Hierarchical classification and feature reduction for fast face detection with support vector machines

We present a two-step method to speed-up object detection systems in computer vision that use support vector machines as classifiers. In the first step we build a hierarchy of classifiers. On the bottom level, a simple and fast linear classifier analyzes the whole image and rejects large parts of the background. On the top level, a slower but more accurate classifier performs the final detection. We propose a new method for automatically building and training a hierarchy of classifiers. In the second step we apply feature reduction to the top level classifier by choosing relevant image features according to a measure derived from statistical learning theory. Experiments with a face detection system show that combining feature reduction with hierarchical classification leads to a speed-up by a factor of 335 with similar classification performance.     

面向高考阅读理解鉴赏题语言风格判别方法

语言风格是高考阅读理解中的重要考察内容,然而不同考察方式所需的分类层次不尽相同,该文将语言风格鉴赏转化为层次分类问题。在类别标签指导下,利用图分割算法,获取与特定类别相对应的原始簇。基于原始簇,利用层次聚类获取语言风格类别层次结构,之后结合层次结构训练SVM层次分类器。在解答语言风格鉴赏题过程中,依据阅读理解题干确定所需分类层次,利用SVM层次分类器完成对阅读材料语言风格判别,最后结合知识库生成语言风格鉴赏题答案。实验结果表明,基于层次结构的语言风格判别方法,可以为高考鉴赏类考题的解答提供技术支撑。     

2D级联CNN模型的放疗危及器官自动分割

目的 精准的危及器官（organs at risk,OARs）勾画是肿瘤放射治疗过程中的关键步骤。依赖人工的勾画方式不仅耗费时力,且勾画精度容易受图像质量及医生主观经验等因素的影响。本文提出了一种2D级联卷积神经网络（convolutional neural network,CNN）模型,用于放疗危及器官的自动分割。方法 模型主要包含分类器和分割网络两部分。分类器以VGG（visual geometry group）16为骨干结构,通过减少卷积层以及加入全局池化极大地降低了参数量和计算复杂度;分割网络则是以U-Net为基础,用双线性插值代替反卷积对特征图进行上采样,并引入Dropout层来缓解过拟合问题。在预测阶段,先利用分类器从输入图像中筛选出包含指定器官的切片,然后使用分割网络对选定切片进行分割,最后使用移除小连通域等方法对分割结果进一步优化。结果 本文所用数据集共包含89例宫颈癌患者的腹盆腔CT（computed tomography）图像,并以中国科学技术大学附属第一医院多位放射医师提供的手工勾画结果作为评估的金标准。在实验部分,本文提出的分类器在6种危及器官（左右股骨、左右股骨头、膀胱和直肠）上的平均分类精度、查准率、召回率和F1-Score分别为98.36%、96.64%、94.1%和95.34%。基于上述分类性能,本文分割方法在测试集上的平均Dice系数为92.94%。结论 与已有的CNN分割模型相比,本文方法获得了最佳的分割性能,先分类再分割的策略能够有效地避免标注稀疏问题并减少假阳性分割结果。此外,本文方法与专业放射医师在分割结果上具有良好的一致性,有助于在临床中实现更准确、快速的危及器官分割。     

A statistical method for automatic labeling of tissues in medical images

A statistically based segmentation method was applied to the recognition of organs in abdominal CT scans. To incorporate prior knowledge of anatomical structure, a stochastic model was used that represented abdominal geometry in three dimensions. Properties of each tissue class with respect to X-ray imaging were modeled by mean grey-value distributions. Twelve different tissues were labeled simultaneously. Deterministic maximization of the a posteriori distribution as well as stochastic optimization by simulated annealing were both applied. Mean segmentation results were determined for a set of 18 scan sequences, using a set of reference contours designated by a radiologist as ground truth. Results appeared to be somewhat better using simulated annealing. The proposed segmentation method, which is fast and fully automatic, seems sufficiently accurate for many clinical applications, such as determination of relative organ volumes.     

DSP-based hierarchical neural network modulation signal classification

This paper discusses a real-time digital signal processor (DSP)-based hierarchical neural network classifier capable of classifying both analog and digital modulation signals. A high-performance DSP processor, namely the TMS320C6701, is utilized to implement different kinds of classifiers including a hierarchical neural network classifier. A total of 31 statistical signal features are extracted and used to classify 11 modulation signals plus white noise. The modulation signals include CW, AM, FM, SSB, FSK2, FSK4, PSK2, PSK4, OOK, QAM16, and QAM32. A classification hierarchy is introduced and the genetic algorithm is employed to obtain the most effective set of features at each level of the hierarchy. The classification results and the number of operations on the DSP processor indicate the effectiveness of the introduced hierarchical neural network classifier in terms of both classification rate and processing time.     

基于级联Vnet-S网络的CT影像单一器官自动分割算法

为了快速准确地对计算机断层扫描（CT）影像中的器官进行分割，提出基于级联Vnet-S网络的单一器官自动分割算法。首先，使用第一个Vnet-S网络对CT影像中的器官进行粗分割；然后，选择分割结果中的最大连接通量做两次膨胀，根据膨胀后的最大连接通量确定器官边界并提取器官区域；最后，使用第二个Vnet-S网络对器官进行细分割。为了验证算法的性能，采用MICCAI 2017 Liver Tumor Segmentation Challenge （LiTS）数据集进行肝脏分割实验，采用ISBI LUng Nodule Analysis 2016（LUNA16）数据集进行肺分割实验。级联Vnet-S算法在LiTS的70例线上测试数据上的Dice系数为0.9600，在LUNA16的288例测试数据上的Dice系数为0.9810，均高于Vnet-S网络和Vnet网络。实验结果表明，基于级联Vnet-S网络的单一器官分割算法可以准确地对器官进行分割，而且级联Vnet-S算法的计算量小于Unet网络和Vnet网络。     

Hierarchical classification of protein function with ensembles of rules and particle swarm optimisation

This paper focuses on hierarchical classification problems where the classes to be predicted are organized in the form of a tree. The standard top-down divide and conquer approach for hierarchical classification consists of building a hierarchy of classifiers where a classifier is built for each internal (non-leaf) node in the class tree. Each classifier discriminates only between its child classes. After the tree of classifiers is built, the system uses them to classify test examples one class level at a time, so that when the example is assigned a class at a given level, only the child classes need to be considered at the next level. This approach has the drawback that, if a test example is misclassified at a certain class level, it will be misclassified at deeper levels too. In this paper we propose hierarchical classification methods to mitigate this drawback. More precisely, we propose a method called hierarchical ensemble of hierarchical rule sets (HEHRS), where different ensembles are built at different levels in the class tree and each ensemble consists of different rule sets built from training examples at different levels of the class tree. We also use a particle swarm optimisation (PSO) algorithm to optimise the rule weights used by HEHRS to combine the predictions of different rules into a class to be assigned to a given test example. In addition, we propose a variant of a method to mitigate the aforementioned drawback of top-down classification. These three types of methods are compared against the standard top-down hierarchical classification method in six challenging bioinformatics datasets, involving the prediction of protein function. Overall HEHRS with the rule weights optimised by the PSO algorithm obtains the best predictive accuracy out of the four types of hierarchical classification method.     

A semi-dependent decomposition approach to learn hierarchical classifiers

In hierarchical classification, classes are arranged in a hierarchy represented by a tree or a forest, and each example is labeled with a set of classes located on paths from roots to leaves or internal nodes. In other words, both multiple and partial paths are allowed. A straightforward approach to learn a hierarchical classifier, usually used as a baseline method, consists in learning one binary classifier for each node of the hierarchy; the hierarchical classifier is then obtained using a top-down evaluation procedure. The main drawback of this naive approach is that these binary classifiers are constructed independently, when it is clear that there are dependencies between them that are motivated by the hierarchy and the evaluation procedure employed. In this paper, we present a new decomposition method in which each node classifier is built taking into account other classifiers, its descendants, and the loss function used to measure the goodness of hierarchical classifiers. Following a bottom-up learning strategy, the idea is to optimize the loss function at every subtree assuming that all classifiers are known except the one at the root. Experimental results show that the proposed approach has accuracies comparable to state-of-the-art hierarchical algorithms and is better than the naive baseline method described above. Moreover, the benefits of our proposal include the possibility of parallel implementations, as well as the use of all available well-known techniques to tune binary classification SVMs.     

医学图像图深度学习分割算法综述

精准分割医学图像中的器官或病灶,是医学图像智能分析领域的重要难题,其在临床上对于疾病的辅助诊疗有着重要应用价值。在解决医学图像信息表征及对非欧空间生理组织结构准确建模等挑战性问题方面,基于图深度学习的医学图像分割技术取得了重要突破,展现出显著的信息特征提取及表征优势,可获得更为精准的分割结果,已成为该领域新兴研究热点。为更好促进医学图像图深度学习分割算法的研究发展,对该领域的技术进展及应用现状做了系统的梳理总结。介绍了图的定义及图卷积网络的基本结构,详细阐述了谱图卷积和空域图卷积操作。根据GCN结合残差模块、注意力机制模块及学习模块三种技术结构模式,归纳并总结了其在医学图像分割中的研究进展。对图深度学习算法在医学图像分割领域的应用和发展做了概要总结和展望,为该领域的技术发展提供参考和新的研究思路。     

Calibrating level set approach by granular computing in computed tomography abdominal organs segmentation

Hybrid methodology that combines granular computing with a level set approach for image segmentation is introduced in this paper. The goal of the study is to provide an adjustable semi-automated method for 3D segmentation of abdominal organs in computed tomography studies. Based on initial guidance in terms of models or seed points the information granule is formulated for each anatomical structure under consideration. The granulation leads via spatial volume resampling and granule-driven image intensity fuzzification to the final segmentation stage employing a hybrid level set approach. Depending on the organ intensity and shape the segmentation runs with selected parameters. The algorithm has been evaluated by 20 computed tomography studies and four various structures delineated by an expert. The experimental part presents a comprehensive analysis of subsequent stages of the algorithm. The voxel-wise segmentation sensitivities reached 94.4% for the right kidney, 93.8% for the left kidney, 94.0% for the spleen, and 94.0% for the liver, with the Dice index at 93.7%, 94.2%, 91.0%, and 92.9%, respectively.     

Development of an IKONOS image classification rule-set for multi-scale mapping of Mediterranean rural landscapes

In this research, a rule-set of object-based classification of IKONOS imagery for fine-scale mapping of Mediterranean rural landscapes was developed. This study was conducted on the Mediterranean island of Crete (Greece). A three-level classification hierarchy was designed in a bottom-up approach containing a total number of 22 classes. The first level was associated with vegetation physiognomy (6 classes), the second level with linear features (6 classes) and the third level with land uses existing in the area (10 classes). Image objects were created with multiresolution segmentation, an algorithm supplied by eCognition software. The segmentation parameters were selected through a trial-and-error approach after visual evaluation of the resulting image objects. The rule-set comprised 100 classification rules described with the ‘Membership Function’ classifier. The classification stability was found to lie between 0.59 and 0.77, inversely proportional to the complexity of each level's classification. For an accuracy assessment, the error matrix method was used in a set of 250 randomly selected points. The overall classification accuracy achieved at the first level was 74%, at the second level 50% and at the third level 64%. The geometric accuracy of the classification was beyond the scope of this research; and moreover, consistent reference data sets were not available. The conclusion is that the use of rules in an object-based image analysis (OBIA) process has the potential to produce accurate landscape maps even in the case of complex environments, in which ancillary data are not available. Future work should focus on testing the transferability of the rule-set in different Mediterranean study sites, in order to draw a conclusion in relation to its potential operational use.     

Multidimensional hierarchical classification

Hierarchical classification can be seen as a multidimensional classification problem where the objective is to predict a class, or set of classes, according to a taxonomy. There have been different proposals for hierarchical classification, including local and global approaches. Local approaches can suffer from the inconsistency problem, that is, if a local classifier has a wrong prediction, the error propagates down the hierarchy. Global approaches tend to produce more complex models. In this paper, we propose an alternative approach inspired in multidimensional classification. It starts by building a multi-class classifier per each parent node in the hierarchy. In the classification phase, all the local classifiers are applied simultaneously to each instance, providing a probability for each class in the taxonomy. Then the probability of the subset of classes, for each path in the hierarchy, is obtained by combining the local classifiers results. The path with highest probability is returned as the result for all the levels in the hierarchy. As an extension of the proposal method, we also developed a new technique, based on information gain, to classifies at different levels in the hierarchy. The proposed method was tested on different hierarchical classification data sets and was compared against state-of-the-art methods, resulting in superior predictive performance and/or efficiency to the other approaches in all the datasets.     

基于多分辨统计模型和曲面恢复的腹部图像分割算法

针对腹部器官边缘模糊、形状差异大、小样本集合难建立统计模型等问题,提出了基于多分辨率统计集成模型和曲面缺失数据恢复的混合图像分割算法。该算法根据器官模型的纹理特征,建立外观轮廓模型;并定义标志点自信度。对于自信度较高的点,使用基于主动图像搜索和模型变形的方法进行分割;将自信度较低的点视为未知点,利用统计模型和自信度高的已知点进行数据恢复。实验结果表明,该混合算法可成功地降低器官分割的平均误差。     

Recursive segmentation and recognition templates for image parsing

In this paper, we propose a Hierarchical Image Model (HIM) which parses images to perform segmentation and object recognition. The HIM represents the image recursively by segmentation and recognition templates at multiple levels of the hierarchy. This has advantages for representation, inference, and learning. First, the HIM has a coarse-to-fine representation which is capable of capturing long-range dependency and exploiting different levels of contextual information (similar to how natural language models represent sentence structure in terms of hierarchical representations such as verb and noun phrases). Second, the structure of the HIM allows us to design a rapid inference algorithm, based on dynamic programming, which yields the first polynomial time algorithm for image labeling. Third, we learn the HIM efficiently using machine learning methods from a labeled data set. We demonstrate that the HIM is comparable with the state-of-the-art methods by evaluation on the challenging public MSRC and PASCAL VOC 2007 image data sets.     

基于混合余弦相似度的中文文本层次关系挖掘

层次关系是中文文本概念间存在的最为重要的关系之一,对层次关系的正确判定是进行领域本体自动构建、文本数据挖掘等信息处理的基础研究内容。先将概念间可能存在的候选层次关系罗列出来,构建词性序列语义余弦相似度和关系词语余弦相似度混合的核函数分类器,将概念间层次关系的挖掘问题转化为分类问题;再通过对文本数据进行模板标注来训练分类器;最后输入预处理后的中文文本,使用核函数分类器对候选层次关系进行判定。以空军武器装备领域的中文文本为测试数据,通过实验表明,该方法简单可靠,具有较好的正确率和召回率。     

融合上下文和多尺度特征的胸部多器官分割

目的 肿瘤周围高危器官的准确分割是图像引导放射治疗中的关键步骤,也是对抗肺癌和食道癌,规划有效治疗策略的重要组成部分。为了解决不同患者之间器官形状和位置的复杂变化情况以及计算机断层扫描（computed tomography,CT）图像中相邻器官之间软组织对比度低等问题,本文提出了一种深度学习算法对胸部CT图像中的高危器官进行细分。方法 以U-Net神经网络结构为基础,将冠状面下的3个连续切片序列即2.5D （2.5 dimention）数据作为网络输入来获取切片联系,同时利用高效全局上下文实现不降维的跨通道交互、捕获单视图下切片序列间的长距离依赖关系、加强通道联系和融合空间全局上下文信息。在编码部分使用金字塔卷积和密集连接的集成提取多尺度信息,扩大卷积层的感受野,并将解码器与编码器每层进行连接来充分利用多尺度特征,增强特征图的辨识度。考虑到CT图像中多器官形状不规则且紧密相连问题,加入深度监督来学习不同层的特征表示,从而精准定位器官和细化器官边界。结果 在ISBI （International Symposium on Biomedical Imaging）2019 SegTHOR （segmentation of thoracic organs at risk in CT images）挑战赛中,对40个胸部多器官训练样本进行分割,以Dice系数和HD （Hausdorff distance）距离作为主要评判标准,该方法在测试样本中食道、心脏、气管和主动脉的Dice系数分别达到0.855 1、0.945 7、0.923 0和0.938 3,HD距离分别为0.302 3、0.180 5、0.212 2和0.191 8。结论 融合全局上下文和多尺度特征的算法在胸部多器官分割效果上更具竞争力,有助于临床医师实现高效的诊断与治疗。     

Range segmentation of large building exteriors: A hierarchical robust approach

There are three main challenging issues associated with processing range data of large-scale outdoor scene: (a) significant disparity in the size of features, (b) existence of complex and multiple structures; and (c) high uncertainty in data due to the construction error or moving objects. Existing range segmentation methods in computer vision literature have been generally developed for laboratory-sized objects or shapes with simple geometric features and do not address these issues. This paper studies the main problems involved in segmenting the range data of large building exteriors and presents a robust hierarchical segmentation strategy to extract fine as well as large details from such data. The proposed method employs a high breakdown robust estimator in a coarse-to-fine approach to deal with the existing discrepancies in size and sampling rates of various features of large outdoor objects. The segmentation algorithm is tested on several outdoor range datasets obtained by different laser rangescanners. The results show that the proposed method is an accurate and computationally cost-effective tool that facilitates automatic generation of 3D models of large-scale objects in general and building exteriors in particular.     

基于深度学习的心脏核磁共振图像自动分割算法

心脏核磁共振成像技术由于其无电离辐射的优点已成为医疗诊断中的主要手段。对左心室、右心室以及左心肌进行准确的分割与识别是心脏手术前的重要一步,手动分割心脏结构耗时且易出错,因此自动分割双心室与心肌至关重要。提出了一种能充分利用心脏图像信息的多尺度特征融合U型神经网络MFF U-Net。首先,选择以U-Net++作为网络基本框架。其次,为了提高特征复用率,解决网络深度增加导致的过拟合问题,在U-Net++的编码部分提出了密集残差模块,使得网络在下采样过程中学习到更多有用特征。此外,在解码部分,为了使网络的分割结果更加符合目标器官之间的物理特征,用多个卷积核来扩大感受野并利用长距离依赖模块共享全局上下文信息,使得网络在编码还原的过程中尽可能地获取到目标器官之间的关系信息,从而使得分割结果更为精准。最后,考虑到双心室与左心肌的连贯性与唯一性,还添加了获取最大连通域与填充细小孔洞的后处理操作。采用的实验数据为ACDC心脏分割挑战数据集,其包含150位志愿者收缩期末期与舒张期末期的短轴心脏磁共振图像。在该数据集的测试集上进行验证,并通过在线提交的方式获取实验结果。实验结果表明,相较于其他算法,所提出的算法能够有效地分割目标器官,特别是舒张期末期的Dice系数分别达到了左心室0.96、右心室0.94和左心肌0.89,收缩期末期的分割精度达到了0.87,0.86和0.89。