具有颜色恒常性的光谱反射率重建

给定物体三刺激值,重建物体反射率在跨媒体颜色复制领域有着重要应用.常见的重建反射率算法包括基向量法、维纳估计法、加权伪逆方法等,这些方法大多都是为了建立由低维度三刺激值或者RGB向高维度光谱反射率的映射关系,以重建的反射率与原始反射率的接近程度为评价指标,并且需要光谱反射率数据进行训练.但是很多工业领域在产品设计时都需...     

Deep Learning and Entropy-Based Texture Features for Color Image Classification

In the domain of computer vision, entropy—defined as a measure of irregularity—has been proposed as an effective method for analyzing the texture of images. Several studies have shown that, with specific parameter tuning, entropy-based approaches achieve high accuracy in terms of classification results for texture images, when associated with machine learning classifiers. However, few entropy measures have been extended to studying color images. Moreover, the literature is missing comparative analyses of entropy-based and modern deep learning-based classification methods for RGB color images. In order to address this matter, we first propose a new entropy-based measure for RGB images based on a multivariate approach. This multivariate approach is a bi-dimensional extension of the methods that have been successfully applied to multivariate signals (unidimensional data). Then, we compare the classification results of this new approach with those obtained from several deep learning methods. The entropy-based method for RGB image classification that we propose leads to promising results. In future studies, the measure could be extended to study other color spaces as well.     

基于Land实验的可见红外伪彩色图像融合方法

由Land在研究人类视觉颜色恒定性现象时从所做的某个实验出发,通过对该实验中所用红、绿色滤波片的理想化假设以及光源的等能分布假设,并建立红外图像与红色透明片、可见光图像与绿色透明片的对应关系,得到Land实验启发下的新的伪彩色融合算法。通过与Toet伪彩色融合算法的结果比较,说明了该方法的有效性。进而借鉴Toet算法中强调图像特有成分的思想,使本算法输出更丰富、更逼近真实景物的色彩。     

基于主动光照的深海图像增强算法

为了改善主动光在水下传播过程中由散射与吸收效应导致的深海图像对比度低下以及颜色失真问题,提出一种水下图像增强算法.不同于传统方法利用最亮点的强度值作为背景光,提出基于物体与背景光非相关性的背景光估计方法,有效避免了前景处的亮像素或白色物体像素对背景光的误判,同时确保了去散射的精确性,提高水下图像的对比度;针对人造光源的颜色增益和光路衰减导致的图像色偏等问题,在去散射图像上选取离光源最近的灰色像素,利用其对光源的敏感性,将光照强度分离出来.最终,通过估计并去除光源本身的颜色增益,同时补偿光在传播过程中的损失,实现图像的颜色校正.实验结果表明,所提算法可以有效去除水下图像的散射效应,较好地恢复图像色彩,进而获得较优的增强图像.相比于其他算法,增强后的图像信息熵和水下图像质量评价指标值较高,说明该算法能显著提升水下图像的质量,同时保留图像有用信息.     

近红外与可见光双通道传感器信息融合的去雾技术

为了对雾霾天气下的图像进行去雾处理,多幅图像去雾算法是常用的方法之一。多幅图像去雾算法也有多种形式,部分算法面临硬件实现困难、获取途径受限或者可实施性弱等问题,而且多幅图像比对处理时常常涉及图像配准,造成算法的实时性差、计算复杂度高等问题。针对以上问题,提出的算法为多幅图像去雾提供了新的思路,基于双目传感器硬件架构能够同时捕获近红外和可见光图像,将近红外传感器图像作为新的数据源,近红外传感器能够在一定程度上穿透雾霾,在雾天捕获可见光传感器无法捕获的图像细节,而且硬件实现简单。可见光图像的颜色信息较丰富,近红外传感器图像对近处场景细节的描述能力较好,捕获的图像稍加校正就能实现完全配准,将近红外图像与可见光图像进行融合,在去雾的同时,可以将近红外传感器图像中的原始细节提取融合到彩色可见光传感器图像中,得到边缘、轮廓等细节信息更加丰富的去雾图像。基于上述思路,借助近红外传感器对边缘细节的描述能力和可见光传感器对颜色信息的反映能力,提出了一种基于近红外与可见光双通道传感器图像融合的去雾算法。首先,将彩色可见光图像转换到HIS彩色空间,分别得到亮度通道图像、色调通道图像和饱和度通道图像。先将其亮度通道图与近红外图像进行融合去雾处理。采用非下采样Shearlet变换（NSST）进行分解,对得到的高频系数进行双指数边缘平滑滤波器保边滤波处理,对低频系数进行反锐化掩蔽处理,通过融合规则和反向变换得到新的亮度通道图像。然后,在对可见光图像的色彩处理中,建立饱和度图的退化模型,采用暗原色原理对参数进行估计,得到估计的饱和度图。最后,将新的亮度通道图像,估计的饱和度图像和原色调图像反映射到RGB空间得到去雾图像。为了验证新算法的有效性,特选取四组雾天拍摄的真实近红外图像与可见光图像进行融合去雾处理,将融合结果与其他两种去雾方法对于彩色可见光图像的去雾效果进行比较。实验结果表明,该算法在提高图像的边缘对比度和视觉清晰度上有较好的效果。并提出将近红外传感器图像作为新的数据源,采用双通道图像融合方法进行去雾处理,为图像去雾提供的新的技术思路是可行的。该算法的优势在于：首先提出将图像融合方法与去雾算法相结合,得到了新的去雾算法的思路。将彩色可见光图像转换到HSI色彩空间,将其亮度通道图与近红外图像采用非下采样Shearlet变换方法进行融合处理,在去雾的同时,可以将近红外传感器图像中的原始细节提取融合到彩色可见光传感器图像中,使得去雾图像中的边缘、轮廓等细节信息更加丰富。其次,提出了在图像去雾算法中采用新的数据源--近红外传感器图像,从图像处理的角度,近红外传感器能够在一定程度上穿透雾霾,对于近处场景细节的描述能力较好,而且硬件实现简单,捕获的图像稍加校正就能实现完全配准,为后续的融合去雾算法带来了便利,为图像去雾提供了新的技术途径和路线。再次,采用的是多幅图像去雾算法,该算法基于双目传感器获取图像,可见光图像的颜色信息较丰富,近红外图像对于近处场景细节的描述能力较好,相对于单幅图像去雾算法,有更好的效果。最后,将可见光传感器图像映射到其他色彩空间,对于每个通道的图像根据其特征有针对性地进行处理。可见光图像的亮度通道图和近红外图像的处理采用了图像融合和增强处理,对于可见光图像饱和度通道的处理采用了图像复原算法,可以从整体上提升去雾效果,对细节特征有了进一步增强。该算法为图像去雾提供了新的技术途径和路线。     

A refined fuzzy time series model for stock market forecasting

Time series models have been used to make predictions of stock prices, academic enrollments, weather, road accident casualties, etc. In this paper we present a simple time-variant fuzzy time series forecasting method. The proposed method uses heuristic approach to define frequency-density-based partitions of the universe of discourse. We have proposed a fuzzy metric to use the frequency-density-based partitioning. The proposed fuzzy metric also uses a trend predictor to calculate the forecast. The new method is applied for forecasting TAIEX and enrollments’ forecasting of the University of Alabama. It is shown that the proposed method work with higher accuracy as compared to other fuzzy time series methods developed for forecasting TAIEX and enrollments of the University of Alabama.     

浅海被动水下偏振成像探测方法

针对传统被动水下偏振成像方法忽略水体对光的吸收效应,成像结果中存在严重的色彩失真,且并未深入发掘利用背景散射光中包含的场景信息的问题.提出浅海被动水下偏振成像探测方法,该方法从水体中背景散射光的传输特性出发,分析场景深度信息与散射光的物理关系,建立基于深度信息的水下Lambertian反射模型,实现无色彩畸变的水下目标场景清晰成像探测.实验结果表明,该方法能够提供接近水下目标真实色彩、符合人眼视觉特性的清晰探测结果,提高水下成像探测能力.     

基于遗传算法选择多光源下的光谱反射率重构研究

为解决基于RGB三通道信息值重构光谱反射率精度不理想的问题,提出了一种优化的基于RGB三通道信息的光谱反射率重构算法。首先编码产生随机选择多个光源的个体,RGB三通道值通过多项式回归算法预测多个光源下的三刺激值,并采用伪逆法进行多光源下的光谱反射率重构,然后将样本的重构精度作为个体的适应度评估值,以优胜劣汰,适者生存为原则对个体进行选择、交叉、变异操作,最后得到适用于颜色样本光谱重构的多个光源与基于这些光源重构得到的光谱反射率。实验选用Munsell颜色集作为训练样本集,RC24色卡、SG140色卡作为检测样本集,8个标准光源和82个发光二极管光源作为实验光源,采用该算法从90个光源中选取最优的光源组合并重构得到样本的光谱数据,并与Zhang提出的基于穷举法选择的多光源下的光谱重构方法和A光源下的伪逆法进行了重构精度对比。实验结果显示该研究提出的方法随着光源个数的增加,光谱反射率重构精度提高,特别是光源个数增加到3时,光谱重构精度提高的幅度最大。在三种重构方法中,该方法重构RC24的平均色差和平均光谱均方根误差分别为0.332 4和0.002 9,而Zhang的方法与伪逆法的平均色差分别为0.429 3和3.266,平均光谱均方根误差分别为0.029 7和0.004 8;该文方法重构SG140的平均色差和平均光谱均方根误差分别为0.486 2和0.007 3,而Zhang的方法与伪逆法的平均色差分别为0.544 8和3.821 9,平均光谱均方根误差分别为0.035 6和0.013 3。结果表明基于多光源下的光谱反射率重构精度明显优于基于单个光源下的重构精度,而基于遗传算法的多光源选择方法又优于穷举法,它能够根据颜色样本自动寻找到最优光源组合,从而基于最优多光源下的三刺激值重构样本的光谱反射率,提高了光谱反射率重构的精度。     

Spectral reflectance estimation for electronic endoscope

A method of spectral reflectance estimation for electronic endoscopes is described. A new approach is proposed for the overfitting problem. We use a small training set. Initially, cross-validation is used to select model parameters. Then, for additional validation of the method a color image acquired under the same illumination as the training data is used. This additional color image gives a clear solution for both the estimation of spectra and the estimation of color reproduced from these spectra. Experiments confirm that the method predicts spectral reflectance and colors well. Accurate reproduction of endoscope images in different viewing conditions and under another light source was implemented using this approach.     

Accurate segmentation of leukocyte in blood cell images using Atanassov's intuitionistic fuzzy and interval Type II fuzzy set theory

In this paper automatic leukocyte segmentation in pathological blood cell images is proposed using intuitionistic fuzzy and interval Type II fuzzy set theory. This is done to count different types of leukocytes for disease detection. Also, the segmentation should be accurate so that the shape of the leukocytes is preserved. So, intuitionistic fuzzy set and interval Type II fuzzy set that consider either more number of uncertainties or a different type of uncertainty as compared to fuzzy set theory are used in this work. As the images are considered fuzzy due to imprecise gray levels, advanced fuzzy set theories may be expected to give better result. A modified Cauchy distribution is used to find the membership function. In intuitionistic fuzzy method, non-membership values are obtained using Yager's intuitionistic fuzzy generator. Optimal threshold is obtained by minimizing intuitionistic fuzzy divergence. In interval type II fuzzy set, a new membership function is generated that takes into account the two levels in Type II fuzzy set using probabilistic T co norm. Optimal threshold is selected by minimizing a proposed Type II fuzzy divergence. Though fuzzy techniques were applied earlier but these methods failed to threshold multiple leukocytes in images. Experimental results show that both interval Type II fuzzy and intuitionistic fuzzy methods perform better than the existing non-fuzzy/fuzzy methods but interval Type II fuzzy thresholding method performs little bit better than intuitionistic fuzzy method. Segmented leukocytes in the proposed interval Type II fuzzy method are observed to be distinct and clear.     

Optimized basis function for spectral reflectance recovery from tristimulus values

Spectral reflectance is a significant physical property of materials. It plays an important role in color constancy, illumination modeling, and color reproduction. Spectral reflectance basis functions are the most important impact factors for spectral reflectance recovery. Previous methods mainly calculated basis functions for the reflectance spectra data sets by employing the principle component analysis (PCA) and its improved methods. In this paper, we present a new method to solve this problem. Specifically, we propose a new cost function and some constraint conditions to convert the problem into an optimization problem by minimizing the cost function. Unlike the PCA method which yields the orthogonal basis functions for approximating the reflectance spectra, our method yields the nonorthogonal basis functions for better approximating the reflectance spectra. Experimental results show that our derived basis functions are better than those obtained by the PCA method for reflectance spectra recovery.     

Color Appearance of a Patch Explained by RVSI for the Conditions of Various Colors of Room Illumination and of Various Luminance Levels of the Patch

Whenever we enter a space illuminated differently from a previous space whether in color or in illuminance, we can quickly adapt to the new atmosphere and can again perceive white for the originally white object; this is known as color constancy. This phenomenon is explained by rotation of the recognition axis of the recognized visual space of illumination (RVSI) toward the illumination color. The explanation then predicts that the color appearance of a test patch changes radically toward the opposite direction from the color of illumination when the physical property of the test patch is kept unchanged at a neutral white. This prediction was confirmed by Experiment 1, where eight different colors of illumination were employed. The test patch appeared very vivid in color and shifted toward the opposite direction from the color of the illumination. In RVSI theory the light source color mode is explained by the release of the test patch from the restriction of RVSI. The release can be achieved by increasing the luminance of the test patch and the color appearance of the patch should then return to its own color as it is no longer controlled by RVSI. In Experiment 2 these predictions were investigated by increasing the luminance of the test patch to a much higher level than that of the objects in the lit room fixed at an illuminance of about 1001x. The color appearance of the test patch indeed became the light source color and returned to the original neutral white. Emphasis was given in the course of the experiments that the subjects were observing the test patch presented in a real 3D space where the subjects also stayed inside so that they could properly construct RVSI for the space.     

用于数码显微镜的白平衡校正方法

提出了一种用于数码显微镜的非线性白平衡方法.以标准D65光源光照下孟赛尔色卡每个色块的XYZ三刺激值和所成图像中的RGB值为训练样本,用神经网络建立不同数码显微镜拍摄的RGB输出与XYZ三刺激值之间的关系;再通过Estevens矩阵把XYZ三刺激值转换成人眼感光器所接受到的刺激值,并进行非线性计算;最后把计算结果转换回RGB空间,从而得到白平衡后的图像.实验结果表明,该方法所得的白平衡校正结果更加接近标准白光照射下的图像颜色.在光源色纯度较高的情况下,采用调整通道增益的方法变化幅度较大,而采用本文提出方法结果更加稳定;特别是当光源的色纯度过大使得图像中某个颜色通道的值接近0的时候,用调整三通道增益的方法已经完全失去了校正作用,而采用本文中的非线型方法仍然可以得到好的校正结果.同时,此方法中建立的神经网络模型反映的是成像设备的RGB空间到XYZ空间的转换特性,对于不同的光照条件,此特性不变,因此对于不同光源不需要分别建立神经网络模型.     

用于数码显微镜的白平衡校正方法

提出了一种用于数码显微镜的非线性白平衡方法.以标准D65光源光照下孟赛尔色卡每个色块的XYZ三刺激值和所成图像中的RGB值为训练样本,用神经网络建立不同数码显微镜拍摄的RGB输出与XYZ三刺激值之间的关系;再通过Estevens矩阵把XYZ三刺激值转换成人眼感光器所接受到的刺激值,并进行非线性计算;最后把计算结果转换回RGB空间,从而得到白平衡后的图像.实验结果表明,该方法所得的白平衡校正结果更加接近标准白光照射下的图像颜色.在光源色纯度较高的情况下,采用调整通道增益的方法变化幅度较大,而采用本文提出方法结果更加稳定;特别是当光源的色纯度过大使得图像中某个颜色通道的值接近0的时候,用调整三通道增益的方法已经完全失去了校正作用,而采用本文中的非线型方法仍然可以得到好的校正结果.同时,此方法中建立的神经网络模型反映的是成像设备的RGB空间到XYZ空间的转换特性,对于不同的光照条件,此特性不变,因此对于不同光源不需要分别建立神经网络模型.     

基于人眼视觉特性的光谱降维模型研究

针对传统光谱降维方法其降维重构后的光谱数据仅是对原始光谱的数学逼近,会出现光谱误差较小但颜色色差较大的缺点,创新性的提出三种将人眼视觉特性与光谱降维相结合的方法。其中,VPCA法直接将光谱光视效率函数加权到原始光谱上再进行降维,LMSPCA方法用LMS视稚响应构建加权矩阵对原始光谱加权后再进行降维,在LMSPCA法中加权矩阵的构建有两种方式,其主要区别在于视稚响应偏差的求取方式不同。方式一中,L,M,S视稚响应偏差是各对应波长上的偏差取绝对值,而方式二中,其偏差是各对应波长上的偏差平方。LMSPCAs法在LMSPCA法基础上再采用PCA(主成分分析)方法对损失的光谱进行降维。实验结果表明VPCA法降维效果较差,LMSPCA法的两种加权矩阵降维效果接近,皆可显著提高降维模型的色度精度,但会降低模型的光谱精度,LMSPCAs法由于针对LMSPCA法因光谱加权引起的光谱损失再进行光谱补偿,其在光谱精度、色度精度以及变光照条件下的色差稳定性这三个方面都能较好地表征原始高维光谱反射率,满足光谱颜色复制的要求。     

New Pythagorean Entropy Measure with Application in Multi-Criteria Decision Analysis

The purpose of this paper is to propose a new Pythagorean fuzzy entropy for Pythagorean fuzzy sets, which is a continuation of the Pythagorean fuzzy entropy of intuitionistic sets. The Pythagorean fuzzy set continues the intuitionistic fuzzy set with the additional advantage that it is well equipped to overcome its imperfections. Its entropy determines the quantity of information in the Pythagorean fuzzy set. Thus, the proposed entropy provides a new flexible tool that is particularly useful in complex multi-criteria problems where uncertain data and inaccurate information are considered. The performance of the introduced method is illustrated in a real-life case study, including a multi-criteria company selection problem. In this example, we provide a numerical illustration to distinguish the entropy measure proposed from some existing entropies used for Pythagorean fuzzy sets and intuitionistic fuzzy sets. Statistical illustrations show that the proposed entropy measures are reliable for demonstrating the degree of fuzziness of both Pythagorean fuzzy set (PFS) and intuitionistic fuzzy sets (IFS). In addition, a multi-criteria decision-making method complex proportional assessment (COPRAS) was also proposed with weights calculated based on the proposed new entropy measure. Finally, to validate the reliability of the results obtained using the proposed entropy, a comparative analysis was performed with a set of carefully selected reference methods containing other generally used entropy measurement methods. The illustrated numerical example proves that the calculation results of the proposed new method are similar to those of several other up-to-date methods.     

Pedestrian intention prediction based on dynamic fuzzy automata for vehicle driving at nighttime

In this paper, we propose a novel algorithm that can predict a pedestrian’s intention using images captured by a far-infrared thermal camera mounted on a moving car at nighttime. To predict a pedestrian’s intention in consecutive sequences, we use the dynamic fuzzy automata (DFA) method, which not only provides a systemic approach for handling uncertainty but also is able to handle continuous spaces. As the spatio-temporal features, the distance between the curbs and the pedestrian and the pedestrian’s velocity and head orientation are used. In this study, we define four intention states of the pedestrian: Standing-Sidewalk (S-SW), Walking-Sidewalk (W-SW), Walking-Crossing (W-Cro), and Running-Crossing (R-Cro). In every frame, the proposed system determines the final intention of the pedestrian as ‘Stop’ if the pedestrian’s intention state is S-SW or W-SW. In contrast, the proposed system determines the final intention of a pedestrian as ‘Cross’ if the pedestrian’s intention state is W-Cro or R-Cro. A performance comparison with other related methods shows that the performance of the proposed algorithm is better than that of other related methods. The proposed algorithm was successfully applied to our dataset, which includes complex environments with many pedestrians.     

高温下数字图像相关散斑最优成像探究

为了测量材料在高温甚至超高温下的力学性能,采用数字图像相关方法,并研究其在高温下的最优成像。采取不同的散斑制作方法,同时加入不同颜色的高温漆,在不同的温度节点,外加不同光源及相应的滤波片,采集并观察图像是否具有良好的对比度。普通的单色光源在800℃以后会逐渐失效,无法获取图像,而紫外光在1 200℃时依然可以获取较好的图像,且直接利用试件本身颜色作为底色效果更佳。采用紫外光照明可以实现DIC在高温环境下的测量。同时利用黑色或者蓝色散斑直接喷涂在试件上有着最佳的对比度,要优于常规的散斑制作方法。     

Fast and robust color constancy algorithm based on grey block-differencing hypothesis

Color constancy is a fundamental research topic in color and vision. Numerous methods have been proposed in recent years. New methods are highly accurate but tend to be more complex. This paper proposes a simple low-level statistical algorithm based on a new hypothesis, the grey block-differencing hypothesis, which states that the average of reflectance differences of adjacent blocks in a scene is achromatic. The new method has almost the same complexity as the simplest methods (i.e., grey world and max-RGB). Experimental results demonstrate that the accuracy of the proposed method is exceptional.