基于最小平均距离免疫算法的模糊红外图像分割

由于犯罪分子利用各种方法来避开传统的刑侦图像技术,因而红外图像逐渐成为获取犯罪现场痕迹的有效手段。然而,从犯罪现场拍摄的红外图像其目标痕迹大多是弱化的,所以在这类红外图像中分割目标是一项具有挑战性的任务。已有基于生物免疫的各类算法尚未明确描述免疫分割作用领域,以及免疫网络算法模型中的免疫识别距离。为实现对目标痕迹弱化红外图像的有效分割,提出了一种新的具有免疫作用领域和最小平均免疫识别距离的人工免疫构架,设计了一种具备最小平均距离免疫域的免疫分割算法。该方法根据红外图像的特点,采用多步分类算法、免疫变异和自适应免疫最小均距识别方法,根据目标区域和背景区域的总体统计特性实现最佳分类。实验结果表明,提出的基于最小平均距离的免疫算法能够有效地分割目标弱化的红外图像。与经典的边缘模板和区域模板方法相比,该算法具有更好的分割效果,尤其是针对目标弱化红外图像的分割,该算法能够较好地给出五个手指的边界轮廓。     

基于WT算法联合混沌理论的图像加密研究

图像加密作为信息加密领域的重要一支,其对于信息安全的重要性显得愈发重要,能够有效地对目标图像信息进行加解密逐步成为了人们的研究热点。为了提高图像加密的安全性,以混沌系统所具有的初值敏感性以及类似随机为基础,提出了采用“混沌变换”方法对图像进行置乱操作的算法,随后以此为基础结合小波理论设计一种图像加密算法。在图像的预处理阶段首先对图像采用小波变换得到四幅小波子图;随后基于混沌置换将四幅子图置乱处理;最后通过小波逆变换恢复出目标加密图像。通过数值仿真实验表明通过该方法解密获得的图像具有与原图像非常高的一致性,并且获得了较高的安全性。     

三次相位板波前编码系统彩色图像恢复的迭代算法

采用三次相位板进行景深延拓的波前编码系统得到非对称的点扩展函数.为了获得最终清晰的彩色图像.研究了一种基于广义极小残差法(GMREs)的迭代算法,结合Tikhonov规整化方法,并利用多通道处理过程对中间图像进行左卷积恢复.为了消除恢复图像边界的振铃效应,推导了新的光学成像过程数学模型,该模型采用反镜像边界条件并利用直积近似对卷积核进行处理.模拟数据的分析表明,采用多通道处理过程对彩色图片进行恢复时,新的算法在给出精确的反巷积结果的同时能有效地抑制噪声的放大;实验结果显示,较之经典的维纳滤波恢复结果,新算法能够更好的消除边界的振铃和图像边缘的振动波纹.     

Blind motion deblurring using multiple images

Recovery of degraded images due to motion blurring is a challenging problem in digital imaging. Most existing techniques on blind deblurring are not capable of removing complex motion blurring from the blurred images of complex structures. One promising approach is to recover the clear image using multiple images captured for the scene. However, in practice it is observed that such a multi-frame approach can recover a high-quality clear image of the scene only after multiple blurred image frames are accurately aligned during pre-processing, which is a very challenging task even with user interactions. In this paper, by exploring the sparsity of the motion blur kernel and the clear image under certain domains, we propose an alternative iteration approach to simultaneously identify the blur kernels of given blurred images and restore a clear image. Our proposed approach is not only robust to image formation noises, but is also robust to the alignment errors among multiple images. A modified version of linearized Bregman iteration is then developed to efficiently solve the resulting minimization problem. Experiments show that our proposed algorithm is capable of accurately estimating the blur kernels of complex camera motions with minimal requirements on the accuracy of image alignment. As a result, our method is capable of automatically recovering a high-quality clear image from multiple blurred images.     

A composite (stacked) picture generation technique for spectral profile representation of dye laser

In this paper, a new technique for data representation format from video recording of spectral profile of tunable dye laser has been formulated. Only a part of the acquired images acquired sequentially in programmed time interval, was used to make the composite picture to measure variation of the parameters. A CCD camera based imaging system was used to confine the spectral pattern. No additional hardware other than image capturing device was used. The technique developed reduces the storage space required in the computer and also enhances the measurement speed significantly. The software algorithm developed has been used effectively to measure various parameters of tunable dye lasers and other optical sources.     

Optical rotation invariant corner detector

In this work we introduce a new approach for corner extraction with rotation invariance. The method is based on an optical processor where the filter in the frequency plane is composed by a spiral phase function and a binary amplitude. The output image of the optical setup presents intensity peaks in each square corner location. The performance of the proposed method was tested on synthetic scenes by numerical simulations and by optical measurements. The presented results show that the method allows the corner extraction, even with rotated scenes.     

Simultaneous surface area measurement for multiple objects by geodesic active contour

A novel scheme to the area measurement of multiple objects is proposed. This scheme is suitable for the simultaneous and exact measurement of multiple objects’ surface areas with different shapes. A geodesic active contour is applied and this scheme consists of three steps. Firstly, the active contour converges to the object's contour edge from an initial position via the iteration of level set functional ?. Secondly, for the convergent active contour, the sub-pixel area measurement and centroid measurement are applied to compute every object's area and centroid simultaneously. Finally, a centroid-self-calibration technology is applied, which measures the pixel's size equivalence with the computed centroid and a dual-frequency laser to measure every object's true surface area. Experiments indicate that this scheme's measurement repetition error can decrease to ±0.4% and that its single measurement error can decrease to ±0.2%. Compared to conventional area measurement schemes, this scheme is highly efficient (i.e., simultaneous measurement of multiple objects’ areas) and highly accurate.     

基于数学形态学的金字塔图像去噪

提出了一种自适应提升拉普拉斯金字塔变换与数学形态算子相结合的图像去噪方法.首先对图像进行分解,采用自适应提升拉普拉斯金字塔,然后利用自适应提升拉普拉斯金字塔变换后系数能量的分布特性以及尺度内和尺度间的依赖性,结合数学形态算子的特点,利用数学形态算子对变换系数进行处理,使得重要变换系数与非重要变换系数分离,分别对两个不同的子集采用不同的阈值进行处理后,再将两个子集合起来,进行逆变换重建.实验结果表明,该算法具有良好的去噪效果,同时保留了图像的细节信息,获得了较高的图像信噪比.     

改进的总变分去噪算法

提出两个改进的模型用以解决标准TV模型在处理纹理丰富的图像时易丢失重要信息以B用梯度检测边缘时易受噪音干扰的缺点.模型A是在标准TV模型的基础上作了两点改进,一是引入边缘检测函数来引导扩散,二是利用小波变换的模来检测边缘.这使得新模型不但根据图像的特征进行平滑,并具有较强的抗噪能力,从而能更好地保护边缘.模型B是基于噪音在小波域中的特性对模型A在计算复杂度上的简化.数值试验表明,这两个模型均比TV模型具有较好的性能.     

基于改进单高斯模型法的交通背景提取

在分析交通背景提取的特点和单高斯模型法的不足的基础上,提出了一种改进的单高斯模型法.该算法利用均值法初始化背景模型,引入判断值进行背景更新,运用邻域判别法实现干扰点抑制.不同天气条件,不同交通场景下的实验表明：与单高斯模型法相比,改进算法在提取背景时可以随车辆的走停及时更新,并且抑制了非静止背景物体的干扰.