Feature-point-extracting-based automatically mosaic for composite microscopic images

Image mosaic is a crucial step in the three-dimensional reconstruction of composite materials to align the serial images. A novel method is adopted to mosaic two SiC/Al microscopic images with an amplification coefficient of 1000. The two images are denoised by Gaussian model, and feature points are then extracted by using Harris corner detector. The feature points are filtered through Canny edge detector. A 40×40 feature template is chosen by sowing a seed in an overlapped area of the reference image, and the homologous region in floating image is acquired automatically by the way of correlation analysis. The feature points in matched templates are used as feature point-sets. Using the transformational parameters acquired by SVD-ICP method, the two images are transformed into the universal coordinates and merged to the final mosaic image.     

Segmentation of tumor ultrasound image via region-based Ncut method

To segment the tumor region precisely is a prerequisite for ultrasound navigation and treatment. In this paper, a normalized cut method to segment tumor ultrasound image is proposed by means of simple linear iterative clustering for presegmentation procedure. The first step, we use simple linear iterative clustering algorithm to divide the image into a number of homogeneous over-segmented regions. Then, these regions are regarded as nodes, and a similarity matrix is constructed by comparing the histograms of each two regions. Finally, we apply the Ncut method to merging the over-segmented regions, then the image segmentation process is completed. The results show that the proposed segmentation scheme handles the strong speckle noise, low contrast, and weak edges well in ultrasound image. Our method has high segmentation precision and computation efficiency than the pixel-based Ncut method.     

区间灰数表征与算法改进及其GM (1, 1)模型应用研究

A fast edge detection method basing on the combination of fuzzy subsets is developed, in which the detection of an edge as a classification problem will be considered, partitioning the image into two portions: the edge portion and the non-edge portion. The latter one, as the main constituent of an image, consists of the object and its background. Removing the non-edge portion from an image, the remainder is nothing but the edge of this image. As far as the fuzziness of the edge of an image is concerned, some fuzzy operations can be made. In this paper, the gray level histogram is partitioned into several sub-regions, and some operations are performed with the associated fuzzy subsets corresponding to those sub-edges in the sub-regions on the gray-level-square-difference histogrmn, and the edge of this image is finally obtained. Practical examples in this paper illustrate that, the described method is simple and effective to achieve an ideal edge image.     

Population information extraction in Chaohu watershed based on RS and GIS

It is always difficult to extract population information of small watershed during the region environment assessment.The report adopted compound areal interpolation to study the population in Chaohu watershed with the GIS technique and landuse interpretation data from remote sensing.The result indicated that the method is effective to extract population information of small watershed.     

Segmentation algorithm of complex ore images based on templates transformation and reconstruction

Lots of noises and heterogeneous objects with various sizes coexist in a complex image, such as an ore image; the classical image thresholding method cannot effectively distinguish between ores. To segment ore objects with various sizes simultaneously, two adaptive windows in the image were chosen for each pixel; the gray value of windows was calculated by Otsu’s threshold method. To extract the object skeleton, the definition principle of distance transformation templates was proposed. The ores linked together in a binary image were separated by distance transformation and gray reconstruction. The seed region of each object was picked up from the local maximum gray region of the reconstruction image. Starting from these seed regions, the watershed method was used to segment ore object effectively. The proposed algorithm marks and segments most objects from complex images precisely.     

An improved automatic detection method for earthquake-collapsed buildings from ADS40 image

Earthquake-collapsed building identification is important in earthquake damage assessment and is evidence for mapping seismic intensity. After the May 12th Wenchuan major earthquake occurred, experts from CEODE and IPSC collaborated to make a rapid earthquake damage assessment. A crucial task was to identify collapsed buildings from ADS40 images in the earthquake region. The difficulty was to differentiate collapsed buildings from concrete bridges, dry gravels, and landslide-induced rolling stones since they had a similar gray level range in the image. Based on the IPSC method, an improved automatic identification technique was developed and tested in the study area, a portion of Beichuan County. Final results showed that the technique＇s accuracy was over 95%. Procedures and results of this experiment are presented in this article. Theory of this technique indicates that it could be applied to collapsed building identification caused by other disasters.     

Lapel Pattern Recognition in Flat Sketches Based on Lapel Model

Taking lapels as examples,the purpose of this paper is to study the problem of recognizing the same clothing component in the flat sketches. Supposing that the four lines in a lapel are concurrent and symmetrical,a mathematical lapel model is built. Then,a novel method for lapel recognition in fashion flat sketches based on the lapel model is proposed. In the image preprocessing stage, the images need to be cropped for two times in order to remove the blank background and extract the region of interest,respectively. In the concurrent recognition stage,line detection,θ selection,and curve fitting are applied to limiting candidate lines. In the symmetrical recognition stage, k-means clustering algorithm is employed to divide the selected lines into four clusters. The threshold values of the difference of corresponding weighted θ are set as lapel recognition criteria. Experiments demonstrate that the recognition accuracy of the proposed method is obtained at about 91. 7%.     

Efficient Approach for Face Detection in Video Surveillance

Security access control systems and automatic video surveillance systems are becoming increasingly important recently, and detecting human faces is one of the indispensable processes. In this paper, an approach is presented to detect faces in video surveillance. Firstly, both the skin-color and motion components are applied to extract skin-like regions. The skin-color segmentation algorithm is based on the BPNN (back-error-propagation neural network) and the motion component is obtained with frame difference algorithm. Secondly, the image is clustered into separated face candidates by using the region growing technique. Finally, the face candidates are further verified by the rule-based algorithm. Experiment results demonstrate that both the accuracy and processing speed are very promising and the approach can be applied for the practical use.     

Adaptive Image Digital Watermarking with DCT and FCM

A novel adaptive digital image watermark algorithm is proposed. Fuzzy c-means clustering （FCM） is used to classify the original image blocks into two classes based on several characteristic parameters of human visual system （HVS）. One is suited for embedding a digital watermark, the other is not. So the appropriate blocks in an image are selected to embed the watermark. The wetermark is embedded in the middle-frequency part of the host image in conjunction with HVS and discrete cosine transform （DCT）. The maximal watermark strength is fixed according to the frequency masking. In the same time, for the good performance, the watermark is modulated into a fractal modulation array. The simulation results show that we can remarkably extract the hiding watermark and the algorithm can achieve good robustness with common signal distortion or geometric distortion and the quality of the watermarked image is guaranteed.     

Improving reflectance estimation by BRDF-consistent region clustering

Previous studies in reflectance estimation generally require prior segmentation of an image into regions of uniform reflectance. Due to the measurement noise and limited sampling of the BRDF (bi-directional reflectance function) directions, such estimated results of reflectance are not accurate. In this paper, we propose a novel method for reducing uncertainty in reflectance estimates by merging image regions which have consistent reflectance observations. Each image region acts as a reflectance subspace, so merging of the image regions can result in subspace reduction. We propose a Bayesian segmentation framework to decrease the reflectance uncertainty by using novel merging criteria. Finally, a maximum likelihood reflectance estimation is made for each resulting image region. Experimental results verify the feasibility and superiority of this reflectance-oriented region merging method.     

文本性的盛宴：美国汉学现当代文学研究的亮点和拐点

由于传统焊缝区域检测算法难以准确提取模糊和对比度低的厚钢管焊缝区域,提出一种新的基于鲁棒PCA模型的焊缝区域检测算法,该算法能克服传统方法的不足,并能准确提取焊缝区域.首先,收集一序列X射线图像,并对其进行空域对齐及亮度归一化预处理.其次,计算得到系列图像的背景图像,并将背景图像与待测试X射线图像张成一个观测矩阵.最后,使用鲁棒PCA算法对观测矩阵进行低秩与稀疏分解,测试图像中的不均匀强度及噪声被消除,焊缝区域被凸显出来,通过全局阈值可将焊缝区域较好地分割出来.实验结果表明,该算法能较大地消除厚钢管X射线图像中噪声及不均匀强度分布带来的干扰、同时增强模糊的焊缝边缘及对比度低的区域,相比传统焊缝区域检测算法,具有更高的检测灵敏度(0.952)和精度(0.989),能更好地将模糊和对比度低的焊缝区域完整检测出来.     

一种基于图像分块特性的数字水印嵌入方法

一种基于图像分块和视觉特性的数字水印嵌入技术将水印信息嵌入到静止图像中，该方法应用双重DCT方法实现了较大的嵌入量和良好的隐蔽性。实验表明，这种方法能够抵御噪声干扰、线性滤波和JPEG压缩编码，具有良好的应用价值。     

一种基于运动背景的视频对象分割算法

随着MPEG-4和MPEG-7的研究发展,基于内容编码和面向对象的存取和交互技术日益得到人们的重视,视频分割技术正在成为当前视频研究领域的热点之一,但是,目前的分割研究大部分是在没有全局运动的情况下进行的,对于运动背景下视频对象的分割研究还不多.为此,提出了一种新的基于改进分水岭和光流的视频分割算法,即先将原始图像使用改进的分水岭算法标记成不同的灰度区域,然后以光流法得到的对象运动信息作为评判准则,将这些区域分别归类于前景对象和背景,达到从运动背景中分割出前景对象的目的.仿真实验表明,此算法能够较精确地分割出视频对象.     

基于多分辨率-分水岭算法的图像分割方法

提出了在小波多分辨率域中使用分水岭算法对图像进行区域分割和融合的新方法,该方法首先使用小波变换理论将原始图像变换为不同层次的金字塔多分辨率图像；然后通过分水岭算法获得最低分辨率下的分割图像；最后。利用逆小波变换将分割的图像映射到原始分辨率上．从而获得分割图像．实验结果表明：此方法可以大大减少噪声存在下的过分割现象．     

基于分水岭算法的双向凝胶电泳图像分割

采用分水岭算法对电泳凝胶图像进行分割,针对其存在的过分割问题以及图像本身的特点,提出了根据拓扑曲率对过分割区域进行合并,以得到有意义的分割结果。用本算法对电泳图像进行了分割实验,结果表明,用本算法适用于电泳图像,并且效果良好。     

剪切力环境下的血管平滑肌细胞图像分割技术研究

定量分析血管平滑肌细胞VSMCs(Vascular Smooth Muscle Cells)在剪切力作用下的形态变化有助于理解VSMCs的生长机制以及病理学研究.图像分割在图像分析中扮演着重要角色，但同时也是图像处理流程中最困难的步骤.针对培养在平板流室中的VSMCs相互交叠的细胞图像，作者首先分析了这类细胞图像的特征，随后提出了一种具有四个步骤的图像分割方法，分别是图像预处理、照明校正、数学形态学重建和分水岭转换.实验结果证明，按照这种方法分割细胞图像，取得了较好的效果.     

基于一种新模糊增强算子的图像边缘检测算法

借鉴广义模糊集的特性,提出了一种新的模糊边缘检测算法:利用一个简单的隶属度函数将空域图像转换到[0,1]的普通模糊域;然后通过一个简单的增强算子先将其扩展到[-1,2];由于一般情况下物体域(物体内部区域)主要以高灰度为主,背景域(物体背景区域)主要以低灰度为主,而过渡域(物体边缘区域)则以高梯度为主,本算法目的是对图像进行边缘检测,先不对它进行转换或截断,而将其转换到图像准灰度域,相当于把图像的物体域和背景域放到了[0,255]的区域以外;然后采用"min"或"max"算子进行所谓的边缘提取,最后将提取的"边缘"数据进行截断处理,从而将图像数据转换到图像的空间域,即图像的灰度域。从边缘检测结果来看,本算法更适合于低对比度、含有较精细部分以及纹理丰富的图像检测,且耗时较小。     

彩色图像分割综述

图像分割是进行图像理解的基础，是图像工程技术中的一个重要问题。近几年来，人们越来越重视彩色图像的分割算法，并期望寻求一种实时性、鲁棒性较好的算法，本总结了近几年出现的彩色图像分割技术进行了讨论。     

基于区域的自动种子区域生长法的彩色图像分割算法

提出了一种基于区域的自动种子区域生长法进行彩色图像分割的算法.该方法首先应用分水岭算法对图像进行初始化分割,形成过分割效果.然后从分水岭算法形成的区域出发,根据一定的规则自动选出一些区域作为种子区域,进行种子区域生长.与传统的种子区域生长法(SRG)以图像中的像素点作为种子进行生长的方法不同,本方法以区域作为种子并以区域作为生长单位进行区域生长.实验结果表明,该算法能够产生较好的分割效果及较快的分割速度.     

融合纹理特征和空间关系的TM影像海岸线自动提取

针对以往研究中简单运用影像单一特征(光谱特征或者空间关系)提取海岸线的不足, 在提取过程中逐步融入区域纹理特征、空间关系和经验知识, 提出一种融合纹理特征和空间关系的海岸线自动提取方法: 首先利用影像的水体指数和纹理特征, 对影像水、陆区域进行二值分割; 然后结合空间关系和经验知识, 对分割结果进行优化和后处理, 并生成矢量海岸线; 最后, 利用基于线匹配的岸线精度评价准则, 分析岸线提取结果。将该方法运用到辽东半岛复杂背景条件下TM影像的海岸线提取实验中, 取得了良好的实验效果, 证明纹理和空间关系在海岸线提取中具有重要作用。