递归趋势分析在汉语语音声韵母切分中的应用研究

基于隐马尔可夫模型(HMM)的连续语音自动切分方法由于较高的切分精度得到了广泛的应用,然而其切分结果还不能够直接应用于基于脚本的语音拼接合成系统,需要音素边界的再调整。本文分析了不同的汉语语音音素的非线性动力学物理模型在其递归图(RP)上的表现,通过递归趋势(RT)这一衡量系统稳定性程度的量化参数,揭示了语音产生过程中的不稳定性。结合基于HMM的连续语音初始切分结果,从定位语音动力学特性突变点的角度,调整声韵母切分边界,在10、20、30毫秒基准范围内,切分精度分别提高了13.88%、4.19%、3.19%。     

轴静强度的非概率可靠性设计

基于不确定非概率凸模型，提出了轴静强度的非概率可靠性设计方法。将具有不确定且有边界限制的应力、载荷等用凸模型中的区间模型来描述，取在标准化区间变量的扩展空间中，从坐标原点到失效面的最短距离作为非概率可靠性指标。该方法可避免求概率密度分布函数，对初始数据要求较低，计算简便。最后给出了计算实例。     

一种改进的基于Viterbi的语音切分算法

主要针对文本提示型说话人识别中语音切分高精确度要求的问题,在利用Viterbi算法的语音切分基础上,提出了向后平滑搜索多帧能量极小值的语音切分方法。该算法首先对0~9的每个数字建立模型,然后利用Viterbi算法对随机数字串进行切分得到初始切分点,最后利用搜索多帧能量极小值的方法更新原始切分点。实验表明,相比于传统的切分算法,在误差范围小于20 ms之内,改进算法的切分准确率由82.1%提高到88%。     

基于图片边界后门嵌入的图像识别攻击研究

采用公开数据集或预训练好的神经网络模型可以快速实现图像分析、语音识别等应用，但存在一定的风险或威胁。攻击者可以通过向开源训练数据或者训练模型中嵌入后门，使模型在接收到带有触发功能的数据时执行指定的后门行为。目前，图像识别的后门攻击采用的后门触发器大多在视觉上容易被发现，为此，文中提出一种基于图片边界后门嵌入的图像识别攻击方法。该方法向训练图片边界添加窄的有色带作为后门触发器，利用隐蔽的外形逃避视觉关注。在MNIST、CIFAR-10等图像识别数据集上对所提方法进行测试，实验结果表明，该后门可以稳定注入，并在毒药率为30%时，攻击成功率达到99.73%。相比于其他两种常见的后门攻击方法，所提方法攻击成功率更高，具有较强的攻击性和鲁棒性。     

英语发音自动评测中单词删除错误的检测方法

基于HMM后验概率分数的自动发音评测方法中,强制匹配获得语音的音素级切分信息是重要的一步。切分结果是否准确直接影响到后验概率分数的可靠性。通过对单词发音网络增加可跨越分支,实现单删除错误检测功能,可降低说话人删除错误对前后单词切分准确性的影响。实验结果显示,单词删除错误检出率达到87．8％。     

基于笔划提取和合并的离线手写体汉字字符切分算法

手写体汉字字符切分是离线汉字字符识别预处理中的一个重要问题，针对离线手写体汉字，提出一种基于笔划提取和合并的手写体汉字字符切分算法。该算法首先基于方向游程提取汉字的笔划，并建立笔划框，再根据汉字笔划的结构知识对笔划框进行合并，得到字符的切分结果。该算法能较好地解决粘连汉字字符的切分问题，对从现场随机采集的2500封手写体信函地址汉字进行切分，单字正确率可达91．5％。     

基于支持向量机的Web文本分类方法

Web文本分类技术是数据挖掘中一个研究热点领域，而支持向量机又是一种高效的分类识别方法，在解决高维模式识别问题中表现出许多特有的优势。文章通过分析Web文本的特点，研究了向量空间模型（VSM）的分类方法和核函数的选取，在此基础上结合决策树方法提出了一种基于决策树支持向量机的Web文本分类模型。并给出具体的算法。通过实验测试表明，该方法训练数据规模大大减少，训练效率较高，同时具有较好的精确率（90．11％）和召回率（89．38％）。     

基于SVM和CRF双层分类器的英文电子病历去隐私化

去隐私化是2014 i2b2/UTHealth中的一个任务,目的在于识别并移除电子病历中的隐私信息.本文提出了一种基于支持向量机(SVMs)和条件随机场(CRFs)双层分类模型的去隐私化方法,经过预处理将病历文本进行词切分(tokenize)处理,并在此基础上抽取4类特征,训练SVM模型对隐私信息实体边界进行划分并将结果作为特征添加到特征集中,通过CRF训练多分类器,并通过该分类器对各个类别的隐私信息进行识别.实验表明双层分类模型对于隐私信息识别是有效的,结果F值达到0.9110.     

一种新的基于直接最小二乘椭圆拟合的肤色检测方法

肤色检测是计算机视觉中的一个重要问题,本文提出了一种新的基于直接最小二乘椭圆拟合的肤色检测方法,其基本思想是根据肤色样本分布区域的边界数据点采用曲线拟合的方法得到肤色分布区域的边界方程。在实现时,为了解决直接在笛卡儿坐标系中提取肤色样本分布区域边界数据的困难,算法采用了一种新的解决思路,即首先把训练肤色样本在色度空间的统计分布转化为图像的形式,然后再利用边缘检测方法得到肤色分布区域的边界数据。根据所得的边界数据点用直接最小二乘椭圆拟合方法便可得到肤色分布区域的椭圆边界,方法简单直观。实践表明,该算法能完成对各种不同环境条件下所拍摄图像的肤色分割,效果理想,其性能明显优于常用的域值界定法和单高斯模型法。     

基于状态码本的准连续隐马尔可夫模型

本文针对经典HMM模型对训练数据要求多且算法复杂的问题，提出了一种改进的模型一基于状态码本的准连续HMM模型（SCBHMM），该模型在有限训练数据的条件下能更加有效地描述语音信号的声学特征．通过将状态转移概率与动态谱变化量相关联，使得SCBHMM能有效地将语音信号的静态特征和动态特征相结合．通过在标准语音数据库USTC94上的大量实验表明了SCBHMM在汉语音节识别中的有效性，它缓减了模型对训练数据的要求，并大大降低了训练、识别的计算量，但同样取得了相当高的识别率．     

Multiresolution segmentation of respiratory electromyographicsignals

Analysis of respiratory electromyographic (EMG) signals in the study of respiratory control requires the detection of burst activity from background (signal segmentation), and focuses upon the determination of onset and cessation points of the burst activity (boundary estimation). The authors describe a new automated multiresolution technique for signal segmentation and boundary estimation. During signal segmentation, a new transitional segment is defined which contains the boundary between background a burst activity. Boundary estimation is then performed within this transitional segment. Boundary candidates are selected and a probability is attributed to each candidate, using an artificial neural network. The final boundary for a given transitional segment is the boundary estimate with the maximum a posteriori probability. This new method has proved accurate when compared to boundaries chosen by two investigators     

基于级联Dense-UNet和图割的肝脏肿瘤自动分割

腹部CT图像肝脏肿瘤分割是进行肝脏疾病诊断、手术规划和放射治疗的重要前提。针对肝脏肿瘤灰度异质、纹理丰富、边界模糊等因素引起的分割困难,该文提出基于级联Dense-Unet和图割的自动精确鲁棒分割方法。首先运用级联的Dense-UNet获取肝脏肿瘤初始分割结果及感兴趣区域,然后利用图像像素级和区域级特征,分别构建可有效区分肿瘤与非肿瘤的灰度模型和概率模型,并将其融入图割能量函数,进一步精确分割感兴趣区域中的肿瘤组织。最后分别采用LiTS和3Dircadb公共数据库作为训练集与测试集进行实验,并与现有多种自动分割方法进行了比较。结果表明,提出方法可有效分割CT图像中灰度、形状、大小、位置各异的肝脏肿瘤,能提取更精确的肿瘤边界,尤其对于对比度低、边界模糊的肿瘤具有明显优势。     

A segmentation model using compound Markov random fields based on a boundary model.

Markov random field (MRF) theory has been widely applied to the challenging problem of image segmentation. In this paper, we propose a new nontexture segmentation model using compound MRFs, in which the original label MRF is coupled with a new boundary MRF to help improve the segmentation performance. The boundary model is relatively general and does not need prior training on boundary patterns. Unlike some existing related work, the proposed method offers a more compact interaction between label and boundary MRFs. Furthermore, our boundary model systematically takes into account all the possible scenarios of a single edge existing in a 3 x 3 neighborhood and, thus, incorporates sophisticated prior information about the relation between label and boundary. It is experimentally shown that the proposed model can segment objects with complex boundaries and at the same time is able to work under noise corruption. The new method has been applied to medical image segmentation. Experiments on synthetic images and real clinical datasets show that the proposed model is able to produce more accurate segmentation results and satisfactorily keep the delicate boundary. It is also less sensitive to noise in both high and low signal-to-noise ratio regions than some of the existing models in common use.     

Extraction and Regularization of Various Building Boundaries with Complex Shapes Utilizing Distribution Characteristics of Airborne LIDAR Points

This study presents an approach for extracting boundaries of various buildings, which have concave boundaries, inner yards, non‐right‐angled corners, and nonlinear edges. The approach comprises four steps: building point segmentation, boundary tracing, boundary grouping, and regularization. In the second and third steps, conventional algorithms are improved for more accurate boundary extraction, and in the final step, a new algorithm is presented to extract nonlinear edges. The unique characteristics of airborne light detection and ranging (LIDAR) data are considered in some steps. The performance and practicality of the presented algorithm were evaluated for buildings of various shapes, and the average omission and commission error of building polygon areas were 0.038 and 0.033, respectively     

改进边界指示函数的水平集活动轮廓模型

由于受成像原理的限制,导致超声图像对比度低、边界模糊,因此基于边界的水平集分割效果很不理想。为了提高超声图像的分割精度和分割效率,提出了一种梯度信息与区域信息相结合的水平集分割算法。首先对基于边界的距离正则化水平集演化(DRLSE)模型进行改进,将区域信息引入到边界指示函数中,并用改进后的边界指示函数代替DRLSE模型中的边界指示函数,最后,得到一个梯度与区域信息相结合的水平集演化模型。结果表明,本文中的模型能准确分割甲状腺肿瘤超声图像,且在分割效率和分割精确度方面均比DRLSE模型有所提高。     

An ultra-fast user-steered image segmentation paradigm: live wire on the fly

We have been developing general user steered image segmentation strategies for routine use in applications involving a large number of data sets. In the past, we have presented three segmentation paradigms: live wire, live lane, and a three-dimensional (3-D) extension of the live-wire method. In this paper, we introduce an ultra-fast live-wire method, referred to as live wire on the fly, for further reducing user's time compared to the basic live-wire method. In live wire, 3-D/four-dimensional (4-D) object boundaries are segmented in a slice-by-slice fashion. To segment a two-dimensional (2-D) boundary, the user initially picks a point on the boundary and all possible minimum-cost paths from this point to all other points in the image are computed via Dijkstra's algorithm. Subsequently, a live wire is displayed in real time from the initial point to any subsequent position taken by the cursor. If the cursor is close to the desired boundary, the live wire snaps on to the boundary. The cursor is then deposited and a new live-wire segment is found next. The entire 2-D boundary is specified via a set of live-wire segments in this fashion. A drawback of this method is that the speed of optimal path computation depends on image size. On modestly powered computers, for images of even modest size, some sluggishness appears in user interaction, which reduces the overall segmentation efficiency. In this work, we solve this problem by exploiting some known properties of graphs to avoid unnecessary minimum-cost path computation during segmentation. In live wire on the fly, when the user selects a point on the boundary the live-wire segment is computed and displayed in real time from the selected point to any subsequent position of the cursor in the image, even for large images and even on low-powered computers. Based on 492 tracing experiments from an actual medical application, we demonstrate that live wire on the fly is 1.3-31 times faster than live wire for actual segmentation for varying image sizes, although the pure computational part alone is found to be about 120 times faster.     

A novel iris segmentation using radial-suppression edge detection

Iris segmentation is a key step in the iris recognition system. The conventional methods of iris segmentation are based on the assumption that the inner and outer boundaries of an iris can be taken as circles. The region of the iris is segmented by detecting the circular inner and outer boundaries. However, we investigate the iris boundaries in the CASIA-IrisV3 database, and find that the actual iris boundaries are not always circular. In order to solve this problem, a new approach for iris segmentation based on radial-suppression edge detection is proposed in this paper. In the radial-suppression edge detection, a non-separable wavelet transform is used to extract the wavelet transform modulus of the iris image. Then, a new method of radial non-maxima suppression is proposed to retain the annular edges and simultaneously remove the radial edges. Next, a thresholding operation is utilized to remove the isolated edges and produce the final binary edge map. Based on the binary edge map, a self-adaptive method of iris boundary detection is proposed to produce final iris boundaries. Experimental results demonstrate that the proposed iris segmentation is desirable.     

Combining static and dynamic features using neural networks and edge fusion for video object extraction

Semantic object representation is an important step for digital multimedia applications such as object-based coding, content-based access and manipulations. The authors propose an image sequence segmentation scheme which provides region information for the semantic object representation of those applications. The objective is to develop a hardware-friendly segmentation algorithm by combining static and dynamic features simultaneously in one scheme. In the initial stage, a multiple feature space is transformed to one-dimensional label space by using self-organising feature map (SOFM) neural networks. The next stage is an edge fusion process in which edge information is incorporated into the neural network outputs to generate more precisely located boundaries of segmentation. The proposed algorithm differs from existing methods as follows: it can segment textured images with low-dimensional features; leads to more meaningful segmentation region boundaries; and is easier to map into hardware than existing methods. Experimental results are compared with an existing segmentation method using evaluation metrics to clarify the advantages of the proposed algorithm objectively.     

基于改进Mask R-CNN的模糊图像实例分割的研究

Mask R-CNN是现阶段实例分割相对成熟的方法,针对Mask R-CNN算法当中还存在的分割边界精度以及对于模糊图片鲁棒性较差等问题,该文提出一种基于改进的Mask R-CNN实例分割方法。该方法首先提出在Mask分支上使用卷积化条件随机场(ConvCRF)来优化Mask分支对于候选区域进一步分割,并使用FCN-ConvCRF分支来代替原有分支;之后提出新锚点大小和IOU标准,使得RPN候选框能够涵盖所有实例区域;最后使用一种添加部分经过转换网络转换的数据进行训练的方法。总的mAP值与原算法相比提升了3%,并且分割边界精确度和鲁棒性都有一定提高。     

基于softmax回归与图割法的脑肿瘤分割算法

从医学图像中分割脑肿瘤区域可以为脑肿瘤的诊断以及放射治疗提供帮助.但肿瘤区域的变化异常且边界非常模糊,因此自动或半自动地分割脑肿瘤非常困难.针对这一问题,本文结合softmax回归和图割法提出一种脑肿瘤分割算法.首先融合多序列核磁共振图像(MRI)并标记训练样本,再用softmax回归训练模型参数并计算每个点属于各个类别的概率,最后将概率融入到图割法中,用最小切/最大流方法得到最终分割结果.实验表明提出的方法可以更好地得到脑肿瘤的边界,并能较准确地分割出脑肿瘤区域.