基于分词的语句相似度计算的改进

语句相似度体现的是两个句子之间的相似程度。语句相似度计算是FAQ和信息检索等方面核心技术之一。针对基于分词的相似度计算过于依赖实际的分词效果,在原相似度计算模型中增加了两个句子不分词时的词形相似度计算,以缓解因为句子分词不准确而导致相似度计算结果偏低的情况。结合“数据结构”课程问答系统的实验,结果表明,改进的方法比原方法有较高的准确率。     

基于云模糊理论的图像纹理分割

为了处理图像纹理的模糊性和随机性,基于云模糊理论提出了纹理特征矢量云模型,并成功地应用于纹理图像分割.该方法在对纹理统计描述符模糊化处理后,逆向生成纹理特征矢量云.矢量云模型的数字特征能够很好地表达纹理的模糊性和随机性,据此通过云距离计算及纹理特征矢量云生长,完成对图像纹理的分割.实验结果表明,该方法较经典的ISODATA算法和K-means簇算法的分割精度高,并且迭代收敛速度快.     

Unsupervised spatio-temporal segmentation for extracting moving objects in video sequences

Extraction of moving objects is an important and fundamental research topic for many video ap-plications. This paper addresses an unsupervised spatio-temporal segmentation scheme to extract moving objects from video sequences. The temporal segmentation localizes moving objects by comparing the motion vector of each block in each frame with the corresponding global motion vector estimated by an outlier rejection (OR) based method. Furthermore, the temporal compensation utilizing the temporal coherence of moving objects is considered in the temporal segmentation to solve the temporarily stopping problem. The detected moving regions usually have discontinuous boundaries and some holes. These regions are then compensated in the spatial domain. In the spatial segmentation, the watershed algorithm considering the global information improves the accuracy of segmentation in the spatial domain. The modified mean filter is presented to suppress some minima. By using a fusion module, moving objects are extracted. Experiments on various sequences have successfully demonstrated the validity of the proposed scheme.     

基于文本挖掘的道路运输安全风险源辨识模型

为了解决当前道路运输安全风险源辨识工作中数据短缺和人员工作量较大的问题，从文本挖掘的角度出发，提出一种能够自动辨识道路运输过程中安全风险源的模型. 该模型首先对道路运输文本进行因果句提取，并对因果句进行分词操作，实现安全风险源特征的增强；其次，进行适应卷积神经网络（convolutional neural networks，CNN）输入的、包含词信息和位置信息的特征构造；然后，将特征构造的结果输入到CNN实现安全风险源的辨识；最后，利用道路交通事故报告进行实验. 实验结果表明:提出的辨识模型能辨识大部分的道路运输安全风险源因素，准确率约为77.321%.      

Segmentation of haustral folds and polyps on haustral folds in CT colonography using complementary geodesic distance transformation

Computer-aided detection （CAD） for CT colonography refers to a scheme that automatically detects polyps in CT images of colon. Current CAD schemes already have a relatively high sensitivity and a low false positive rate. However, misdiagnosis and missed diagnosis are still common to happen, mainly due to the existence of haustral folds （HFs）. An innovative idea of segmenting semilunar HFs from the smooth colonic wall and then using different methods to detect polyps on HFs and those on the smooth colonic wall is proposed in this paper to reduce the false positives and false negatives caused by HFs. For the polyps on HFs, a novel segmentation method is specially developed based on complementary geodesic distance transformation （CGDT）. The proposed method is tested on four different models and real CT data, The property of CGDT is proved and our method turns out to be effective for HF segmentation and polyp segmentation. The encouraging experimental results primarily show the feasibility of the proposed method and its potential to improve the detection performance of CAD schemes.     

细胞图像自动分割方法研究

提出一种细胞图像的自动分割与定位方法．首先将直方图分割所得二值图像的距离变换作为边缘提取的指导知识，进行细胞图像的精确分割；然后利用分割边缘点的梯度方向和几何知识确定细胞的中心．该方法既能对非重叠细胞也能对重叠细胞进行细胞的分割与定位．     

FULLY AUTOMATIC FRAMEWORK FOR SEGMENTATION OF BRAIN MRI IMAGE

Automaticbraintissuesegmentationfrommag neticresonanceimages(MRI)isofgreatimportance forresearchandclinicalstudyofmuchneurological pathology.Duringthepastdecade,theMRIhashad agreatimpactonthediagnosticimagingofmosthu manorgansystem.Thesegmentationofbrai…     

Automatic identification of cracks from borehole image under complicated geological conditions

Identifying cracks from the spread image of a borehole wall is one of the most common usages of borehole imaging method. The manual identification of cracks is time-consuming and can be easily influenced by objective judgment. In this study, firstly, the image translation from RGB color model to HSV color model is done to highlight the structural plane region, which is closer to the color recognition of human sight; secondly, the Saturation component is filtered for further processing and a twice segmentation method is proposed to improve the accuracy of automatic identification. The primary segmentation is based on the statistics of saturation over a longer borehole section and can give a rough estimation of a crack. Then, the pixels are shifted in the reverse direction to the sine curve estimated and make the centerline of the crack flat. Based on the shifted image, the secondary segmentation is done with a small rectangle region that takes the baseline of the roughly estimated crack as its centerline. The result of the secondary segmentation can give a correction to the first estimation. Through verifying this method with actual borehole image data, the result has shown that this method can identify cracks automatically under very complicated geological conditions.     

噪声符合高斯分布的曲面图像分割

针对噪声符合高斯分布的曲面图像分割问题,提出了一种基于变分水平集方法的隐式曲面上噪声符合高斯分布的图像分割模型。为定义隐式曲面上图像区域划分的动态轮廓线,引入另一个动态水平集函数,其零水平集与隐式曲面的交线即为曲面上区域划分的轮廓线。该模型采用N个水平集函数划分2^N个区域的区域竞争策略。通过变分方法得到图像分割空间轮廓线演化的梯度降方程。通过显式差分格式对演化方程进行离散。实验结果表明,该模型对曲面上噪声符合高斯分布的图像具有良好的分割效果,并且在图像分割的同时能够完全保持曲面固有的几何特征。     

高分辨率遥感影像分割提取构筑物评估

面向对象的影像分析已成为针对高分辨率遥感影像信息提取和行业应用的一种先进方法．该方法的实用效果,基础和关键在于对影像中所关注地物目标的影像分割提取效果．针对高分辨率遥感影像地面杓筑物的提取,采用经验性符合统计比较法,把自动分割结果与人工分割提取的地面构筑物区域轮廓相对符合程度分为3个等级：良好、可以接受和不可接受,对当前应用广泛的两种多尺度影像分割方法进行了地面构筑物分割提取效果评估．     

基于球面图像的三角网格模型分割

为了从局部形状的角度实现对三角网格模型的管理和重用,提出了一种基于球面图像的三角网格模型分割方法.通过球面参数化及球面划分,将三角网格模型的表面属性信息映射到球面图像中;利用成熟的区域生长、区域合并图像分割算法对球面图像进行分割;将球面图像的分割结果转换为三角网格模型的分割结果.实验结果表明:该分割方法可以对不均匀的低分辨率三角网格模型进行有效分割,降低了几何属性估算对分割结果的影响,不会发生过分割现象,不需要进行分割的后续处理.     

融合边缘检测与区域生长的交通图像分割方法

在交通监控中,如何从复杂的背景中分割运动物体是至关重要的一步,针对车辆的运动阴影对图像分割产生的不利影响,提出了一种新的融合边缘检测与区域生长的彩色图像分割算法,算法同时考虑了图像的彩色信息和空间信息.该算法首先对彩色图像边缘检测,并根据检测结果设置种子像素;再基于颜色相似性生长准则,结合边缘检测结果,对每个种子点进行区域生长;最后,利用区域合并算法对剩余的像素进行合并.实验结果表明该算法很大程度上克服了阴影给图像分割带来的不利影响.     

双注意力引导的跨层优化交通场景语义分割

针对交通场景目标分割边缘不平滑以及小目标难以准确分割等问题，本文提出一种双注意力引导的跨层优化交通场景语义分割算法。首先，构建多分支特征提取编码网络，并利用串行非比例式空洞卷积实现空间上下文信息提取，进而改善小目标信息的丢失；其次，构建基于空间对齐的跨层特征融合解码网络，实现语义信息和细节信息的融合，增强不同尺度目标的表达能力；最后，提出通道和空间注意力机制，建模全局通道相关性和长距离位置相关性，增强网络对关键特征的学习能力。交通场景数据集Cityscapes和CamVid上的实验结果表明，所提特征提取编码网络、跨层特征融合解码网络以及注意力机制模块是有效的；所提语义分割算法获得了77.79%和78.66%的平均交并比，能够平滑目标分割边缘，尤其对细长条形目标具有鲁棒性。     

交际翻译理论与中国菜名翻译

本文以纽马克的文本类型和交际翻译理论为基础,探讨中国菜名的翻译策略。笔者认为,菜名翻译具有信息传递,文化交流和审美价值三大功能,交际翻译理论对中国菜名的翻译具有很强的适用性和指导意义。     

A Bintree Energy Approach for Colour Image Segmentation Using Adaptive Channel Selection

A new hierarchical approach called bintree energy segmentation was presented for color image seg-mentation. The image features are extracted by adaptive clustering on multi-channel data at each level and used as the criteria to dynamically select the best chromatic channel, where the segmentation is carried out. In this approach, an extended direct energy computation method based on the Chan-Vese model was proposed to segment the selected channel, and the segmentation outputs are then fused with other channels into new images,from which a new channel with better features is selected for the second round segmentation. This procedure is repeated until the preset condition is met. Finally, a binary segmentation tree is formed, in which each leaf represents a class of objects with a distinctive color. To facilitate the data organization, image background is employed in segmentation and channels fusion. The bintree energy segmentation exploits color information involved in all channels data and tries to optimize the global segmentation result by choosing the "best" chan-nel for segmentation at each level. The experiments show that the method is effective in speed, accuracy and flexibility.     

基于离散水平集方法的Chan-Vese模型

采用离散形式的水平集函数代替传统的符号距离函数,提出基于离散水平集方法的Chan-Vese模型。采用变分方法得到水平集函数演化的梯度降方程,对得到的演化方程进行半隐式离散求解和数值算例验证,并与传统的符号距离函数方法进行对比,以验证离散水平集方法在图像分割中的优势。实验结果表明,基于离散水平集方法的Chan-Vese模型分割速度和精度比用符号距离函数实现的Chan-Vese模型有明显提高。     

Image series segmentation and improved MC algorithm

A semiautomatic segmentation method based on active contour is proposed for computed tomography （CT） image series. First, to get initial contour, one image slice was segmented exactly by C-V method based on Mumford-Shah model. Next, the computer will segment the nearby slice automatically using the snake model one by one. During segmenting of image slices, former slice boundary, as next slice initial contour, may cross over next slice real boundary and never return to right position. To avoid contour skipping over, the distance variance between two slices is evaluated by an threshold, which decides whether to initiate again. Moreover, a new improved marching cubes （MC） algorithm based on 2D images series segmentation boundary is given for 3D image reconstruction. Compared with the standard method, the proposed algorithm reduces detecting time and needs less storing memory. The effectiveness and capabilities of the algorithm were illustrated by ,experimental results.     

Automatic Airway Deletion in Pulmonary Segmentation

A method of removing the airway from pulmonary segmentation image was proposed. This method firstly segments the image into several separate regions based on the optimum threshold and morphological operator,and then each region is labeled and noted with its mean grayscale. Therefore, most of the non-lung regions can be removed according to the tissue‘s Hounsfield units (HU) and the imaging modality. Fifially, the airway region is recognized and deleted automatically through using the priori information of its HU and size. This proposed method is tested using several clinical images, yielding satisfying results.