Study on traffic scene semantic segmentation method based on convolutional neural network

In order to improve the semantic segmentation accuracy of traffic scene,a segmentation method was proposed based on RGB-D image and convolutional neural network.Firstly,on the basis of semi-global stereo matching algorithm,the disparity map was obtained,and the sample library was established by fusing the disparity map D and RGB image into the four-channel RGB-D image.Then,with two different structures,the networks were trained by using two different learning rate adjustment strategy respectively.Finally,the traffic scene semantic segmentation test was carried out with RGB-D image as the input,and the results were compared with the segmentation method based on RGB image.The experimental results show that the proposed traffic scene segmentation algorithm based on RGB-D image can achieve higher semantic segmentation accuracy than that based on RGB image.     

Modelling and segmentation of lip area in face images

The problem of automatic segmentation of face images to extract the lip area is considered in the context of lip tracking. A novel segmentation method is proposed for lip tracker initialisation which is based on a Gaussian mixture model of the pixel RGB values. The model is built using the predictive validation technique advocated by Kittler (1997) which has been modified to allow modelling with full, rather than just diagonal covariance matrices. A subsequent grouping of the mixture components provides the basis for a Bayesian rule labelling of the pixels as lip or non-lip. The proposed method is tested on a database of 145 images and demonstrates that its accuracy is significantly better than the segmentation obtained by k-means clustering. Moreover, the proposed method does not require the number of segments to be specified a priori.     

Robust Environmental Change Detection Using PTZ Camera via Spatial-Temporal Probabilistic Modeling

This paper proposes a novel procedure for detecting environmental changes by using a pan-tilt-zoom (PTZ) camera. Conventional approaches based on pixel space and stationary cameras need time-consuming image registration to yield pixel statistics. This work proposes an alternative approach to describe each scene with a Gaussian mixture model (GMM) via a spatial-temporal statistical method. Although details of the environment covered by the camera are lost, this model is efficient and robust in recognizing scene and detecting scene changes in the environment. Moreover, the threshold selection for separating different environmental changes is convenient by using the proposed framework. The effectiveness of the proposed method is demonstrated experimentally in an office environment.     

基于改进的2-D Otsu方法和YCgCr空间的肤色分割

为了弥补基于固定阈值的肤色分割方法存在的缺陷,在对多种彩色空间和肤色模型进行分析的基础上,提出采用改进的2-D Otsu方法和YCgCr彩色空间进行肤色分割。首先将光照补偿之后的肤色样本图像从RGB彩色空间转换到YCgCr彩色空间,并利用样本图像上的179221个肤色点建立2维高斯模型;进而将待分割的图像进行光照补偿并转换到YCgCr彩色空间,利用已经建立的高斯模型计算图像的肤色相似度,得到肤色相似度图像;最后,结合像素的空间邻域信息,使用改进的2-D Otsu方法对肤色相似度图像进行2值化处理。对这种方法进行了理论分析和实验验证。结果表明,该肤色分割算法有效地克服了使用固定阈值法进行图像分割时缺乏针对性和抗噪性的缺陷,该算法是可行的。     

基于全卷积神经网络的非对称并行语义分割模型

针对RGB图像具有丰富的色彩细节特征,红外图像对目标轮廓、尺寸、边界等外形特征有较高敏感度的特点,提出了一种非对称并行语义分割模型APFCN（Asymmetric Parallelism Fully Convolutional Networks）.APFCN上路设计了一个卷积核尺寸非统一的五层空洞卷积网络来提取红外图像目标高层轮廓特征;下路沿用卷积加池化网络提取RGB图像三个尺度上的细节特征;后端将红外图像高层特征与RGB图像三个尺度的细节特征进行融合,并将4倍上采样后的融合特征作为语义分割输出.结果表明,APFCN在像素精度和交并比等方面均优于FCN（输入为RGB图像或红外图像）,适用于背景一致下地面目标的语义分割任务.     

一种用于视频目标检测的时空背景建模方法

针对传统时域统计背景模型只关注于单个像素的统计特征而忽略背景像素之间的相互关联,同时不能很好地适应目标进入场景并静止的情形,在高斯背景模型基础上,融合背景的空间统计特征,建立一个时空联合的背景模型。结合背景的空间区域特征和像素的亮度变化更新背景,使得背景更新过程中保持背景的空间特征。实验结果表明,利用该方法建立的背景模型进行目标检测,能有效检测移入静止的目标,克服了传统检测方法检测目标中的空洞,检测到的目标完整性有较大提高,有利于后续的进一步处理。      

图像非广延高斯熵分割的阈值选取方法

图像分割是实现图像目标与背景分离的基础和关键,其中阈值化是最为流行和有效的技术之一。考虑到图像处理作业场景的复杂境况,实时有效的阈值分割方法的提出一直是研究和应用中的极大挑战。针对复杂的工业无损检测图像及红外图像目标分割问题,基于非广延高斯熵提出一种新的图像阈值分割方法。对于图像中像素灰度级不均衡、不规则的随机分布特点,提出方法体现了较好的处理能力。在与实践常用方法的比较实验中,实验结果表明了提出方法的较优异性能和推广应用前景。     

红外小目标检测的滞后阈值分割法

为提高红外小目标检测的精度,针对小目标的阈值分割,提出滞后阈值分割法。通过快速非最大值抑制预提取红外小目标的中心像素,最大程度缩减前景像素数。利用邻域梯度方差判别小目标是否为杂波,自适应地调整灰度阈值,完成对小目标的低虚警分割。分析一般场景下分割所得前景、背景的组成,给出了滞后阈值分割的阈值取值方法。实验表明滞后阈值分割法将虚警概率限制在很小的范围内,并取得了优于其他分割法的检测概率。在小目标检测中,该算法具有较强的实用性。     

基于区域限制的EM和图割的非监督彩色图像分割方法

提出一种基于区域限制的EM(Expectation Maximization)和图割的非监督彩色图像分割方法,以解决自动确定分割类数问题.首先,生成图像的超像素,提取图像的CIE Lab颜色特征和多尺度四元数Gabor滤波特征;为了高效自动地确定分割类数,同时避免因直接使用超像素造成的奇异值问题,对每一个超像素采样并使用采样像素表示超像素;然后采用高斯混合模型对采样像素集合进行建模,使用加入区域限制的分量EM自动获取模型组件数及参数,最后使用图割结合高斯混合模型对图像进行优化,获取最终分割结果.实验结果表明,该方法在分割效率和分割质量上均得到较大提升.     

Background subtraction based on circulant matrix

Detecting moving objects in a scene is a fundamental and critical step for many high-level computer vision tasks. However, background subtraction modeling is still an open and challenge problem, particularly in practical scenarios with drastic illumination changes and dynamic backgrounds. In this paper, we present a novel background modeling method focused on dealing with complex environments based on circular shift operator. The background model is constructed by performing circular shifts on the neighborhood of each pixel, which forms a basic region unit. The foreground mask is obtained via two stages. The first stage is to subtract the established background from the current frame to obtain the distance map. The second is to adopt the graph cut on the distance map. In order to adapt to the background changes, the background model is updated with an adaptive update rate. Experimental results on indoor and outdoor videos demonstrate the efficiency of our proposed method.     

基于视景仿真技术的虚拟激光投影动态交互方法北大核心CSCD

虚拟激光投影在动态交互时,易用性差、鲁棒性低都会导致交互方法的交互性能变差,为提高激光投影交互效果,提出基于视景仿真技术的虚拟激光投影动态交互方法。利用VEW方法采集激光投影视频图像,利用加权平均法对采集到的视频图像背景去噪,并基于红绿蓝三原色建立视频的RGB彩色空间模型,对背景去噪后的图像灰度转换后,进行阈值分割;依据预处理结果确定激光投影中心点位置,通过投影的中心点映射,实现投影动态交互;使用视景仿真技术对交互场景图像的交互效果实施增强,提高激光投影的动态交互性能。实验结果表明,运用该方法开展激光投影动态交互时,交互性高、易用性好、鲁棒性高。     

Segmentation of video background regions based on a DTCNN-clustering approach

We propose a novel algorithm for segmentation of video background models in time-variant scenarios. It is robust to gradual or abrupt illumination changes, diverse kind of noises, and even scenario variation. The algorithm generates regions according to the scene composition by keeping region segmentation coherence. The proposed method based on a discrete-time cellular neural network estimates the number regions in the current background model, and then, a modified k-means algorithm is used to achieve segmentation. The findings demonstrate the robustness of the method and its superiority over two state of the art scene segmentation algorithms.     

光照突变下融合多类特征的场景分割方法

为解决场景模型在快速光照变化下失效的问题,提出了一种新的前景目标分割方法。该方法共包括三个步骤。首先,利用全局光照函数建立高斯混合模型;其次,提取当前帧中的纹理、ZNCC 及轮廓特征;最后,将提取到的特征分两阶段与高斯混合模型进行融合(第一阶段:融合纹理及ZNCC 特征;第二阶段:融合轮廓特征),得到最终的场景分割结果。实验结果表明:该算法具有较好的鲁棒性,并且相较于基于全局光照建模的方法具有更高的精度值及召回值。     

基于改进加权增强局部对比度测量的红外小目标检测方法

红外弱小目标检测技术是红外搜索与跟踪系统的重要组成部分(IRST)。一般来说,在复杂背景环境下,红外弱小目标检测往往会有高虚警率和低检测率的问题。为了解决这一问题,提出一个改进的加权增强局部对比度测量（IWELCM）检测框架,具有重要意义。首先,通过将局部对比度机制与信杂比(SCR)的计算相结合,提出一个增强的局部对比度测量方法,在增强图像中疑似红外弱小目标区域的同时也提高图像的SCR。其次,通过利用红外图像中弱小目标的特性,以及目标与周围背景的统计差异,提出一个改进的加权函数来进一步增强目标和抑制背景。最后,采用一个自适应阈值分割的方法去获取检测的目标。在不同场景的数据集上的对比实验表明,与七种现有流行的方法相比,提出方法在复杂背景下能够有效地从干扰对象中提取真实的红外弱小目标,具有更好的检测性能。     

A New Method for the Estimation of Mass Functions in?the?Dempster�CShafer��s Evidence Theory: Application to Colour Image Segmentation

In this paper, the problem of colour image segmentation is addressed using the Dempster–Shafer (DS) theory. Examples are provided showing that this theory is able to take into account a large variety of special situations that occur and which are not well solved using classical approaches. Modelling both uncertainty and imprecision, and computing the conflict between images and introducing a priori information are the main features of this theory. Consequently, the performance of such a segmentation scheme is largely conditioned by the appropriate estimation of mass functions in the DS evidence theory. In this paper, a new method of automatically determining the mass function for colour-image segmentation problems is presented. The mass function of each pixel is determined by applying possibilistic c-means (PCM) clustering to the grey levels of the three primitive colours. A reliability criterion, associated with each pixel and the mass functions of its neighbouring pixels, is used into a fuzzy based reasoning system in order to decide on the appropriate segmentation. Experimental segmentation results on medical and textured colour images highlight the effectiveness of the proposed method.     

Independent Component Analysis-Based Background Subtraction for Indoor Surveillance

In video surveillance, detection of moving objects from an image sequence is very important for target tracking, activity recognition, and behavior understanding. Background subtraction is a very popular approach for foreground segmentation in a still scene image. In order to compensate for illumination changes, a background model updating process is generally adopted, and leads to extra computation time. In this paper, we propose a fast background subtraction scheme using independent component analysis (ICA) and, particularly, aims at indoor surveillance for possible applications in home-care and health-care monitoring, where moving and motionless persons must be reliably detected. The proposed method is as computationally fast as the simple image difference method, and yet is highly tolerable to changes in room lighting. The proposed background subtraction scheme involves two stages, one for training and the other for detection. In the training stage, an ICA model that directly measures the statistical independency based on the estimations of joint and marginal probability density functions from relative frequency distributions is first proposed. The proposed ICA model can well separate two highly-correlated images. In the detection stage, the trained de-mixing vector is used to separate the foreground in a scene image with respect to the reference background image. Two sets of indoor examples that involve switching on/off room lights and opening/closing a door are demonstrated in the experiments. The performance of the proposed ICA model for background subtraction is also compared with that of the well-known FastICA algorithm.      

高动态条件下增量惯导信息辅助的空地红外弱小移动目标检测算法（特邀）

红外弱小移动目标检测技术是计算机视觉的研究热点和难点。针对机载高动态条件下的空地目标检测存在的场景变化动态、背景干扰强度大、目标运动规律未知等挑战,提出了一种新型的基于增量惯导信息辅助的空地红外弱小移动目标检测算法。为了解决传统惯导信息预测的漂移误差问题,提出了增量惯导信息概念,设计了增量惯导信息的位置预测模型,实现了对目标点的准确预测。构建了基于增量惯导信息辅助与背景差分的移动目标检测框架,通过增量惯导信息对不同位姿下的成像进行校正,引入基于爬山法互相关匹配算法计算校正后图像的平移参数,采用高斯加权对背景图像进行估计,最后通过图像分割检测弱小移动目标。仿真实验验证了文中设计检测算法的有效性和精确性。     

基于层次聚类的多颜色空间融合分割方法

针对自然图像分割中,由于单一的颜色空间难以表示复杂多变的场景信息以及目标与背景的低对比度等产生的过分割和误分割问题,提出了选择RGB、XYZ和LUV这3个颜色空间的增强图像进行基于层次聚类的融合分割的新方法.对Berkeley分割图像库中的多幅图像进行了多组分割实验,并与Mean-shift等多种经典分割方法进行了定性...     

Automatic body segmentation with graph cut and self-adaptive initialization level set (SAILS)

In this paper, we propose an automatic human body segmentation system which mainly consists of human body detection and object segmentation. Firstly, an automatic human body detector is designed to provide hard constraints on the object and background for segmentation. And a coarse-to-fine segmentation strategy is employed to deal with the situation of partly detected object. Secondly, background contrast removal (BCR) and self-adaptive initialization level set (SAILS) are proposed to solve the tough segmentation problems of the high contrast at object boundary and/or similar colors existing in the object and background. Finally, an object updating scheme is proposed to detect and segment new object when it appears in the scene. Experimental results demonstrate that our body segmentation system works very well in the live video and standard sequences with complex background.     

一种改进的自适应背景建模视频分割方法

针对传统的自适应背景建模难以解决背景复杂以及外界光照变化等因素影响视频分割质量的问题,提出了一种改进的自适应背景建模视频分割方法。该方法首先对彩色图像建立高斯背景模型,然后对背景模型进行更新,最后通过对高斯分布准则进行改进使前景目标分割性能得到提高。仿真实验表明,该方法能够准确建立背景模型,准确分割前景目标,较传统算法具有更强的鲁棒性。