基于改进的混合高斯模型的运动目标检测

提出了一种改进的混合高斯背景模型方法,克服了传统混合高斯背景建模方法计算时间长的缺点。通过对视频图像中运动目标区域进行背景建模,减小了每一帧的背景建模区域,同时在提取运动目标区域前先对初提取的前景目标进行中值滤波,减小运动目标区域的范围,进一步压缩了背景建模的时间。最后通过与时间平均背景建模和传统混合高斯背景建模方法进行比较,验证了本文算法的高效性。     

基于改进混合高斯模型的井下目标检测算法

煤矿井下监控视频图像质量差、噪点多、光照易突变,采用传统混合高斯模型进行目标检测存在运行速度慢、算法复杂度高、易受光照影响等问题。针对该问题,提出了一种基于改进混合高斯模型的井下目标检测算法。使用改进的暗通道去雾算法对井下图像进行预处理,对井下雾图的缩略图求暗通道图,并采用双线性插值得到去雾图像;在混合高斯模型的基础上,使用改进的块建模策略降低建模复杂度,提高算法运行速度;结合三帧差分法,根据图像前景所占比例对高斯建模前期和建模后期设定不同的学习率,以抑制光照对目标检测的影响,提高建模速度和准确度。实验结果表明,当光照发生突变时,该算法能较好地描述检测对象,对光照变化有明显抑制作用;与三帧差分法、传统混合高斯模型相比,该算法可有效提高处理速度。     

改进混合高斯模型的自适应烟雾图像分割算法

为了解决烟雾分割算法中灰白(白)烟和黑色烟雾的同时提取以及分割阈值的自适应选取问题,利用烟雾属于前景目标的特征以及灰白(白)烟和黑色烟雾的颜色特征,提出一种改进混合高斯模型的自适应烟雾图像分割算法.该算法在混合高斯背景模型的基础上删除长期没有匹配的"过期"高斯成分,缩短背景建模时间;根据烟雾"团聚"特点以及灰白(白)烟和黑色烟雾各自不同的R,G,B三分量关系对烟雾图像进行分块,通过对不同的块使用自适应差分阈值提取烟雾可疑区域;针对光线突变、物体移入场景并静止下来等使场景可能发生渐变的问题,对背景图像进行全局和局部更新,并引入修正因子?对已经更新后的背景图像再次修正,确保背景图像更加接近真实场景;最后通过计算当前图像与背景图像的差异程度作为块差分阈值更新基准,实时更新块差分阈值.实验结果表明,文中算法不但能够同时提取灰白(白)烟和黑色烟雾,而且目标边缘不规则信息保留完整,分割准确度比已有算法提高了35.9%,运行速度有小幅度提升.     

基于视频的车流量检测算法

实现了一种基于虚拟检测线的道路车流量检测算法.利用高斯混合背景模型进行背景建模,检测出包含运动目标的图像块,系统根据超过检测线的连通区域图像面积占整个连通区域面积的比值来判断是否有车辆经过,进而实现车流量统计.     

基于图像块和边缘增强的运动目标检测

针对复杂背景下的噪声、动态背景和阴影等影响,提出一种基于图像块和边缘增强的运动目标检测算法。用图像块建模,按掩膜结构特点修补断桥使掩膜连续,采用高层图像金字塔建模得到目标质心,用于提取目标,用Sobel平移算子得到细化的边缘,按边缘信息对掩膜进行增强处理从而得到完整的目标。实验结果表明,该算法可有效提取目标,消除单高斯建模法去除不了的复杂背景和阴影,克服传统帧差法常见的边缘不完整的缺点。     

基于混合高斯模型的Mean-Shift模板更新算法

针对基于Mean-Shift目标跟踪算法中遇到的不能对模板进行实时更新的问题,提出一个基于混合高斯背景建模的目标模板更新算法.该算法将目标视为背景,对目标中的每一个像素点利用三个高斯函数对它进行建模,利用每次Mean-Shift跟踪到的目标区域来对先前建立的混合高斯模型进行实时更新,将混合高斯模型得到的目标模板作为下一帧跟踪的目标模板.该算法较好地解决了基于Mean-Shift算法的模板更新问题,实验证明该算法是有效的.     

结合单高斯与光流法的无人机运动目标检测

针对无人机场景下运动目标检测对实时性要求高,运动背景、环境光照易变化等问题,提出一种结合单高斯与光流法的运动目标检测算法.首先,对运动相机捕捉的图像采用改进的单高斯模型进行背景建模,并融合前一帧图像的多个高斯模型来进行运动补偿,然后,将得到的前景图像作为掩模来提取特征点和进行光流跟踪,并对稀疏特征点的运动轨迹进行层次聚类.实验结果表明,该算法能有效地处理运动相机造成的前景对背景模型的干扰,背景建模速度快,对光照变化不敏感,检测出的目标接近真实目标.     

体育视频中的运动员检测与跟踪

利用自适应高斯混合模型对视频图像进行建模,从图像序列中获取背景图像并提取运动区域,利用像素的颜色信息从背景图像中提取绿色球场。为提高运动员检测的准确度,利用纹理相似性度量方法消除运动区域中的阴影,用形态学方法消除区域内的裂缝,根据球场信息去除球场外的噪声。改进了CamShift算法,并应用该算法对运动员进行跟踪。     

融合子块梯度与线性预测的单高斯背景建模

针对单高斯背景模型不能适应非平稳场景且对初期保持静止后期运动的物体造成"鬼影"现象的问题,提出了融合子块梯度与线性预测的单高斯背景建模方法。首先,对每个像素点进行单高斯背景建模,并实现像素级的自适应更新,运用子块梯度算法将梯度在阈值内的子块作为背景以消除"鬼影";然后,将子块梯度法获得的前景与单高斯模型确定的前景做与运算,提高在非平稳场景下对背景的判断能力;最后,运用线性预测方法处理获得的前景点,将面积小于阈值的连通区域还原为背景。采用CDNET 2012 Dataset和Wallflower Dataset进行仿真实验:当场景变化幅度较大时,所提算法与混合高斯模型(GMM)相比,虽然检测率稍有下降,但检测精度提高了40%;在其他场景中检测率虽只提高约10%,检测精度却能提高25%以上。实验结果表明,融合子块梯度与线性预测的单高斯背景建模能够适应非平稳场景并消除"鬼影"现象,获得的背景比混合高斯模型更精确,提取的前景细节更丰富。     

基于高斯混合模型的纹理图像分割

纹理图像分割是图像处理的一个基本问题。由于基于高斯混合模型的纹理图像分割方法．大多采用单像素的方法，因此分割精度和效率都较低。为了更好地进行纹理图像分割，在子空间思想的基础上，提出了一个基于图像块的分割算法及其改进算法，即先取图像块的均值、标准差、最大值、最小值以及中间像素的像素值等5个特征作为纹理特征，再利用高斯混合模型进行纹理图像分割，实验结果表明，该新算法的分割精度和分割效率较原分割算法都有较大提高。     

Luminance contrast and chromaticity contrast preference on the colour display for young and elderly users

The human visual system changes with aging and one of the most important changes is the decrease of spatial contrast sensitivity. We investigated this change both for luminance contrast and chromaticity contrast, and both for threshold contrast and preferred contrast, (preferred by users to carry out a visual recognition task), in a series of psycho-physical experiments with achromatic and chromatic sinusoid gratings of different values of spatial frequency, hue, and luminance level, and with two observer groups: young and elderly observers. We investigated the spatial frequency range of 0.1–10 cycles per degrees. Our results indicate that, beyond the expected luminance contrast sensitivity decline of the elderly observers, the difference between the preferred luminance contrast of the elderly and the preferred luminance contrast of the young is even more significant than the threshold difference. The small preference differences between the age groups for chromaticity contrast compared to luminance contrast suggests that while with increasing age both the chromatic and the achromatic contrast sensitivity drops, preferred contrast stays more stable for chromaticity contrast than for luminance contrast.     

Contrast of a fuzzy relation

In this paper we address a key problem in many fields: how a structured data set can be analyzed in order to take into account the neighborhood of each individual datum. We propose representing the dataset as a fuzzy relation, associating a membership degree with each element of the relation. We then introduce the concept of interval-contrast, a means of aggregating information contained in the immediate neighborhood of each element of the fuzzy relation. The interval-contrast measures the range of membership degrees present in each neighborhood. We use interval-contrasts to define the necessary properties of a contrast measure, construct several different local contrast and total contrast measures that satisfy these properties, and compare our expressions to other definitions of contrast appearing in the literature. Our theoretical results can be applied to several different fields. In an Appendix A, we apply our contrast expressions to photographic images.     

Efficient Mining of Discriminating Relationships Among Attributes Involving Arithmetic Operations

Contrast patterns describe differences between two or more data sets or data classes; they have been proven to be useful for solving many kinds of problems, such as building accurate classifiers, defining clustering quality measures, and analyzing disease subtypes. This article investigates the mining of a new kind of contrast patterns, namely discriminating inter‐attribute functions (DIFs), which represent arithmetic‐expression‐based inter‐attribute relationships that distinguish classes of data. DIFs are an expressive and practical alternative of item‐based contrast patterns and can express discriminating relationships such as “weight/(height)² is more likely to be ≤25 in one class than in another class.” Besides introducing the DIF mining problem, this article makes theoretical and algorithmic contributions on the problem. We prove that DIF mining is MAX SNP‐hard. Regarding how to efficiently mine DIFs, we present a set of rules to prune the search space of arithmetic expressions by eliminating redundant ones (equivalent to some others). We give two algorithms: one for finding all DIFs satisfying given thresholds and another for finding certain optimal DIFs using genetic computation techniques. The former is useful when the number of attributes is small, whereas the latter is useful when that number is large; both use the redundant arithmetic‐expression pruning rules. A performance study shows that our techniques are effective and efficient for finding DIFs.     

Chromaticity contrast,luminance contrast,and legibility of text

Abstract— The purpose of this investigation was to determine if color differences can improve legibility and compensate for insufficient luminance contrast. Several authors previously have answered this question positively on the basis of performance experiments. We argue that a valid answer cannot be given unless subjective data are used. In four experiments, subjective ratings were collected pertaining to multicolor CRT text displays. The displays were systematically varied with respect to both luminance contrast and chromaticity contrast. The results indicate that chromaticity contrast and luminance contrast are additive only under specific conditions. Chromaticity contrast cannot improve legibility if an acceptable level of luminance contrast is present. Generally, chromaticity contrast cannot be substituted for luminance contrast.     

遥感图像理想均衡化及图像质量定量评价

大多数原始的遥感影像由于其灰度分布集中在较窄的范围内,影像的细节不够清晰,对比度较低。为了使影像的灰度范围拉开或使灰度均匀分布,从而增大反差,增强影像细节信息,通常采用的方法为直方图均衡化。通过对信息熵定义的阐述,引出直方图均衡化的图像增强算法。通过分析传统直方图均衡化算法中存在的缺陷,进而基于分段映射思想提出一种改进的理想直方图均衡化算法。同时,为了对传统算法和改进算法进行定量化地分析比较,基于同时对比度以及人类视觉对比度分辨率限制和模糊数学的相关思想,分别提出基于加权几何平均数法的合成平均对比度和细节评价参数的定义。最后,采用同时对比度、基于加权几何平均数法的合成平均对比度以及细节评价参数作为定量评价的指标,对所提出的改进算法进行了定量评价。评价结果表明,该改进算法的图像增强效果优于传统的直方图均衡化算法。     

Assessment of drivers' visual perception of information displayed in LED traffic signs at road construction sites

This study investigated drivers' visual perception of information displayed in LED traffic signs under different levels of environmental luminance. Concerning the information displayed, the environmental luminance comprised four luminance contrasts, two arrow types, and three environmental luminances, which were used to identify the visibility of traffic signs at the reading distance. Regarding visible distance, visibility involved reading starting and end points, extent of glare, comfortable distance, and glare distance. According to the study results obtained by using a highly reliable glare assessment method and studying the LED traffic signs, the lower the luminance contrast of traffic signs was the higher glare the participants perceived. A luminance contrast of 6200:2066 cd/m² provided the farthest comfortable and glare distances, enabling drivers to notice the signs comparatively earlier and have sufficient time to react accordingly. Overall, arrow type “<<<” outperformed “←” because the former was easier for drivers to read and created less glare. Regarding environmental luminance, traffic signs displayed in high and medium luminance environments (i.e., extremely sunny and cloudy days, respectively) each featured their own advantage, whereas those displayed in low luminance environments (nights) have to be further improved.     

非均匀光照图像的子块部分重叠对数变换

在非均匀光照下获得的图像往往具有高动态范围。大多数现有的全局对比度增强算法对此种图像效果并不好。一种简化的对数变换公式被提出来解决这一问题,更进一步子块部分重叠方法也被采纳进来实现局部的对比度增强。为了达到优化的目的,两种对比度测度用来寻找优化的变换参数。这些方法合并成一种新的算法POSLT（子块部分重叠的对数变换）。POSLT尽管计算复杂性比较高,但是展示出了良好的性能和可控性。实验结果也显示出了该算法在对比度增强中的局部适应性,有力支持了POSLT。     

打卡图像的分层人脸比对识别方法研究

具有身份的人脸图像比对要求具有高的识别率和实时性。本文针对打卡人脸图像,提出了一种基于加权模板匹配和SVM的分层人脸识别方法。该方法利用Gabor小波变换进行人脸图像特征提取,采用贡献分析法分析特征的贡献权重,在待测人脸图像比对识别时,采用加权模板匹配进行比对,通过两个阈值的判断,在既不能认为比对正确和不正确的情形下,再采用SVM和库中人脸图像进行识别比对。实验结果表明,基于该方法的人脸比对识别率高、实时性好,可用于实时打卡人脸比对。     

Text – background polarity affects performance irrespective of ambient illumination and colour contrast

In a series of experiments, proofreading performance was consistently better with positive polarity (dark text on light background) than with negative polarity displays (light text on dark background). This positive polarity advantage was independent of ambient lighting (darkness vs. typical office illumination) and of chromaticity (black and white vs. blue and yellow). A final experiment showed that colour contrast (red text on green background) could not compensate for a lack of luminance contrast. Physiological measures of effort and strain (breathing rate, heart rate, heart rate variability and skin conductance level) and self-reported mood, fatigue, arousal, eyestrain, headache, muscle strain and back pain did not vary as a function of any of the independent variables, suggesting that participants worked equally hard in all experimental conditions, so that the interpretation of the primary performance measure was unlikely to be contaminated by a performance–effort trade-off.     

VB中"调节图形对比度”功能的实现方法

使用Windows系统的API函数和VB编程环境,编制了调节图形对比度的VB程序。