Music symbol recognition by a LAG-based combination model

Most of optical music recognition (OMR) systems work under the assumption that the input image is scanner-based. However, we propose in this paper, camera based OMR system. Camera based OMR has a challengeable work in un-controlled environment such as a light, perspective, curved, transparency distortions and uneven staff-lines which tend to incur more frequently. In addition, the loss in performance of binarization methods, line thickness variation and space variation between lines are inevitable. In order to solve these problems, we propose a novel and effective staff-line removal method based on following three main ideas. First, a state-of-the-art staff-line detection method, Stable Path, is used to extract staff-line skeletons of the music score. Second, a line adjacency graph (LAG) model is exploited in a different manner over segmentation to cluster pixel runs generated from the run-length encoding (RLE) of an music score image. Third, a two-pass staff-line removal pipeline called filament filtering is applied to remove clusters lying on the staff-line. A music symbol is comprised of several parts so-called primitives, but the combination of these parts to form music symbol is unlimited. It causes difficulty applying the state-of-the-art method for music symbol recognition. To overcome these challenges and deal with primitive parts separately, we proposed a combination model which consists of LAG model, Graph model, and Set model as a framework for music symbol recognition. Our method shows impressive results on music score images captured from cameras, and gives high performance when applied to the ICDAR/GREC 2013 database, and a Gamera synthetic database. We have compared to some commercial software and proved the expediency and efficiency of the proposed method.     

Automated PET-guided liver segmentation from low-contrast CT volumes using probabilistic atlas

The use of the functional PET information from PET-CT scans to improve liver segmentation from low-contrast CT data is yet to be fully explored. In this paper, we fully utilize PET information to tackle challenging liver segmentation issues including (1) the separation and removal of the surrounding muscles from liver region of interest (ROI), (2) better localization and mapping of the probabilistic atlas onto the low-contrast CT for a more accurate tissue classification, and (3) an improved initial estimation of the liver ROI to speed up the convergence of the expectation-maximization (EM) algorithm for the Gaussian distribution mixture model under the guidance of a probabilistic atlas. The primary liver extraction from the PET volume provides a simple mechanism to avoid the complicated pre-processing of feature extraction as used in the existing liver CT segmentation methods. It is able to guide the probabilistic atlas to better conform to the CT liver region and hence helps to overcome the challenge posed by liver shape variability. Our proposed method was evaluated against manual segmentation by experienced radiologists. Experimental results on 35 clinical PET-CT studies demonstrated that our method is accurate and robust in automated normal liver segmentation.     

Structure-Sensitive Superpixels via Geodesic Distance

Segmenting images into superpixels as supporting regions for feature vectors and primitives to reduce computational complexity has been commonly used as a fundamental step in various image analysis and computer vision tasks. In this paper, we describe the structure-sensitive superpixel technique by exploiting Lloyd’s algorithm with the geodesic distance. Our method generates smaller superpixels to achieve relatively low under-segmentation in structure-dense regions with high intensity or color variation, and produces larger segments to increase computational efficiency in structure-sparse regions with homogeneous appearance. We adopt geometric flows to compute geodesic distances amongst pixels. In the segmentation procedure, the density of over-segments is automatically adjusted through iteratively optimizing an energy functional that embeds color homogeneity, structure density. Comparative experiments with the Berkeley database show that the proposed algorithm outperforms the prior arts while offering a comparable computational efficiency as TurboPixels. Further applications in image compression, object closure extraction and video segmentation demonstrate the effective extensions of our approach.     

Blood Vessel Segmentation via Neural Network in Histological Images

In this paper we utilize the Kohonen's self-organizing feature map to segment blood vessels from biopsies in tumor tissue. The ability of this kind of neural network to recognize very complex patterns makes it an effective computational tool for the segmentation. We propose a strategy of blood vessels segmentation using a neural network, taking into account the quality of our images and its features: complexity in shape and variability in size. Segmentation results are contingently manually corrected. The proposed segmentation strategy is tested on manual segmentation, where segmentation errors of less than 3.5% are observed. This work is a part of a global image analysis process and these images will be subject to a further morphometrical analysis in order to diagnose and prognosticate automatically malignant tumours.     

Segmentation and classification of brain tumors using modified median noise filter and deep learning approaches

The most vital challenge for a radiologist is locating the brain tumors in the earlier stage. As the brain tumor grows rapidly, doubling its actual size in about twenty-five days. If not dealt properly, the affected person’s survival rate usually does no longer exceed half a year. This can rapidly cause dying. For this reason, an automatic system is desirable for locating brain tumors at the early stage. In general, when compared to computed tomography (CT), magnetic resonance image (MRI) scans are used for detecting the diagnosis of cancerous and noncancerous tumors. However, while MRI scans acquisition, there is a chance of appearing noise such as speckle noise, salt & pepper noise and Gaussian noise. It may degrade classification performance. Hence, a new noise removal algorithm is proposed, namely the modified iterative grouping median filter. Further, Maximum likelihood estimation-based kernel principal component analysis is proposed for feature extraction. A deep learning-based VGG16 architecture has been utilized for segmentation purposes. Experimental results have shown that the proposed algorithm outperforms the well-known techniques in terms of both qualitative and quantitative evaluation.

     

Preserving boundaries for image texture segmentation using grey level co-occurring probabilities

Texture analysis has been used extensively in the computer-assisted interpretation of digital imagery. A popular texture feature extraction approach is the grey level co-occurrence probability (GLCP) method. Most investigations consider the use of the GLCP texture features for classification purposes only, and do not address segmentation performance. Specifically, for segmentation, the pixels in an image located near texture boundaries have a tendency to be misclassified. Boundary preservation when using the GLCP texture features for image segmentation is important. An advancement which exploits spatial relationships has been implemented. The generated features are referred to as weighted GLCP (WGLCP) texture features. In addition, an investigation for selecting suitable GLCP parameters for improved boundary preservation is presented. From the tests, WGLCP features provide improved boundary preservation and segmentation accuracy at a computational cost. As well, the GLCP correlation statistical parameter should not be used when segmenting images with high contrast texture boundaries.     

An Optimized Image Segmentation Approach Based on Boltzmann Machine

Image segmentation with complex background is a tedious task. In our study, a convex spline is constructed based on Good Features to Track (GF2T) method’s region-based salient feature (i.e., corner) set. For an optimized edge-based segmentation, an ellipse shape prior based on this convex spline is useful in edge regularization procedure with region-based features. This kind of optimization is achieved by Boltzmann machine (BM) to automatically form an elliptical foreground mask of the GrabCut method. We demonstrated our approach’s usability through traveling salesman problem (TSP), thus, we consider that the TSP’s valid tour’s path solved by BM can be taken as an optimized convex spline for edge-based segmentation. In our experiments, proposed BM-based approach has the performance improvement of segmentation to stand-alone GF2T as 29.79% improvement based on bounding boxes evaluation and as 38.67% improvement based on the overlapping pixel regions for a quantitative evaluation via objective metrics.     

Nonlinear model and constrained ML for removing back-to-front interferences from recto–verso documents

In this paper, we approach the removal of back-to-front interferences from scans of double-sided documents as a blind source separation problem, and extend our previous linear mixing model to a more effective nonlinear mixing model. We consider the front and back ideal images as two individual patterns overlapped in the observed recto and verso scans, and apply an unsupervised constrained maximum likelihood technique to separate them. Through several real examples, we show that the results obtained by this approach are much better than the ones obtained through data decorrelation or independent component analysis. As compared to approaches based on segmentation/classification, which often aim at cleaning a foreground text by removing all the textured background, one of the advantages of our method is that cleaning does not alter genuine features of the document, such as color or other structures it may contain. This is particularly interesting when the document has a historical importance, since its readability can be improved while maintaining the original appearance.     

Recognition of organs in CT-image sequences: a model guided approach

A method is proposed for segmentation of organs in CT-image sequences. An important feature of the method is the use of search areas to guide the segmentation process. Time consuming, data-directed operations are restricted to these areas, instead of being applied to the whole image. A search area for a particular anatomical structure is determined by constraints, derived from a model of the imaged scene and by results already obtained in the recognition process. The method has been successfully applied on recognition of the spleen in abdominal X-ray CT scans.     

SurvSurf: human retrieval on large surveillance video data

The volume of surveillance videos is increasing rapidly, where humans are the major objects of interest. Rapid human retrieval in surveillance videos is therefore desirable and applicable to a broad spectrum of applications. Existing big data processing tools that mainly target textual data cannot be applied directly for timely processing of large video data due to three main challenges: videos are more data-intensive than textual data; visual operations have higher computational complexity than textual operations; and traditional segmentation may damage video data’s continuous semantics. In this paper, we design SurvSurf, a human retrieval system on large surveillance video data that exploits characteristics of these data and big data processing tools. We propose using motion information contained in videos for video data segmentation. The basic data unit after segmentation is called M-clip. M-clips help remove redundant video contents and reduce data volumes. We use the MapReduce framework to process M-clips in parallel for human detection and appearance/motion feature extraction. We further accelerate vision algorithms by processing only sub-areas with significant motion vectors rather than entire frames. In addition, we design a distributed data store called V-BigTable to structuralize M-clips’ semantic information. V-BigTable enables efficient retrieval on a huge amount of M-clips. We implement the system on Hadoop and HBase. Experimental results show that our system outperforms basic solutions by one order of magnitude in computational time with satisfactory human retrieval accuracy.     

面向部件分割的PointNet注意力加权特征聚合网络

为解决PointNet最大池化损失次要特征导致部件分割精度降低的问题,提出一种面向部件分割的PointNet注意力加权特征聚合网络,能够充分利用点云的不同特征进行部件分割。首先利用多层感知机提取点云的空间几何特征,将特征乘以共享权重矩阵,以获取每个点的每一个特征的注意力分数;接着把归一化的分数作为权重乘以对应的全局特征并求和,得到聚合的全局特征;最后使用多层感知机将聚合的特征映射到部件分割结果。实验结果表明,相比于传统PointNet方法,该方法提升了部件分割的总平均交并比,同时在网络鲁棒性和计算复杂度方面具有显著优势。该方法有效优化了PointNet。     

Smart objects identification system for robotic surveillance

Video surveillance is an active research topic in computer vision.In this paper,humans and cars identifcation technique suitable for real time video surveillance systems is presented.The technique we proposed includes background subtraction,foreground segmentation,shadow removal,feature extraction and classifcation.The feature extraction of the extracted foreground objects is done via a new set of afne moment invariants based on statistics method and these were used to identify human or car.When the partial occlusion occurs,although features of full body cannot be extracted,our proposed technique extracts the features of head shoulder.Our proposed technique can identify human by extracting the human head-shoulder up to 60%–70%occlusion.Thus,it has a better classifcation to solve the issue of the loss of property arising from human occluded easily in practical applications.The whole system works at approximately 16 29 fps and thus it is suitable for real-time applications.The accuracy for our proposed technique in identifying human is very good,which is 98.33%,while for cars identifcation,the accuracy is also good,which is 94.41%.The overall accuracy for our proposed technique in identifying human and car is at 98.04%.The experiment results show that this method is efective and has strong robustness.     

Use of Spectral Clustering Combined with Normalized Cuts (N-Cuts) in an Iterative k-Means Clustering Framework (NKSC) for Superpixel Segmentation with Contour Adherence

Superpixel segmentation methods are generally used as a pre-processing step to speed up image processing tasks. They group the pixels of an image into homogeneous regions while trying to respect existing contours. In this paper, we propose a fast Superpixels segmentation algorithm with Contour Adherence using spectral clustering, combined with normalized cuts in an iterative k-means clustering framework. It produces compact and uniform superpixels with low computational costs. Normalized cut is adapted to measure the color similarity and space proximity between image pixels. We have used a kernel function to estimate the similarity metric. Kernel function maps the pixel values and coordinates into a high dimensional feature space. The objective functions of weighted K-means and normalized cuts share the same optimum point in this feature space. So it is possible to optimize the cost function of normalized cuts by iteratively applying simple K-means clustering algorithm. The proposed framework produces regular and compact superpixels that adhere to the image contours. On segmentation comparison benchmarks it proves to be equally well or better than the state-of-the-art super pixel segmentation algorithms in terms of several commonly used evaluation metrics in image segmentation. In addition, our method is computationally very efficient and its computational complexity is linear.     

On the optimality of image processing pipeline

The optimisation of image processing steps such as segmentation and feature extraction individually in an application does not yield an optimal pipeline. In this paper we demonstrate how the use of different image segmentation algorithms directly impacts upon the quality of texture measures extracted from segmented regions and final classification ability. The difference between the best and the worst possible performances by choosing different algorithms is found to be significant. We then develop the methodology for determining the optimal pipeline for scene analysis and show our experimental results on the publicly available benchmark “MINERVA”.     

结合相似性拟合与空间约束的图像分割

目的 图像中的目标一般含有很多子类,仅仅利用某个子类的特征无法完整地分割出目标区域。针对这一问题,提出一种结合相似性拟合与空间约束的图像交互式分割方法。方法 首先,通过手工标记的样本组成各个目标的字典,通过相似度量搜寻测试样本与各个目标的字典中最相似的原子建立拟合项;再结合图像的空间约束项,构建图像分割模型;最后利用连续最大流算法求解,快速实现图像分割的目的。结果 通过对比实验,本文方法的速度比基于稀疏表示的分类方法的速度提高约13倍,而与归一化切割(N-Cut),逻辑回归(logistic regression)等方法相比,本文方法能取得更稳定和准确的分割结果。此外,本文方法无需过完备字典,只需要训练样本能体现各个子类的信息即可得到稳定的图像分割结果。结论 本文交互式图像分割方法,通过结合相似性拟合以及空间约束建立分割模型,并由连续最大流算法求解,实现图像的快速准确的分割。实验结果表明,该方法能够胜任较准确地对自然图像进行分割以及目标提取等任务。     

融合语义—表观特征的无监督前景分割

目的 前景分割是图像理解领域中的重要任务,在无监督条件下,由于不同图像、不同实例往往具有多变的表达形式,这使得基于固定规则、单一类型特征的方法很难保证稳定的分割性能。针对这一问题,本文提出了一种基于语义-表观特征融合的无监督前景分割方法（semantic apparent feature fusion,SAFF）。方法 基于语义特征能够对前景物体关键区域产生精准的响应,但往往产生的前景分割结果只关注于关键区域,缺乏物体的完整表达;而以显著性、边缘为代表的表观特征则提供了更丰富的细节表达信息,但基于表观规则无法应对不同的实例和图像成像模式。为了融合表观特征和语义特征优势,研究建立了融合语义、表观信息的一元区域特征和二元上下文特征编码的方法,实现了对两种特征表达的全面描述。接着,设计了一种图内自适应参数学习的方法,用于计算最适合的特征权重,并生成前景置信分数图。进一步地,使用分割网络来学习不同实例间前景的共性特征。结果 通过融合语义和表观特征并采用图像间共性语义学习的方法,本文方法在PASCAL VOC（pattern analysis,statistical modelling and computational learning visual object classes）2012训练集和验证集上取得了显著超过类别激活映射（class activation mapping,CAM）和判别性区域特征融合方法（discriminative regional feature integration,DRFI）的前景分割性能,在F测度指标上分别提升了3.5%和3.4%。结论 本文方法可以将任意一种语义特征和表观特征前景计算模块作为基础单元,实现对两种策略的融合优化,取得了更优的前景分割性能。     

Gaussian-Process-Based Real-Time Ground Segmentation for Autonomous Land Vehicles

Ground segmentation is a key component for Autonomous Land Vehicle (ALV) navigation in an outdoor environment. This paper presents a novel algorithm for real-time segmenting three-dimensional scans of various terrains. An individual terrain scan is represented as a circular polar grid map that is divided into a number of segments. A one-dimensional Gaussian Process (GP) regression with a non-stationary covariance function is used to distinguish the ground points or obstacles in each segment. The proposed approach splits a large-scale ground segmentation problem into many simple GP regression problems with lower complexity, and can then get a real-time performance while yielding acceptable ground segmentation results. In order to verify the effectiveness of our approach, experiments have been carried out both on a public dataset and the data collected by our own ALV in different outdoor scenes. Our approach has been compared with two previous ground segmentation techniques. The results show that our approach can get a better trade-off between computational time and accuracy. Thus, it can lead to successive object classification and local path planning in real time. Our approach has been successfully applied to our ALV, which won the championship in the 2011 Chinese Future Challenge in the city of Ordos.     

Weight adaptation and oscillatory correlation for imagesegmentation

We propose a method for image segmentation based on a neural oscillator network. Unlike previous methods, weight adaptation is adopted during segmentation to remove noise and preserve significant discontinuities in an image. Moreover, a logarithmic grouping rule is proposed to facilitate grouping of oscillators representing pixels with coherent properties. We show that weight adaptation plays the roles of noise removal and feature preservation. In particular, our weight adaptation scheme is insensitive to termination time and the resulting dynamic weights in a wide range of iterations lead to the same segmentation results. A computer algorithm derived from oscillatory dynamics is applied to synthetic and real images, and simulation results show that the algorithm yields favorable segmentation results in comparison with other recent algorithms. In addition, the weight adaptation scheme can be directly transformed to a novel feature-preserving smoothing procedure. We also demonstrate that our nonlinear smoothing algorithm achieves good results for various kinds of images.     

基于区块特征融合的点云语义分割方法

为降低室外大规模点云场景中多类三维目标语义分割的计算复杂度,提出一种融合区块特征的语义分割方法。采用方形网格分割方法对三维点云进行区块划分、采样以及组合,求取简化的点云组合区块集,将其输入至区块特征提取和融合网络中从而获得每个区块的特征修正向量。设计点云区块全局特征修正网络,以残差的方式融合特征修正向量与原始点云全局特征,修正因分割造成的错误特征。在此基础上,将方形网格分割尺寸作为神经网络的参数引入反向传播过程中进行优化,从而建立高效的点云语义分割网络。实验结果表明,反向传播算法可以优化分割尺寸至最佳值附近,所提网络中的全局特征修正方法能够提高语义分割精度,该方法在Semantic3D数据集上的语义分割精度达到78.7%,较RandLA-Net方法提升1.3%,且在保证分割精度的前提下其点云预处理计算复杂度和网络计算时间明显降低,在处理点数为10万~100万的大规模点云时,点云语义分割速度较SPG、KPConv等方法提升2~4倍。     

基于链式竞争遗传算法的KSW熵的图像分割

针对最佳熵阈值图像分割算法过程中计算复杂度高的问题,提出了一种基于链式竞争遗传算法的最佳熵阈值确定法(KSW熵法)的图像分割算法.通过将3个邻域的链式竞争引入到常规遗传算法框架下,实现特征选择过程;将改进的遗传算法应用到最佳阈值图像分割算法中,完成对阈值的寻优过程.仿真实验结果与分析表明:算法在分割速度和效果上均优于传统的最佳阈值图像分割算法和单纯的遗传优化最佳阈值图像分割算法.