基于DCT变换的图像融合方法研究

提出了一种基于离散余弦变换(DCT)以及一种结合小波变换与DCT变换的图像融合新方法。前者将源图像进行分块DCT变换,依据DCT系数的高频能量,对源图像的对应区域进行融合。后者利用DCT系数的高频能量对小波分解后得到的低频子图进行融合,同时以此为依据对小波最高分解层的小波高频系数进行选择,其他分解层的小波高频系数依据最大局部方差准则进行融合。依照平均误差、峰值信噪比以及均方根误差等客观评价标准,将新方法与其他常用的基于小波变换或DCT变换的融合方法进行了比较。实验结果表明,结合小波变换与DCT变换的图像融合新方法获得的融合效果优于其他方法。该方法与常用的基于小波变换的融合方法相比,其平均误差减少了40.8%~69.5%,峰值信噪比提高了9.9%~15.6%,均方根误差减少了34.8%~47.5%,评价结果与目视效果相吻合,表明该方法能有效地提高图像融合的质量。基于DCT变换的图像融合新方法的融合效果仅次于结合小波变换与DCT变换的图像融合新方法且其计算量相对较少,适用于实时处理。     

PCA-based method for 3D shape recovery of microscopic objects from image focus using discrete cosine transform

This article introduces a new algorithm for shape from focus (SFF) based on discrete cosine transform (DCT) and principal component analysis (PCA). DCT is applied on a small 3D neighborhood for each pixel in the image volume. Instead of summing all focus values in a window, AC parts of DCT are collected and then PCA is applied to transform this data into eigenspace. The first feature, containing maximum variation is employed to compute the depth. DCT and PCA are computationally intensive; however, the reduced data elements and algorithm iterations have made the new approach competitive and efficient. The performance of the proposed approach is compared with other methods by conducting experiments using image sequences of a synthetic and two microscopic objects. The evaluation is gauged on the basis of unimodality, monotonicity, and resolution of the focus curve. Two other global statistical metrics, root mean square error (RMSE) and correlation have also been applied for synthetic image sequence. Besides, noise sensitivity and computational complexity are also compared with other algorithms. Experimental results demonstrate the effectiveness and the robustness of the new method.     

Automatic defect inspection for LCDs using singular value decomposition

Thin film transistor liquid crystal displays (TFT-LCDs) have become increasingly popular and dominant as display devices. Surface defects on TFT panels not only cause visual failure, but result in electrical failure and loss of LCD operational functionally. In this paper, we propose a global approach for automatic visual inspection of micro defects on TFT panel surfaces. Since the geometrical structure of a TFT panel surface involves repetitive horizontal and vertical elements, it can be classified as a structural texture in the image. The proposed method does not rely on local features of textures. It is based on a global image reconstruction scheme using the singular value decomposition (SVD). Taking the image as a matrix of pixels, the singular values on the decomposed diagonal matrix represent different degrees of detail in the textured image. By selecting the proper singular values that represent the background texture of the surface and reconstructing the matrix without the selected singular values, we can eliminate periodical, repetitive patterns of the textured image, and preserve the anomalies in the restored image. In the experiments, we have evaluated a variety of micro defects including pinholes, scratches, particles and fingerprints on TFT panel surfaces, and the result reveals that the proposed method is effective for LCD defect inspections.     

自适应量化嵌入数字水印的音频隐密通信方法

本文提出了一种基于数字水印的音频隐秘通信方法。鉴于隐藏效果的直观性。原始水印信息以二值图像为例,也可为其他类型如语音。该算法首先对有具体含义的原始水印信息进行预处理。使之成为适于嵌入音频的一维数宇水印序列;然后对音频载体分段,根据各段能量大小选定隐藏段。再对各段进行离散余弦变换（DCT）;考虑到不可见性和鲁棒性的折中。本文依据噪声模型确定的噪声敏感度因子在DCT域的中低频系数区选定具体嵌入位置;并采用量化嵌入方法,可实现盲提取。实验结果表明该方法具有良好的不可见性和鲁棒性,能抵抗滤波、重采样、重量化、加噪声等常规信号处理攻击。     

低比度钢板微小缺陷的图像增强和分割

针对低对比度、微小的钢板表面缺陷,提出了基于图像增强的图像分割算法来有效分割和识别缺陷目标。采用小波变换与同态滤波结合算法对图像进行增强处理,不仅消除照度不均的影响,还突出了缺陷细节的信息,达到图像增强的效果。最后,利用粒子群算法优化最大类间方差法参数（PSO-Otsu）确定增强后图像的最佳阈值,并结合Canny算子进行缺陷检测。对比其他算法,该算法在检测低对比度的微小缺陷上取得了良好的效果。     

Filtering of images of small-sized objects in systems with circular microscaning

This paper describes a model for detection of images obtained in circular microscanning. To detect objects on a low background with the use of an image detection system based on circular microscanning, an approach to image filtering is proposed. The calculation of the noise covariance matrix with respect to the data obtained by the numerical simulation method is used. The calculated matrix is used in filter construction. It is shown that the applied approach makes it possible to increase the signal/noise ratio in image processing with a low background component.     

改善激光共焦扫描显微镜像质的方法

为解决传统光学显微镜样本上每一点的图像都受到邻近点衍射或散射光干扰的问题,研发了一套基于C#WinForm控制平台进行连续扫描方式的激光共焦扫描显微镜(LCSM)系统,并且成功地对生物细胞进行了扫描成像。针对共焦显微镜图像像质不高的问题,提出合理选取探测器针孔直径,并通过高斯低通滤波、盲解卷积的方法,确保实现高像质。实验结果表明,基于上述方法改进后的LCSM具有较高图像质量,该方法简单易行,便于实施。     

基于超像素的感知哈希交通场景图像去重方法

针对构建典型交通场景样本库采集数据过程中存在数据重复、相似的问题,提出了一种基于超像素分割下感知哈希的交通场景图像去重算法。首先,对图像超像素分割;然后记录每个超像素区域的像素均值作为图像特征,在分割后的图像中提取像素点组成整幅图像的描述集,并对其进行离散余弦变换（DCT）采用Zigzag模式顺序对变换的系数矩阵编码作为图像DCT特征;最终,采用Jaccard 距离衡量图像相似度,根据权重确定个特征对图像相似度的贡献,确定合理的阈值建立图像去重系统。在KITTI、DeepTesla等数据集上选择部分图像进行实验,实验结果表明,该方法对图像去重的准确率达98.55%,同时具有较高的鲁棒性和稳定性。     

Image noise reduction using Wiener filtering with pseudo-inverse

We present a preliminary design and experimental results of a Gaussian noise reduction method for ultrasound images. Our method utilizes a Wiener filtering algorithm with pseudo-inverse technique. The method is capable of solving the Gaussian noise problem in ultrasound image by setup a constant dB of noise function. The key idea of the Wiener filtering algorithm is to process the given ultrasound signal by making the filtering less sensitive to slight changes in input conditions. In this paper, we investigate the possibility of employing this approach for pre-processing ultrasound image application. The application of the proposed method for reducing Gaussian noise is demonstrated by four examples. Meanwhile, we also made the comparisons with median filter, mean filter and adaptive filter; the results reveal that the proposed method has the best noise filtering capability than other three methods. The results also show that the proposed method produces recovery images with quiet high peak-signal-to-noise ratio.     

基于小波变换的人脸光照补偿

提出了一种用于非均匀光照条件下人脸识别的光照补偿算法。该算法首先将人脸图像变换到对数域,并在对数域中计算2维小波变换,通过舍弃低频子带图像中的系数来实现人脸图像的非均匀光照补偿。由于人脸光照补偿的目的是为了提高人脸识别性能,所以光照补偿的效果利用人脸识别率来表征。在Yale B人脸库中,与对数域离散余弦变换（DCT）光照补偿算法进行了比较,实验结果表明,本文方法的人脸识别平均误识率可以达到0.18%,比对数域DCT方法具有更好的性能。另外,在CAS_PEAL人脸库中的实验结果表明,本文方法的性能与对数域DCT方法相近。     

基于对比度敏感度的无参考图像清晰度评价

提出一种基于对比度敏感度的无参考图像清晰度评价方法,用于评价光测设备无参考电视图像的清晰度.首先通过低通滤波器构造原始图像的参考图像,并对两者进行离散余弦变换(DCT),其次将变换后的中频系数和高频系数各分为高低两档后进行离散余弦反变换处理得到各自子图像,然后计算相应子图像的结构相似度SSIM,最后将各子图像结构相似度...     

Automated Surface Inspection Using Gabor Filters

In this paper we present a machine vision system for the automatic inspection of defects in textured surfaces found in industry. The defects to be inspected are those that appear as local anomalies embedded in a homogeneous texture. The proposed method is based on a Gabor filtering scheme that computes the output response of energy from the convolution of a textured image with a specific Gabor filter. The best parameters of a Gabor filter are selected so that the energy of the homogeneous texture is zero, and any unpredictable defeats will generate significantly large energy values. A simple thresholding scheme then follows to discriminate between homogeneous regions and defective regions in the filtered image. This transforms texture differences into detectable filter output. The experiments on structural textures such as leather and sandpaper have shown the effectiveness of the proposed method.     

基于小波和灰度共生矩阵的带钢表面缺陷识别

提出了一种基于小波变换和灰度共生矩阵的带钢表面缺陷识别方法。采用小波变换分解缺陷图像并提取其低频子带信息。通过在低频子带上构造0°、45°、90°和135°四个方向的灰度共生矩阵,分别计算角二阶矩、熵、对比度和逆差矩4个特征值,共获得16个特征值,并将其输入支持向量机,完成对6类共1800张带钢表面缺陷图像的识别,总体识别精度大于96%。实验结果表明,小波变换与灰度共生矩阵结合能有效描述带钢表面缺陷纹理特征,具有较好的识别效果。     

基于视觉的铁路货车滚动轴承表面缺陷分类研究

用CCD代替人眼对轴承表面缺陷进行图像采集,采用卷积滤波与开、闭运算相结合的图像处理方法,有效去除了缺陷周围边缘点的干扰。在提取传统特征基础上增加了压缩度、线度、距离极值比、NMI特征和不变矩等特征量,增强了缺陷分类的依据;对BP神经网络的输入矩阵和归一化方法的改进,提高了神经网络的记忆能力及识别速度;通过试验对缺陷分类系统识别结果进行检测,确定了该系统的可靠性。     

基于混沌映射与差分进化自适应教与学优化算法的太赫兹图像增强模型

为消除功率起伏效应引起的太赫兹(THz)图像局部伪影,构建了基于同态滤波的THz图像增强模型.然而,模型各参数取值差异大且耦合性强,给其参数确定带来了困难.为此,本文提出了混沌映射与差分进化自适应教与学优化算法以求解增强模型最优参数.首先,改进了标准Logistic混沌映射,提高了种群多样性.其次,引入适应度更新率,构...     

Automatic optical inspection for detecting defects on printed circuit board inner layers

This paper studies automatic optical inspection for detecting defects on the printed circuit board inner layer. The development of this study can be divided into five stages, they are reference image rebuilding, inspection image normalization, image subtraction, defects separation and defect classification. In the image subtraction stage, the difference between the reference image from the printed circuit board design and the inspected image is checked for defects. Each defect region is separated using a defect outer boundary tracing method. A boundary state transition method is proposed to classify the defect types. This system can recognize eight defect types, open, mouse bite, pinhole, missing conductor, short, spur, excess copper and missing hole. In addition, a comparison with the methods described in the literature is made, proving that the proposed method produces better results .     

Automatic surface inspection for directional textures using nonnegative matrix factorization

A global image restoration scheme using nonnegative matrix factorization (NMF) is proposed in this paper. This NMF-based image restoration scheme can be used for inspecting the defects in directional texture surfaces automatically. Decomposing the gray level of image pixels into an ensemble of row vectors, we first reduce the data set from original data space into a lower-dimensional NMF space. The repetitive and periodical primitives are well reconstructed by two lower-dimensional basis and weight matrices with nonnegative elements, named nonnegative matrix approximation (NMA). Then the local defects will be revealed by applying image subtraction between the original image and the NMA. As a consequence, the directional textures are eliminated, and only local defects are preserved if they initially are embedded in the surface. A supervised heuristic, elbow of residual curve rule, is devised which helps users to determine a proper basis space size of a specific image. Experiments on a variety of directional texture surfaces are given to demonstrate the effectiveness and robustness of the proposed method.     

基于分块DCT变换的数字盲水印算法的应用

为了解决图片类信息加密易破解的问题,更好地对相关产品的版权加以保护,采用一种基于分块离散余弦变换(DCT)的数字盲水印算法。图像经分块后先进行DCT变换,从空域变换到频域,然后进行水印的嵌入和提取,最后与基于小波变换域的数字水印算法的实验进行对比。研究表明,该算法在峰值信噪比为52.588 3 dB时,可以为图像提供良好的水印透明性,且对噪声、剪切等攻击有明显的抵抗效果,同时具有良好的鲁棒性。     

发动机缸体视觉图像定位方法研究

针对发动机缸体孔系实现在线自动测量的难题,提出了基于视觉图像的发动机缸体定位方法。通过面阵CCD获取发动机缸体定位销孔图像信息,利用图像处理提取定位销孔中心位置,以此建立每个被测缸体的测量基准。使用环形双峰阈值法与中值滤波对图像进行预处理,有效地解决了定位销孔具有倒角影响图像边缘提取精度的问题。为验证此定位方法的准确性与可行性,分别进行了发动机缸体定位孔直径与位置度重复性测量与发动机缸体上表面孔组直径与位置度重复性测量实验。实验结果表明所提出的视觉定位方法稳定可靠,实现了发动机缸体的快速、精确定位,为实现发动机缸体的在线检测打下了可靠基础。     

结合多尺度圆形滤波与MS聚类 的疑似结节分割

肺部CT图像中疑似病灶感兴趣区域(ROI)的准确分割是肺部计算机辅助检测/诊断(CAD)的重要环节。本文提出结合Hessian矩阵滤波的均值漂移聚类肺部疑似病灶ROI区域分割算法。对原图像进行多尺度Hessian矩阵圆形滤波,图像中圆形的疑似结节病灶区域得到滤波增强、直线形的气管/血管区域得到抑制,将Hessian矩阵滤波后的形状特征、灰度、空间位置3种信息引入特征空间,将均值漂移聚类的核函数分解为3种特征信息所分别对应的核函数乘积形式,最后采用自适应计算带宽的方法确定每个待分割疑似区域的带宽进行均值漂移聚类分割。对来自LIDC等127个包含不同类型肺结节的病例进行实验,实验结果表明引入Hessian矩阵圆形滤波信息的均值漂移聚类能够分割出与血管或气管相连或者交叉的结节区域,去除ROI中包含的非结节区域,能有效分割出基于灰度信息难以分割的毛玻璃型(GGO)结节;对于3种类型的结节区域:血管相连结节(VPN)、毛玻璃型结节(GG0)、孤立性结节(SPN)分割平均准确率分别为92.80%、86.13%、95.08%。