背景渐变的视频对象分割算法研究及实现

提出了一种有效的背景渐变的视频对象分割算法.首先将前一帧分成前景和背景两部分,然后采用灰度投影匹配算法对当前帧进行全局运动估计和补偿,将当前帧与上一帧进行差分运算,便可得到差分图像.通过对差分图像进行二值化处理,得到运动模板并与前景信息进行相与计算,再结合当前帧信息便可得到运动目标.在TI公司的TMS320DM642芯片上验证了该算法,实验结果表明该算法不仅对亮度变化和环境变化具有鲁棒性,而且可独立、精确地分割出运动目标.     

应用粒子滤波及先验概率模型进行图像分割的新算法

基于粒子滤波在非线性非高斯情况下具有较好的预测结果,本文提出了一种自适应背景图像分割新算法,该算法利用粒子滤波对下一帧的前景区域进行预测,进而计算出下一帧各像素点属于背景的概率以指导下一帧图像分割;在前景像素值与背景像素值相近的情况下利用先验知识进行图像分割是一种较好的方法,本文以粒子滤波预测结果与先验概率模型计算结果的均值作为当前像素点属于背景的概率来进行图像分割,实验结果表明,该方法在背景变化范围较大的情况下,可以减少前景点误分割为背景点的概率.     

基于复杂场景下特征融合与估计的目标跟踪算法

针对图像跟踪过程中的背景干扰、光照条件变化与目标特性漂移等问题,提出一种粒子滤波框架下灰度和纹理特征融合与估计的目标跟踪算法.首先建立基于灰度直方图和边缘梯度直方图的特征集合,以增加目标描述的信息量和稳定性;然后在跟踪过程中依据每个特征对目标与背景的区分度赋予其相应的权值,突出贡献较大的特征,并将该加权系数用于目标和候选目标的相似性度量;最后,为了克服复杂场景对单幅图像中特征可信度的影响,将跟踪的连续性引入权值的自适应调整过程,利用当前帧的特征区分度和前一帧的特征权值对当前帧的特征权值进行估计,从而实现特征权值的在线更新.实验结果表明,通过对特征的融合及其权值的在线估计,实现了复杂场景下的稳定目标跟踪.     

基于邻域统计分布变化分析的UWB SAR隐蔽目标变化检测

该文针对载机不同航迹条件下所得多时相UWB SAR图像灰度值存在较大起伏,严重影响了基于像素灰度值差异的变化检测算法性能,提出了一种基于邻域统计分布变化分析的UWB SAR隐蔽目标变化检测方法。该方法将Gram-Charlier展开理论同秩序滤波器相结合对多时相图像中每个像素邻域的统计分布进行估计,进而借助K-L散度理论对多时相图像邻域统计分布变化进行定量分析以检测目标对应的变化区域。实验结果表明,该文方法能够更好地适应不同航迹UWB SAR图像间灰度起伏的影响,取得更好的检测结果。     

多参数亮度值重映射的颜色传递方法

为了提高灰度图像中人眼视觉的情景感知和目标的识别率,提出了多参数亮度值重映射的颜色传递方法,将灰度图像赋予色彩。该方法首先将灰度目标图像和彩色参考图像由RGB颜色空间转换到亮度与色彩分离的YCbCr颜色空间。然后,引入由彩色参考图像和灰度目标图像的列数和亮度共同决定的参数,对彩色参考图像的亮度值进行运算,使彩色参考图像与灰度目标图像具有相同的亮度值动态范围。最后,根据目标像素和它周围的像素存在相关性,进行灰度目标图像与彩色参考图像的像素匹配后,将得到的结果再逆变至可以表示出各种色彩的RGB颜色空间。该颜色传递方法降低了色彩误传现象,提高了颜色传递的准确度并缩短了程序的运行时间。     

红外运动目标的灰度分层检测算法

为了减少复杂背景下红外序列图像运动目标检测中的漏检与误检,采用了基于灰度分层的瞬时帧差检测算法。该算法根据红外图像的特点,使用自适应门限将当前帧图像分为两层,采用基于区域的像素间差分代替传统的基于空间坐标的像素间差分,通过对高灰度层检测时得到的高置信度的检测结果去引导对低灰度层的检测。实验表明,该算法可对多类多个运动目标进行有效检测,且具有较高的实时性与鲁棒性。     

改进的高斯背景建模在车辆检测中的应用

研究实时环境下运动车辆的检测算法,针对存在渐变及重复性运动的车辆运动背景采用改进的高斯背景建模法,对背景进行更新,将所得的当前背景与前一帧视频图像进行相减,得到当前时刻车辆的运动图像。将所得图像进行数学形态学去噪处理得到较为理想的车辆检测效果图,并根据处理后的图像判断车辆是否运动以及其运动轨迹。实验证明该方法能去除噪声对图像产生的影响,对判断某一时刻车辆是否运动行之有效。     

基于Mean Shift的抗遮挡运动目标跟踪算法

采用Mean shift和Kalman滤波器相结合来处理动态背景下目标跟踪问题.首先利用Kalman滤波器进行预估计获得每帧Mean Shift算法的起始位置.由图像差分法得到物体轮廓.同时定义了相似因子判断物体是否发生遮挡.当发生遮挡时,根据物体运动状态不同.对颜色信息和运动信息分别赋予不同权值来预测物体在当前帧的位置并作为下一帧预测的起点.此时,目标位置的线性预测替代了Kalnlan滤波器的作用.实验证明,新算法可实现对快速运动目标的跟踪,对遮挡也有很好的稳健性.     

嵌入Kalman滤波的Camshift人体跟踪研究

针对复杂背景下Camshift算法跟踪运动人体容易丢失目标的情况,提出了一种Kalman滤波和Camshift算法相结合的改进算法。Camshift算法利用颜色直方图做反向投影得到色彩的概率分布图,利用初始化的搜索窗口和位置并结合上一帧跟踪结果自适应调整跟踪窗口从而跟踪人体。采用Kalman滤波可以对运动人体进行估计以克服复杂背景下色彩的干扰,同时对Camshift的迭代结果进行校正。实验表明,本文方法在复杂背景下能更好地跟踪运动人体目标。     

一种改进的红外弱小目标快速检测方法

针对复杂背景下红外弱小目标检测率低、目标跟踪困难的问题,提出一种改进的红外弱小目标快速检测方法。该方法采用改进的形态学滤波抑制背景噪声,对处理后的多帧图像进行方差估计初步突出目标像素,然后对其进行信噪比估计得到整个图像序列像素得分,图像中像素信噪比高的被标记为目标像素,再对标记过的图像进行分块分析,最终准确提取出连续图像序列中的目标像素。检测出的目标像素作为Hough变换的目标跟踪算法的输入,设置双阈值实现目标的有效跟踪。实验结果表明,在复杂背景下的红外弱小目标提取中,基于噪声方差估计的目标检测拥有较高的检测概率和较低的虚警概率,将其获得的目标像素作为Hough变换的输入,不仅可以有效跟踪目标,而且简化了算法的复杂度,实现目标的快速提取和跟踪,具有很高的应用价值。     

奇异值分解域差异性度量的低景深图像显著性目标提取方法

针对低景深图像(DOF)目标提取过程中,容易出现目标提取不完整或背景被误识为目标等现象,该文提出一种奇异值分解 (SVD)域差异性度量的低景深图像目标提取方法。先对低景深图像进行高斯模糊,以图像中每一个像素点为中心,利用滑动窗口分别截取模糊前后图像上相同位置的图像块并进行奇异值分解,再构造两奇异值之间的差异特征向量,针对此向量定义低中高全频段信息加权的差异性度量算子,计算对应像素点的显著性特征值,逐像素处理得到显著性结果图并进行阈值化处理,实现低景深图像目标的有效提取。对大量低景深图像进行处理,并与几种现有方法进行比较,提出方法的F度量值最大可提高54%,平均绝对误差减少76%～87%,可完整提取目标并有效去除背景,具有较强的可靠性。     

一种有效的混合噪声滤波算法

以噪声特点和图像结构分析为基础,提出了一种有效的混合噪声滤除算法。算法首先通过极值判断和像素间的相容性检测,分离出脉冲噪声并以中值滤波滤除;然后对含有高斯噪声的图像以模糊滤波算法进行降噪处理。实验结果表明,本算法能有效地滤除图像中脉冲与高斯混合噪声,且较好地保护了图像细节特征。     

Adaptive-integration-time image sensor with real-time reconstruction function

A computational image sensor is proposed in which the pixel controls its integration time to light intensity. The integration time of each pixel is selected from among several lengths of integration time and the integration time is shortened if the pixel intensity becomes saturated. Although the integration time of each pixel varies, the pixel intensity is adjusted on the sensor in real time. The dynamic range of the pixel value output from the proposed sensor is greatly widened. A prototype of 64/spl times/48 pixels has been fabricated by using 2-poly 2-metal 0.8-/spl mu/m CMOS process. The proposed sensor has simple functions for the comparison of intermediate integration value and threshold to control the integration time and nonlinear image reconstruction. Because the maximum number of the comparison-reset operations during a frame is three, one of the four integration times can be selected pixel by pixel. The circuit and layout design of the prototype which has computational elements based on column parallel architecture are described and the fundamental functions have been verified. By the experiments, it has been verified that the sensor can achieve a wide dynamic range by adapting to light.     

利用Zernike矩对心脏超声图进行光流场计算的邻域优化算法

本文针对心脏超声图的特性 ,介绍了一种计算二维光流场的邻域优化算法 ,通过对Zernike矩的正交性、旋转不变性等特性的分析 ,论证了Zernike矩在心脏超声图中作为运动不变特征的优越性 ,从而将传统光流场计算中的亮度不变假设推广到了亮度各阶Zernike矩不变 .另外 ,由于Zernike矩计算了像素点附近其余像素的贡献 ,使得单个像素点引起的误差对特征的影响大大减小 ,从而提高了算法的鲁棒性     

DMD结合图像传感器的高动态场景成像探测

微镜阵列DMD空间光调制器结合图像传感器能够解决高动态场景中强弱目标同时探测的问题,获得高动态图像数据,能使光电成像设备动态范围扩展到130 dB以上。为了实现快速的像素级光强控制以及获得高质量的实际场景高动态图像数据,根据实际场景成像其非边缘区域的相邻像素值变化缓慢的特点,提出了一种适合DMD驱动方式的像素级调光算法-自适应空间区域法,它具有较快的收敛速度,最坏的情况下,经过N/I次即可完成调光权值的搜索;在以FPGA为驱动核心的高动态场景成像探测实验系统平台上,对DMD的驱动时序进行了分析研究。并通过此实验平台验证了自适应空间区域法的有效性,该方法能够实现对高动态场景中强弱目标的同时探测,具有很好的光强控制质量,对场景光强局部变化的适应性较强。     

Reversible data hiding using multi-pass pixel value ordering and prediction-error expansion

Pixel value ordering (PVO) prediction has become the most efficient method for high-fidelity reversible data hiding (RDH). In this approach, only the maximum and minimum of pixel block are predicted and modified to embed data and the preservation of pixel value order guarantees the reversibility. To achieve larger embedding capacity and superior performance, more blocks suitable for RDH are utilized in recent improved schemes. However, their performance is still unsatisfactory. In this paper, a novel RDH scheme is proposed by extending original PVO into multi-pass PVO embedding. Specially, the k largest or smallest pixels are taken as independent data bit carriers to fulfill k-pass PVO embedding. Although the pixel value order may change after data embedding, reversibility still can be guaranteed and image redundancy can be far better exploited. Moreover, embedding performance can be further enhanced by optimal combined embedding. Experimental results verify that the proposed scheme outperforms previous PVO-based schemes and some other state-of-the-art works.     

Efficient PVO-based reversible data hiding using multistage blocking and prediction accuracy matrix

In recent years, pixel value ordering based reversible data hiding has become a hot research topic for its high-fidelity. In this approach, only the maximum and minimum of pixel block are predicted and modified to embed data and the preservation of pixel values order guarantees the reversibility. So far, the optimal block size can only be exhaustively searched until Wang et al. propose the dynamic blocking strategy which enables the combination of two various-sized blocks. By further dividing flat block into four sub-blocks to retain larger embedding capacity, dynamic blocking can employ less high complexity blocks for a given embedding capacity. However, the lack of host image dependent automatic block classification mechanism still exposes the fact that their work is far from efficient and comprehensive. In this paper, to address this drawback and to better exploit image redundancy, a really efficient and more comprehensive blocking strategy namely multistage blocking is proposed. High efficiency lies in prediction accuracy matrix based thresholds determination, which enables infinitely extended multistage blocking in theory. The superiority of the proposed scheme is also experimentally verified.     

Analysis of dynamic magnetic resonance images

Dynamic magnetic resonance (MR) imaging with contrast agents is a very promising technique for studying tissue perfusion in vivo. A temporal series of magnetic resonance images of the same slice are acquired following the injection of a contrast agent into the blood stream. The image intensity depends on the local concentration of the contrast agent, so that tissue perfusion can be studied by the image series. A new method of analyzing such series is described here. Nonparametric linear regression is used for modeling the image intensity along the series on a pixel by pixel basis. After modeling, some relevant quantities describing the time series are obtained and displayed as images. Due to its flexibility, this approach is preferred to parametric modeling when pathology is present since this can induce a wide spread of patterns for the pixel image intensity along time. Results of the application of the method to series of dynamic magnetic resonance images from ischaemic rat brains after the injection of the susceptibility agent Sprodiamide Inj. (Dy-DTPA-BMA) are shown and compared to results from a related known method.     

An image pixel based variational model for histogram equalization

In this paper, we develop an image pixel based histogram equalization model for image contrast enhancement. The approach is to propose a variational model containing an energy functional to adjust the pixel values of an input image directly so that the resulting histogram can be redistributed to be uniform. This idea is different from existing histogram equalization algorithms where a histogram based on the input image is constructed, a mapping is determined to output a uniform histogram and then the pixel values of the input image are adjusted based on the mapping. In the variational model, a mean brightness term is incorporated to preserve the brightness of the input image, and a geometry constraint can also be added to keep the geometry structure of the input image. Theoretically, the existence of the minimizer of the proposed model, and the convergence of the proposed algorithm are given. Experimental results are reported to demonstrate that the performance of the proposed model are competitive with the other testing histogram equalization methods for several testing images.