基于DDE技术分析的红外图像细节增强算法

在光电设备中,高位宽红外探测器由于具有高灵敏度、低噪声的特点而得到广泛应用。而通常的显示设备或快速处理只要求8位数据宽度,所以需要将高位宽的图像数据压缩至8位位宽。但是如果压缩方法处理不当,可能会造成原有信息丢失,在显示图像中表现为图像细节的损失。通过对FLIR公司数字图像细节增强（DDE）技术的分析,比较了DDE技术与以往常用红外图像动态范围压缩算法的区别,总结并提出了符合DDE思想的红外图像动态范围压缩细节增强算法,该算法在全面增强图像细节的同时能够抑制图像背景中的杂波,增强弱小目标,既满足图像显示的视觉效果,又有利于在图像上进行目标提取。     

Time-dependent crosstalk effects for image sensors with different isolation structures

Photo-generated carriers may diffuse into the adjacent cells to form the electrical crosstalk, which is especially noticeable after the pixel cell size has been scaled down. The electrical crosstalk strongly depends on the structure and electrical properties of the photosensitive areas. In this work, time-dependent crosstalk effects considering different isolation structures are investigated. According to the different depths of photo-diode(PD) and isolation structure, the transport of photo-generated carriers is analyzed with different regions in the pixel cell. The evaluation of crosstalk is influenced by exposure time. Crosstalk can be suppressed by reducing the exposure time. However, the sensitivity and dynamic range of the image sensor need to be considered as well.     

基于电子轰击式CCD的大动态条纹相机研究

为了实现对更弱、以及物理量跨度更大的信号探测, 满足材料、生物、信息、半导体物理以及能源等重大科学领域对诊断精密化的进一步需求, 需要提高条纹相机的动态范围、空间分辨率和信噪比. 为此, 本文研制了基于电子轰击式CCD(EBCCD)的大动态条纹相机, 条纹变像管采用时间和空间方向分别聚焦的矩形框电极和电四极透镜结构, 可降低空间电荷效应. 并提高电子加速电压, 减小电子渡越时间以降低空间电荷相互作用时间. 采用基于电子轰击读出技术的背照式CCD(BCCD)作为读出器件, 取代传统的像增强CCD(ICCD)以缩短图像转换链, 较大地降低了超快诊断设备转换过程中的图像衰减, 从而提高条纹相机图像的信噪比、空间分辨率和动态范围. 实验得到静态空间分辨率高于35 lp/mm, 动态空间分辨率达到20 lp/mm, 偏转灵敏度为60.76 mm/kV, 动态范围达到2094:1, 扫描速度非线性为5.04%, 条纹相机的电子轰击半导体(EBS)增益达到3000以上.     

高动态图像获取技术

高动态图像获取技术可以提高复杂光照条件下的成像质量。简要介绍了高动态图像获取的基本原理。对近年来发展起来的各种高动态成像方法和重建技术进行了总结,尤其是采用普通图像传感器通过扩展重建获取高动态图像的方法。     

基于对数解调的递变能量CT成像方法

复杂结构件由于有效厚度差异大和成像系统动态范围受限, 单一能量下的投影数据信息不完整, 常用CT重建算法及不完全数据重建算法无法在数据缺失严重的情况, 有效实现复杂结构件的CT重建. 为此论文提出基于对数解调的递变能量CT成像方法. 该方法在分析直接高动态CT成像所存在问题的基础上, 提出利用对数变换压缩递变能量投影序列动态范围, 并利用现有的基于图像灰度一致性的融合方法, 计算融合加权系数, 再经常规重建算法实现复杂构件的CT成像. 论文并以某复杂仪表为对象, 进行实验, 相比传统的固定能量成像方式, CT信息完整, 质量高. 从而说明论文所提出的方法, 能够实现CT系统动态范围的扩展, 实现复杂结构件的高动态CT成像.     

Elimination of environmental noise in interferometric wavelength shift demodulation for dynamic fiber Bragg grating sensor array

We present a novel reference compensation method for eliminating environmental noise in interferometric wavelength shift demodulation for dynamic fiber Bragg grating (FBG) sensors. By employing a shielded wavelength-division-multiplexed reference FBG in the system, the environmental noise is measured from the reference channel, and then subtracted from the demodulation result of each sensor channel. An approximate 40 dB reduction of the environmental noise has been experimentally achieved over a frequency range from 20 Hz to 2 kHz. This method is also suitable for the elimination of broadband environmental noise. The corresponding FBG sensor array system proposed in this paper has shown a wavelength resolution of .     

Development of High Performance Quantum Image Algorithm on Constrained Least Squares Filtering Computation

Recent development of computer technology may lead to the quantum image algorithms becoming a hotspot. Quantum information and computation give some advantages to our quantum image algorithms, which deal with the limited problems that cannot be solved by the original classical image algorithm. Image processing cry out for applications of quantum image. Most works on quantum images are theoretical or sometimes even unpolished, although real-world experiments in quantum computer have begun and are multiplying. However, just as the development of computer technology helped to drive the Technology Revolution, a new quantum image algorithm on constrained least squares filtering computation was proposed from quantum mechanics, quantum information, and extremely powerful computer. A quantum image representation model is introduced to construct an image model, which is then used for image processing. Prior knowledge is employed in order to reconstruct or estimate the point spread function, and a non-degenerate estimate is obtained based on the opposite processing. The fuzzy function against noises is solved using the optimal measure of smoothness. On the constraint condition, determine the minimum criterion function and estimate the original image function. For some motion blurs and some kinds of noise pollutions, such as Gaussian noises, the proposed algorithm is able to yield better recovery results. Additionally, it should be noted that, when there is a noise attack with very low noise intensity, the model based on the constrained least squares filtering can still deliver good recovery results, with strong robustness. Subsequently, discuss the simulation analysis of the complexity of implementing quantum circuits and image filtering, and demonstrate that the algorithm has a good effect on fuzzy recovery, when the noise density is small.     

In situ quantification of noise as a function of signal in digital images

An efficient and accurate algorithm for determining the magnitude of noise as a function of signal in an arbitrary digital image is presented and demonstrated. The algorithm is robust and largely independent of the form of the image, returning the noise function with subcount error across the full dynamic range of a synthetic test image where noise of a known form has been added. The noise performance of a CCD under different image recording and processing conditions is examined using the algorithm. The effect of different noise functions on pattern-matching measurements of electronic structure by quantitative convergent beam electron diffraction is investigated.     

微通道板像增强器动态范围测试技术研究

研究了微通道板像增强器动态范围特性，分别分析了以亮度增益、信噪比和分辨力等参数作为判断微通道板像增强器动态范围依据的可行性，形成了以分辨力和输出亮度为评判依据的微通道板像增强器动态范围测试方法，搭建了相应的动态范围测量装置，并对测量装置的组成部分进行了限制设计。对微通道板像增强器动态范围测试装置的分析显示：建立的测量装置可以实现10-5～103lx的较宽动态范围测量，能够满足微通道板像增强器动态范围测试的要求。     

用于MR图像非局域均值去噪的边缘增强策略

非局域均值(NLM)滤波有很好的去噪效果并已成功地应用于磁共振图像的去噪中,但与所有去噪方法相同,总是会在一定程度上模糊图像细节. 该文提出将从原始图像中提取出来的高频信息与NLM去噪图像相融合,来还原在去噪过程中丢失的细节. 首先利用一种基于拉普拉斯金字塔的多分辨率方法,从原始图像中提取出包含丰富的边缘信息的高频组分. 然后利用作者提出的一种新的基于SUSAN算子的边缘检测算子产生一幅连续的边缘图,并利用该边缘图将高频组分与NLM方法去噪的图像相融合. 该方法在图像的平滑区域取得了良好的去噪效果,同时可以保留甚至增强图像的细节. 同时,该方法对图像的增强不会导致增强图像中常见的伪影.     

相干激光雷达距离像的噪声抑制算法研究

激光雷达能同时成距离像和强度像,较其他的成像体制给出更多的有用信息。距离像的去噪是激光雷达图像处理的关键步骤,去噪的好坏直接关系激光雷达在跟踪、定位等方面的精确性。根据距离像噪声来源,采用结合强度像的多级中值滤波对实际的雷达图像进行了去噪处理。首先利用强度像3×3邻域均值和图像平均值消除失落信息影响,然后采用多级中值滤波消除逸出值。分别给出了各步处理结果,最终获得了较清晰的距离像。研究结果表明运用结合强度像的多级中值滤波方法能够有效地抑制相干激光雷达距离像的噪声。     

Time history speckle pattern under statistical view

The dynamic speckle analysis has been done because speckle interference began to be evaluated as an important source of information, especially those related to biological samples under laser beam. The time history speckle pattern, a THSP image, is an approach to analyze, which allows a summary to represent the activity monitored. THSP adoption has compelled the research to evaluate the information and the reliability of its inner pattern. This study presents a statistical approach to analyze the data using white noise tests and cross-spectrum analysis. The data analyzed was a set of THSP from animal sperm samples. The evaluation of white noise among lines was conducted using the Fischer test over eight THSP and the cross-spectral approach has been conducted by comparing the information in each THSP. The results presented that the THSP of the sperm did not behave as a white noise in a global evaluation, and specifically in the Gaussian white noise test over 4096 lines; only six lines presented the behavior of a white noise pattern. The strong rejection of white noise test confirms that the THSP pattern allows reliable information. The results obtained with the cross-spectral analysis presented differences between lines within the same THSP, showing that the information of an inner pattern varies in relation to space, which is against the findings in literature.     

Image recovery by backward iteration of the non-standard finite difference time domain algorithm

The time-reversed finite difference time domain (FDTD) algorithm is a useful tool for image recovery. By using a negative time step, a scattered wave field can propagated backwards to extract information about the size and shape of a scatterer, but errors, in both the FDTD algorithm and noise in scattered field data, reduce the quality of the recovered image. Using a new high accuracy FDTD algorithm derived from a nonstandard (NS) finite difference model, we can greatly reduce FDTD error. Using NS-FDTD we have investigated image recovery as a function of noise in numerical experiments.     

基于CCD传感的激光粒度测试技术研究

颗粒测试在工业生产和科学研究中涉及的领域非常广泛,常用的颗粒粒度及其分布的测试方法是激光粒度测试法,其具有测试精度高、测量速度快、重复性好和可测粒径范围宽等突出优点。CCD传感器有灵敏度高、分辨率高、噪声小和较大的动态范围等优点,其作为激光粒度仪的探测器提高光强分辨率的应用已经很普遍了。为提高测量精度,通过对CCD传感技术的研究,应用图像处理的方法来设计光电探测器,搭建了基于米氏散射原理的激光粒度测试系统。实验结果表明,用CCD传感器采集光散射图像,再对图像进行处理,D50与D10误差在6%之内,D90误差在1%之内,降低了测量的重复误差。     

基于场景坏点检测的面阵CCD暗电流估计方法

利用CCD中的部分坏点作为温度敏感器,提取CCD温度估计值,建立温度估计值与暗电流的函数模型,并结合均值滤波的方法,在真实场景图像中提取CCD温度估计值,实现了面阵CCD的暗电流估计。在方法分析的基础上,以真实的暗电流数据为基准参考,对暗电流的估计结果进行了比对实验。实验结果表明,在不同的积分时间及大动态范围的成像条件下暗电流的估计结果十分准确,偏差小于0.4%,并且具有一定的抗噪性。该方法利用场景本身的特征信息对拍摄时刻的暗电流进行估计,不需要额外采集暗电流数据,节省了图像采集时间,十分适合于积分时间较长的高光谱成像或天文观测领域,采用无温度控制的低成本CCD成像探测系统进行实时图像采集。     

推扫式多挡增益光谱成像模拟及噪声分析

综合性遥感监测对载荷的灵敏度和动态范围都有较高的要求,为此提出一种基于逐像元多挡增益切换的光谱成像方法。不同于帧增益或者行列增益切换的成像方式,该方法利用4T-APS CMOS探测器无破坏读出的特点,可以进行像元级的增益优化。这种方式可以兼顾水体、植被、云层等多要素的成像需求,极大地提高了载荷的研制效益。其基本原理为探测器先进行全局曝光,然后根据多级积分电容的饱和判断,选取不饱和的最高增益信号以增益码加信号的形式下传。地面根据增益码对应的定标系数反推出像元真实辐射值。由于谱段多且为分段响应,为保证系统的定量化应用,建立多挡增益光谱成像模型并进行噪声分析具有重要意义。通过对噪声类型的分析,建立了多挡增益下的泊松-高斯噪声模型。基于该模型计算了像元受噪声影响以低增益读出的概率。结果表明,虽然噪声会影响读出增益的变化,但影响区间极小,入瞳辐亮度在5 mW·cm-2·μm-1·sr-1以内,信号比增益挡位辐亮度分界值小0.05 mW·cm-2·μm-1·sr-1即可保证正常读出的概率大于99.6%。随着信号增强,光子噪声增大,增益减小,影响区间扩大。根据多挡增益信噪比模型,分析得出光谱模式与合并通道模式下的信噪比变化。最后,利用宽波段成像光谱仪（WIS）数据作为入瞳辐亮度进行了四挡增益的推扫式光谱成像模拟,分析了多挡增益光谱图像的固有特点。根据噪声模型对中心波长为0.443 μm的光谱图像添加1~3σ的随机噪声,分析了噪声对地物目标所处增益的影响。结果显示,在满足信噪比指标的前提下,该系统的单挡动态范围为74 dB,总动态范围可达114 dB。该方法不但提高了水体等弱信号的信噪比,而且可以保证建筑、云层等亮目标不饱和。成像模拟及噪声分析不仅有利于该载荷的后续研制,也可以为同类光谱仪器的设计提供参考。     

TDI CCD光子响应非均匀性噪音分析与测量

探测器光子响应非均匀性噪音会降低低照度情况下遥感成像系统的成像质量.针对这一现象,本文首先结合探测器的物理性质,对各种噪音源进行了研究;建立了TDI CCD不同级数下的光子响应非均匀性噪音模型,随着曝光量的增加,光子响应非均匀性噪音也线性增加.其次根据曝光级数越多TDI CCD对非均匀性噪音的平滑效应越明显这一现象,提出一种光子响应非均匀性系数与曝光级数之间的关系式,并给出了利用TDI CCD输出图像提取光子响应非均匀性噪音的方法.最后建立了试验系统,通过试验对测试获得的光子响应非均匀性噪音与理论分析计算得出的结果进行了分析.     

Hybrid multi-resolution detection of moving targets in infrared imagery

A hybrid moving target detection approach in multi-resolution framework for thermal infrared imagery is presented. Background subtraction and optical flow methods are widely used to detect moving targets. However, each method has some pros and cons which limits the performance. Conventional background subtraction is affected by dynamic noise and partial extraction of targets. Fast independent component analysis based background subtraction is efficient for target detection in infrared image sequences; however the noise increases for small targets. Well known motion detection method is optical flow. Still the method produces partial detection for low textured images and also computationally expensive due to gradient calculation for each pixel location. The synergistic approach of conventional background subtraction, fast independent component analysis and optical flow methods at different resolutions provide promising detection of targets with reduced time complexity. The dynamic background noise is compensated by the background update. The methodology is validated with benchmark infrared image datasets as well as experimentally generated infrared image sequences of moving targets in the field under various conditions of varying illumination, ambience temperature and the distance of the target from the sensor location. The significant value of F-measure validates the efficiency of the proposed methodology with high confidence of detection and low false alarms.     

On noise reduction in strain maps obtained with the grid method by averaging images affected by vibrations

Any image-based contactless measurement system has a limited resolution because of sensor noise. If the sensor is rigorously static with respect to the imaged object, a possibility is to reduce noise by averaging images acquired at different times. This paper discusses images of a pseudo-periodic grid used in experimental solid mechanics to give estimations of in-plane displacement and strain components of a deformed flat specimen. Because of the magnification factor which is employed, the grid images are often affected by residual vibrations, thereby invalidating the assumption that the sensor is static. The averaged grid image is thus a biased estimator of the unknown noise-free image. In spite of this, we prove that the retrieved displacement and strain components still benefit from noise reduction by time-averaging. A theoretical model is discussed, and experiments on real and synthetic data sets are provided.     

No-reference image noise estimation based on noise level accumulation

In this paper, a method of no-reference image noise assessment is presented, which utilizes the estimated noise level accumulation (NLA) index value. The affine reconstruction model is applied after segmenting the noisy image into several patches. Boundary blur process is conducted to smooth the segmentation edges. For each image patch the mean value standing for brightness and the standard deviation value indicating the noise standard deviation are computed to give the noise samples estimation. The accurate image noise standard deviation is estimated by integrating NLA index value of several overlapped intervals combined with different visual weights. Experiment results are provided to demonstrate that the proposed method performs well for images with different contents over a large range of noise levels both monotonously and accurately. Comparisons against other conventional approaches are also carried out to exhibit the superior performance of the proposed algorithm.