高速公路路口汽车图像噪声大的特点,提出了基于复合顺序形态变换的车牌边缘检测。该方法不但能有效地改善有噪图像边缘检测的效果,鲁棒性好,而且并行快速,便于硬件实现。     

Specimen preparation and image processing and analysis techniques for automated quantification of concrete microcracks and voids

Specimen preparation and image processing/analysis techniques were developed for use in automated quantitative microstructural investigation of concrete, focusing on concrete microcracks and voids. Different specimen preparation techniques were developed for use in fluorescent and scanning electron microscopy (SEM) of concrete; then techniques produce a sharp contrast between microcracks/voids and the body of concrete. The image processing/analysis techniques developed specifically for use with concrete address the following usages: automatic threshold; development of intersecting microcracks/voids and connected voids; distinction of microcracks form voids based on geometric attributes; and noise filtration.     

基于二维直方图分析的二元子图微光图像增强处理

提出了基于二维直方图分析的二元子图微光图像增强处理算法，根据原始图像的二维直方图分布特征，依据相邻像素的相关性数据，将图像分割成二元子图像，对每个子图像分别进行直方图均衡增强处理，最后合成得到完整的处理图像，通过用信息容量等图像评价参数对实验结果进行分析，表明此算法克服了对整幅图像进行直方图均衡时难以把握局部细节的缺点，得到了具有较好视觉质量的处理后微光图像。     

Pedestrian detection and tracking in infrared imagery using shape and appearance

In this paper, we present an approach toward pedestrian detection and tracking from infrared imagery using joint shape and appearance cues. A layered representation is first introduced and a generalized expectation-maximization (EM) algorithm is developed to separate infrared images into background (still) and foreground (moving) layers regardless of camera panning. In the two-pass scheme of detecting pedestrians from the foreground layer: shape cue is first used to eliminate non-pedestrian moving objects and then appearance cue helps to locate the exact position of pedestrians. Templates with varying sizes are sequentially applied to detect pedestrians at multiple scales to accommodate different camera distances. To facilitate the task of pedestrian tracking, we formulate the problem of shot segmentation and present a graph matching-based tracking algorithm that jointly exploits the shape, appearance and distance information. Experimental results with both OSU Infrared Image Database and WVU Infrared Video Database are reported to demonstrate the accuracy and robustness of our algorithm.     

污染扩散瞬时浓度场测量方法的研究

提出了一种以粒子成像、粒子散射和统计光学为理论基础的污染扩散非定常瞬时浓度场的测量原理，设计了大范围瞬时浓度场测量系统，利用该系统对瞬时扩散的浓度场进行了实际测量，通过数字图像分析和处理得到污染扩散的非定常瞬时浓度场。     

产品图象智能检测技术研究

工业界对图象智能检测技术有着迫切的需要。本文通过对火腿肠产品在线感官质量检测关键性技术的深入研究，证实了该项技术住火腿肠类产品在线感官检测中应用的可行性，文中重点讨论了产品图象智能判别系统的硬软件结构、图象检测判别万法、检测实验数据及实验结果分析。     

Near-wall porosity characteristics of fixed beds packed with wood chips

Packed beds of fuel wood chips are commonly found in thermal conversion processes such as combustion or gasification. Wood chips in particular are mostly used as fuel for small-scale domestic heating boilers but also for commercial-scale combustion units. The characterization of spatial voidage distribution inside the wood chip beds is of great importance for flow and reactor modelling. This study focuses on the radial porosity variations of cylindrical beds of three different types of commercially available wood chips including chips classified as G30 size class. The conventional technique of consolidating packed beds with a resin was chosen as the experimental procedure. The radial voidage distribution in different cylindrical beds is determined by image analysis of sections of the solidified packings. Additionally, a packing of monosized spheres was investigated in order to assess the selected procedure in comparison with widely available literature data for spheres. The results are discussed and summarized in a mathematical expression correlating the radial voidage distribution depending on average wood chip size, packing core porosity and dimensionless distance from the tube wall.     

像全息的水下应用

基于开发海洋事业的需要，对如何拍摄好水中物体的全息图进行了详细的理论分析，并利用像全息的方法成功拍摄到了一系列富有工艺美感的水下物体的全息图。针对拍摄的全息图进行了细致的定性、定量分析，最后对仪器开发和应用前景作了初步的探讨。     

一种基于对象几何形状的视频抗误码方法

针对压缩视频流中的误码,提出了一种新颖的基于对象几何形状的视频抗误码方法.首先,在编码端通过比较帧间编码帧内邻近宏块的运动向量差异获得粗略的对象几何形状信息,以对象标示图表示;再把它以与标准码流兼容的方式嵌入到压缩码流里去;最后,在解码端利用它帮助精确的构造错误宏块的运动向量.试验结果表明,该方法对于即使大量连续slice出错的情况仍能获得比较好的掩盖效果,而增加的运算复杂度和传输负载量都不大.