Fast image enlargement using predefined codebook and eigenspace backprojection for lost pixel

In this paper, a new image enlargement method applying the backprojection for lost pixel (BPLP) to the predefined codebook-based method is proposed. BPLP is a method for image restoration. In BPLP, the eigenspace reflecting the characteristics of an input image is generated from the remained pixels and is used to restore the missing pixels. In the proposed method, the eigenspace is replaced by one generated from the predefined codebook (PDC). PDC represents edge-blurring properties in a small image patch and consists of pairs of low- and high-frequency image patches on various edge patterns. By replacing the PDC-based estimation of lost high-frequency components with BPLP, a fast image enlargement method retaining its performance can be developed. Through some experiments, the effectiveness of the proposed method was demonstrated. Especially, it was confirmed that the processing time of the proposed method was shortened to about 1/50 that of the PDC-based method.     

Image super-resolution based on local self-similarity

A new image super-resolution method based on a codebook mapping is proposed. The codebook mapping represents the internal relationship between low-and high-frequency image components, and is used for the estimation of highfrequency image components lost in the sampling process. In the proposed method, codebooks are first generated by the low-and high-frequency image components of the original image itself that is to be processed. Then, a resultant superresolution image, that is, an enlarged image, is obtained by combining the estimated high-frequency image components with the low-frequency ones obtained by a typical interpolation-based method. The effectiveness of the proposed method is verified by comparing it against some conventional methods.     

Hopfield neural network and its applications on image edge detection

A method of image edge detection using the Hopfield neural network (HNN) is proposed in this paper.The image edge parameters are introduced in detail, and the energy function of HNN is given based on the edge parameters. Tests on the image edge detection show that images detected by the proposed method have good edge closeness and true edge, at the same time it has good anti-noise performance. The image edge detection using HNN is better than that obtained by some other edge detection operators.     

PSF-based restoration with edge adaptive regularization for structured light measurement

Edge information is important for measurement based on structured light. The presented effort puts forward an image restoration method with edge regularization for structured light measurement, which detects the edge of image, then updates the parameters of image degradation model and restores results in iteration. The diffraction limit of optics and nonlinear distortion of sensors are calculated as prior knowledge for semi-blind deconvolution. Blur metric is introduced for constraints of deconvolution iterations. Images before and after restoration are sent to shape recognition and automatic calibration modules for comparison. From experimental results we can conclude that the proposed approach can effectively enhance image quality and edge details, so that greater precision can be achieved.     

Single-image super-resolution with joint-optimization of TV regularization and sparse representation

A super-resolution (SR) reconstruction framework is proposed using regularization restoration combined with learning-based resolution enhancement via sparse representation. With the viewpoint of conventional learning methods, the original image can be split into low frequency (LF) and high frequency (HF) components. The reconstruction mainly focuses on the process of HF part, while the LF one is founded simply by typical interpolation function. For the severely blurred single-image, we first use regularization restoration technology to recover it. Then the regularized output remarkably betters the quality of LF used in traditional learning-based methods. Last, image resolution enhancement with characteristic of edge preserving can implement based on the acquired relatively sharp intermediate image and the pre-constructed over-complete dictionary for sparse representation. Specifically, the regularization can favorably weaken the dependence of atoms on the course of degradation. With both techniques, we can noticeably eliminate the blur and the edge artifacts in the enlarged image simultaneously. Various experimental results demonstrate that the proposed approach can produce visually pleasing resolution for severely blurred image.     

Efficient greyscale image compression technique based on vector quantization

In this paper, a novel greyscale image coding technique based on vector quantization (VQ) is proposed. In VQ, the reconstructed image quality is restricted by the codebook used in the image encoding/decoding procedures. To provide a better image quality using a fixed-sized codebook, the codebook expansion technique is introduced in the proposed technique. In addition, the block prediction technique and the relatively address technique are employed to cut down the required storage cost of the compressed codes. From the results, it is shown that the proposed technique adaptively provides better image quality at low bit rates than VQ.     

红外气动退化图像复原校正的复合算法研究

鉴于气动效应图像退化因素的复杂性和随机性,单一复原算法难以满足要求,提出构造复合算法来提高湍流退化图像复原质量的思路,综合各种基于不同理论模型和应用对象的实用算法,实现优势互补。给出了现有算法结合的主要途径,主要探讨了基于统计理论的MAP复原算法与基于正则化理论的总变分算法复合后各算法结构的自适应调整和改造以及各种参数的自适应性选择问题。充分利用各算法的优势,完成基于算法结构相对稳定的智能化组合。在微机上进行了一系列的复原对比实验。实验结果表明：复合算法提高了图像的总体复原质量。     

结合边缘信息和图像特征信息的曲波域遥感图像融合

曲波变换是一种更适合于图像处理的多尺度几何分析方法,具有比小波变换更强的方向选择和辨识能力,而且对图像边缘的表达更优于小波.结合色度-饱合度-亮度变换将其应用于合成孔径雷达图像和多光谱图像融合可以更好地表示图像中的有用特征.首先对多光谱图像进行色度-饱合度-亮度变换,得到亮度分量Ⅰ,对雷达图像和Ⅰ分量进行曲波变换得到粗尺度系数和细节尺度系数;将雷达图像的粗尺度系数和细节尺度系数进行叠加,计算归一化的曲波系数直方图,定义边缘有效因子,利用合成孔径雷达图像的特征信息将曲波变换系数分为均匀区、非均匀区和亮点目标区.然后采用相应的融合规则对融合图像的粗尺度系数进行处理,对细节尺度系数采用简单的直接取大方法,逆变换后得到新的亮度分量.用新的亮度分量替代原亮度分量进行逆色度-饱合度-亮度变换得到最终融合结果,利用统计类指标对融合结果进行评价.实验结果表明,该方法在保持光谱信息和提高空间分辨率上都有较好的效果.     

结合边缘信息和图像特征信息的曲波域遥感图像融合

曲波变换是一种更适合于图像处理的多尺度几何分析方法,具有比小波变换更强的方向选择和辨识能力,而且对图像边缘的表达更优于小波.结合色度-饱合度-亮度变换将其应用于合成孔径雷达图像和多光谱图像融合可以更好地表示图像中的有用特征.首先对多光谱图像进行色度-饱合度-亮度变换,得到亮度分量I,对雷达图像和I分量进行曲波变换得到粗尺度系数和细节尺度系数;将雷达图像的粗尺度系数和细节尺度系数进行叠加,计算归一化的曲波系数直方图,定义边缘有效因子,利用合成孔径雷达图像的特征信息将曲波变换系数分为均匀区、非均匀区和亮点目标区.然后采用相应的融合规则对融合图像的粗尺度系数进行处理,对细节尺度系数采用简单的直接取大方法,逆变换后得到新的亮度分量.用新的亮度分量替代原亮度分量进行逆色度-饱合度-亮度变换得到最终融合结果,利用统计类指标对融合结果进行评价.实验结果表明,该方法在保持光谱信息和提高空间分辨率上都有较好的效果.     

空间变化PSF非盲去卷积图像复原法综述

传统的图像复原一般认为点扩散函数(PSF)是空间不变的,实际光学系统由于受到像差等因素的影响,并非严格的线性空间不变系统,基于空间变化PSF的非盲去卷积图像复原法逐渐体现其优越性。空间变化PSF的非盲去卷积图像复原法先准确估计图像空间变化的PSF,再利用非盲去卷积算法对图像进行复原,有利于恢复出高质量图像。本文从算法的角度综述了近几年提出的基于空间变化PSF的非盲去卷积图像复原方法,并对比了基于强边缘预测估计PSF的非盲去卷积法、基于模糊噪声图像对PSF估计非盲去卷积法等算法的优缺点,各算法分别在PSF估计精确度、振铃效应抑制效果、适用范围等方面体现出各自的优劣。空间变化PSF的非盲去卷积图像复原法的研究,有利于推进图像复原技术向更高水平发展,使光学系统往轻小型化方向发展,从而在多个科学领域发挥其重要作用。     

Region-specialized underwater image restoration in inhomogeneous optical environments

Underwater optical imaging and image processing play an important role in oceanic scientific research and practical applications. However, as light is absorbed and scattered by the water which it passes through, the underwater image acquired is seriously degraded. The resolution and contrast are lowered by light scattering and color is distorted by the wavelength-selective absorption. In order to solve these problems, various artificial sources are widely used in underwater imaging systems. However, additional sources except the parallel background light will cause pollution to the underwater optical environments, making it inhomogeneous. In this condition, the popular image processing methods based on the assumption of the homogeneous ambient light are weak or even futile. Aimed to solve this problem, a region-specialized method for underwater image restoration is proposed in this paper. Region-varied changes in the ambient light are taken into consideration for dehazing and color compensation. In contrast to the references including the state of art methods for both underwater image enhancement and restoration, the enhanced clarity and color fidelity are obtained through visual evaluation. Moreover, the highest score of no-reference perceptual quality assessment is obtained by the proposed method as well.     

CCD细分技术方法研究及应用

依据线阵CCD和面阵CCD细分技术的原理，提出了九点四线曲面拟合的面阵CCD细分方法，并通过与均值法对比以及在图像放大和图像定位中的实际应用，证明了该方法不仅计算简便、图像相关度高、边缘趋势保持较好，而且可大大提高图像的分辨率和目标图像的定位精度。     

A novel image fusion method using WBCT and PCA

A novel image fusion algorithm based on wavelet-based contourlet transform （WBCT） and principal component analysis （PCA） is proposed. The PCA method is adopted for the low-frequency components. Using the proposed algorithm to choose the greater of the active measures, the region consistency test is performed for the high-frequency components. Experiments show that the proposed method works better in preserving the edge and texture information than wavelet transform method and Laplacian pyramid （LP） method do in image fusion. Four indicators for the fusion image are given to compare the proposed method with other methods.     

基于边缘软判决的小波域自适应图像去噪

提出了一个新的图像去噪方法。该方法基于非抽样小波变换的多分辨分解,在各尺度下对小波系数进行了边缘和非边缘分类,并根据它们的不同统计特性运用了不同的估计技术。鉴于边缘分类的不确定性,提出了依概率的软分类技术,通过计算边缘发生的概率,判决当前系数应该采用哪一种估计。仿真结果表明:该方法在滤除图像噪声的同时,边缘得到了保持,较目前存在的一些方法更具有优越性。     

运动与离焦模糊图像的复原

在运动和离焦所引起的图像模糊的情况中,本文提出了一种新的基于霍夫变换区分离焦模糊和运动模糊两类模糊的方法.该方法通过比较霍夫变换矩阵中的亮点数来区分两类模糊,不仅正确率达到100%,而且抗干扰性能好;其次通过对运动模糊图像做两次方向微分,估计其模糊方向,提高了模糊方向的估计准确度;最后利用改进的Prewiit算子和费米函数计算模糊图像的刃边函数,进而得到图像的调制传递函数,再利用维纳滤波复原图像.实验结果表明:本文算法不仅具有有效性和强抗噪音能力,而且对图像的信噪比要求可以低到20 dB;与传统算法相比,提高了图像的复原质量.     

X射线探测器图像的恢复研究

图像退化是引起X射线探测器图像质量下降的主要原因。图像恢复,可以提高系统的分辨率和对比度灵敏度。通过对射线成像系统图像退化的分析,介绍了反滤波信号恢复方法,并提出了一种近似恢复算法,该算法将反滤波恢复近似为除法运算的恢复。给出了恢复前后的图像及标准差的对比,实验结果表明了该算法的有效性。     

基于指导滤波的图像盲复原算法

作为图像处理领域的重要分支和研究热点之一, 图像复原方法 的研究始终具有重要理论意义和广泛的应用价值, 图像盲复原一直以来都 是图像复原中比较困难的问题之一. 针对相机与所拍摄景物之间由于相对 位置移动而使所获得图像发生运动模糊的情况, 本文提出了一种基于指导滤波 的图像盲复原算法. 我们首先通过频域迭代算法对点扩散函数 进行估计. 然后, 由于指导滤波具有较好的保持图像边缘的特性, 我们应用基于指导滤波的图像非盲复原算法恢复目标图像. 对以上两步进行反复迭代, 直到获得最终的清晰图像. 为了验证本文所提算法的有效性, 给出了多组对比实验. 实验结果表明, 本文所提算法能够在有效地抑制噪声和振铃 效应的同时, 还能够更好的保持图像的边缘和纹理细节. 因此, 本文算法可以获得更高质量的复原图像.     

基于迁移学习的原子力显微镜成像恢复方法

受限于探针针尖结构尺寸,用原子力显微镜进行微纳测量时会产生图像边缘失真.提出了一种基于迁移学习的原子力显微镜成像恢复方法,通过迁移学习训练源模型和靶模型实现一维栅格成像恢复.该方法采用数学形态法中的腐蚀算法生成栅格点云数据,通过U-Net网络源模型从点云中提取针尖卷积效应的特征向量,将权重参数迁移至U-Net网络靶模型,靶模型在自适应正则化方法下进行监督学习.实验结果表明,该方法能有效恢复一维栅格的原子力显微镜测量图像,提高横向分辨力,可用于纳米栅格的线宽检测上.     

一种基于改进频域图像配准技术的超分辨力图像处理方法

为提高光电成像系统的空间分辨力,提出了一种基于改进的频率域图像配准技术的超分辨力图像处理方法。首先利用改进的频域图像配准方法估算出低分辨力图像之间的微位移量,然后采用Papoulis-Gerchberg超分辨力处理方法完成图像复原。利用不同重构方法进行了仿真及实验研究,给出了评价参数。模拟和实际显微热图像的处理结果表明:该算法可使图像质量得到改善,分辨的细节更多,可有效地提高光电成像系统的空间分辨力;处理算法简单,计算量小,可实现快速处理。该算法还可应用于其他不可控光学微扫描成像系统中,具有广泛的应用前景。     

An adaptive regression method for infrared blind-pixel compensation

Blind pixel compensation is an ill-posed inverse problem of infrared imaging systems and image restoration. The performance of a blind pixel compensation algorithm depends on the accuracy of estimation for the underlying true infrared images. We propose an adaptive regression method (ARM) for blind pixel compensation that integrates the multi-scale framework with a regression model. A blind-pixel is restored by exploiting the intra-scale properties through the nonparametric regressive estimation and the inter-scale characteristics via parametric regression for continuous learning. Combining the respective strengths of a parametric model and a nonparametric model, ARM establishes a set of multi-scale blind-pixel compensation method to correct the non-uniformity based on key frame extraction. Therefore, it is essentially different from the traditional frameworks for blind pixel compensation which are based on filtering and interpolation. Experimental results on some challenging cases of blind compensation show that the proposed algorithm outperforms existing methods by a significant margin in both isolated blind restoration and clustered blind restoration.