Chest X-ray enhancement to interpret pneumonia malformation based on fuzzy soft set and Dempster–Shafer theory of evidence

Image enhancement algorithms are commonly used to increase the contrast and visual quality of low-dose x-ray images. This paper proposes an automated enhancement method using soft fuzzy sets with a new decision-making scheme based on Dempster-Shafer theory of evidence for the visual interpretation of pneumonia malformation in low-dose x-ray images, called as XEFSDS. The XEFSDS model first generates an original source x-ray image into a complementary image, then each original and complement image is applied to the characterized image object and background areas of fuzzy space. The S-function is utilized to define fuzzy soft sets for the classification of gray level ambiguity in both images, and hence a decision criterion via Dempster-Shafer approach and fuzzy interval has been adapted to discriminate uncertainties on the pixel intensity and the spatial information. Modified membership grade operations have been performed on each object/background area, and Werner’s AND/OR operator (an aggregation operator) has been utilized to build a new membership function from two modified membership functions. Finally, an enhanced image is obtained from the new membership function via defuzzification. Experiments on different pneumonia X-ray images demonstrate that the XEFSDS scheme produces better results than the existing methods. To show the advantages of the XEFSDS scheme, we have executed a segmentation based examination on enhanced image for the detection of pneumonia malformation as well as abnormal lobe (lobar pneumonia) or bronchopneumonia.     

基于最优模糊变换和Retinex理论的静脉图像增强方法

一种新的结合模糊变换和retinex理论静脉图像增强方法,可以解决近红外静脉图像所存在的低对比度,动态范围狭窄和强度分布不对称问题。最优模糊变换用于加强全局对比度,引入的Retinex方法可以增强图像细节信息,弥补最优模糊变换的细节缺失。由于图像从空间域向模糊域转换时使用一个参数优化隶属函数,处理的图像不具有最佳性,文中提出一种双参数的隶属函数的优化方法,同时提出一种自适应的选择控制参数方法。实验结果表明,该方法可以有效提高静脉图像与背景的对比度,与其他方法的实验结果相比较,可以看出该办法具有更好的图像增强性能。     

A rank ordered filter for medical image edge enhancement and detection using intuitionistic fuzzy set

This paper gives a novel scheme using intuitionistic fuzzy set theory to enhance the edges of medical images. Medical images contain lots of uncertainties, as they are poorly illuminated and fuzzy/vague in nature. So, direct segmentation techniques will not produce better results. There are lots of researches on edge enhancement starting from non-fuzzy to fuzzy set, but proper enhancement (highlighting important structures) is not obtained. Enhancement of edges helps in recovering the important structures that are not visible properly. Even minute pathological blood vessels/cells are not visible properly and in that case edge enhancement will enhance these blood vessels/cells. Intuitionistic fuzzy set theory is found suitable in medical image processing as it considers more (two) uncertainties as compared to fuzzy set theory. In the processing phase, image is initially converted to intuitionistic fuzzy image and intuitionistic fuzzy entropy is used to obtain the optimum value of the parameter in the membership and non-membership functions. Then it computes the total variation of the pixels with respect to the median value of the image window (rank order filtering). This enhances the borders or the edges of the image. The resulting image is then segmented (edge detected) using standard Canny's edge detector, when simply using Canny's edge detector does not give better result. From the result it is observed that on comparing with non-fuzzy and fuzzy methods, the proposed method gives better information about the images, which is helpful to the pathologists in accurate diagnosing of diseases.     

Development of intuitionistic fuzzy special embedded convolutional neural network for mammography enhancement

This article proposes a novel mammogram enhancement approach using adaptive intuitionistic fuzzy special set (IFSS) with deep convolutional neural network (called MECNNIFS) for visual interpretation of mammography lesions, lumps, and abnormal cells in low‐dose X‐ray images. The proposed MECNNIFS scheme utilizes the membership grade modification by IFSS on low‐dose X‐ray images (mammography). The suggested model attempts to increase the underexposed and abnormal structural regions such as breast lesions, lumps, and nodules on the mammogram. The proposed algorithm initially separates mammograms using convolutional neural networks (CNNs) into foreground and background areas and then fuzzifies the image by intuitionistic fuzzy set theory. Low‐level features of a mammogram of the adjacent part are integrated with CNN in pixel classification during the separation task stage to improve the performance. Hyperbolic regularization and hesitant score have been applied on fuzzy plane to quantify the uncertainty and fuzziness in spatial domain for the proposed contrast enhancement. Finally, an enhanced mammogram is acquired through the process of defuzzification. The results show better quality and performance for improvement of contrast and visual quality in mammograms compared with other state‐of‐the‐art methods.     

Image segmentation based on histogram analysis utilizing the cloud model

Both the cloud model and type-2 fuzzy sets deal with the uncertainty of membership which traditional type-1 fuzzy sets do not consider. Type-2 fuzzy sets consider the fuzziness of the membership degrees. The cloud model considers fuzziness, randomness, and the association between them. Based on the cloud model, the paper proposes an image segmentation approach which considers the fuzziness and randomness in histogram analysis. For the proposed method, first, the image histogram is generated. Second, the histogram is transformed into discrete concepts expressed by cloud models. Finally, the image is segmented into corresponding regions based on these cloud models. Segmentation experiments by images with bimodal and multimodal histograms are used to compare the proposed method with some related segmentation methods, including Otsu threshold, type-2 fuzzy threshold, fuzzy C-means clustering, and Gaussian mixture models. The comparison experiments validate the proposed method.     

基于直觉模糊散度的噪声图像边缘检测

针对传统图像边缘检测算法抑制噪声能力差的问题,提出一种基于直觉模糊集（Intuitionistic Fuzzy Set,IFS）的边缘检测算法。该算法设定了一个表示平坦区域的模板图像,并在图像窗口内构造了一种同时考虑了图像梯度和图像窗口的方差信息的隶属度函数,然后通过计算图像窗口与模板图像之间的模糊直觉散度（Intuitionistic Fuzzy Divergence,IFD）对边缘进行定位和输出。实验结果表明,对于被高斯噪声或均匀噪声严重污染的图像,该算法能够得到较好的检测结果。     

一种基于粒子群算法的模糊隶属函数优化方法

在分析图像模糊增强算法对于隶属函数及其模糊区域选择方法不足的基础上,提出一种新的基于粒子群算法的模糊隶属函数优化方法。该方法给出一个新模糊熵的定义,这个新模糊熵定义不仅考虑到图像在模糊域中划分区域时随隶属函数变化而变化的情况,同时又考虑到图像在空域中划分区域时随隶属函数变化而变化的情况。这样就使得图像依照最大熵准则变换到模糊域更能够有效地反映图像的固有信息。另外,根据图像增强算法中使用double型数据类型的特点,采用改进粒子群优化算法寻求隶属函数的最优参数。将新算法应用于图像增强中,取得了优于现有大多数模糊增强算法的效果。     

自适应图像模糊增强快速算法

Pal-King算法的隶属度函数复杂,图像增强速度慢,且渡越点难以设置。针对上述问题,提出一种自适应图像模糊增强快速算法。采用新的隶属度函数使模糊增强函数的增强幅度更大、速度更快,通过改进OTSU算法的自适应阈值计算公式,使渡越点的设置更合理。实验结果表明,与Pal-King算法相比,改进算法具有更快的增强速度和更好的增强效果。     

区间二型模糊集彩色遥感图像边缘检测方法

针对一型模糊集其隶属度函数是确定的,不具有柔性,很难满足图像的多方面边缘检测要求,及传统PalKing算法采用单一阈值对图像进行增强难以满足灰度变化丰富且含大量信息的彩色遥感图像处理的要求。提出了一种新的基于区间二型模糊集的彩色遥感图像边缘检测方法。实验结果表明,它能较好地检测出彩色遥感图像边缘,因此是一种实用有效的彩色遥感图像边缘检测方法。     

Color fidelity and visibility enhancement of underwater image de-hazing by enhanced fuzzy intensification operator

This paper presents an optimization based algorithm for underwater image de-hazing problem. Underwater image de-hazing is the most prominent area in research. Underwater images are corrupted due to absorption and scattering. With the effect of that, underwater images have the limitation of low visibility, low color and poor natural appearance. To avoid the mentioned problems, Enhanced fuzzy intensification method is proposed. For each color channel, enhanced fuzzy membership function is derived. Second, the correction of fuzzy based pixel intensification is carried out for each channel to remove haze and to enhance visibility and color. The post processing of fuzzy histogram equalization is implemented for red channel alone when the captured image is having highest value of red channel pixel values. The proposed method provides better results in terms maximum entropy and PSNR with minimum MSE with very minimum computational time compared to existing methodologies.     

基于局部空间信息的可变类模糊阈值光学遥感图像分割

阈值法分割在光学遥感图像分析中被得到广泛的应用,然而传统阈值法也存在诸多局限性,如对噪声敏感,需人为设定类别数,计算复杂度高等.针对传统闽值法的局限性,提出一种基于局部空间信息的可变类模糊阈值光学遥感图像分割方法.首先,以图像光谱的一阶矩为初始类中心,利用二分法原理和区域间最大相似度准则来快速确定类别数及其中心.然后,...     

基于小波变换和模糊理论的裂纹图像增强算法

针对传统图像增强算法的缺陷, 提出了一种基于小波分析和模糊理论的图像增强算法, 该算法先对原始图像进行小波变换得到图像的高频和低频小波系数, 再定义新的模糊隶属度函数对低频系数进行模糊增强, 对不同方向上的高频系数进行小波阈值去噪, 通过小波重构得到增强后的图像, 所有算法通过Matlab编程验证, 能有效的增强图像, 改善图像的视觉效果. 实验结果表明, 算法是可行有效的.     

自适应直觉模糊相异直方图裁剪的图像增强算法

针对经典的直方图均衡化图像增强算法可能存在的对比度过度增强、亮度分布不均匀和细节信息不突出等问题,提出自适应直觉模糊相异直方图裁剪的图像增强算法.基于直觉模糊集的“投票模型”,引入直觉模糊相异直方图的概念,并基于此提取图像像素的空间位置信息.同时,利用S型隶属度函数对图像直觉模糊相异直方图进行自适应裁剪,采用分段策略对裁剪后的直觉模糊相异直方图进行均衡化处理.最后,利用直觉模糊集的犹豫度刻画原图像的未知信息,修正由引导滤波获得的细节图像,从而保留图像丰富的细节信息.针对3种类型的图像,即自然图像、MRI脑图像及近红外图像的实验结果表明,所提出算法能够有效提高图像的对比度,保留图像的细节信息,使图像呈现较自然的视觉效果,改善图像的质量评价指标.     

A new fuzzy clustering algorithm for the segmentation of brain tumor

This paper introduces a new method of clustering algorithm based on interval-valued intuitionistic fuzzy sets (IVIFSs) generated from intuitionistic fuzzy sets to analyze tumor in magnetic resonance (MR) images by reducing time complexity and errors. Based on fuzzy clustering, during the segmentation process one can consider numerous cases of uncertainty involving in membership function, distance measure, fuzzifier, and so on. Due to poor illumination of medical images, uncertainty emerges in their gray levels. This paper concentrates on uncertainty in the allotment of values to the membership function of the uncertain pixels. Proposed method initially pre-processes the brain MR images to remove noise, standardize intensity, and extract brain region. Subsequently IVIFSs are constructed to utilize in the clustering algorithm. Results are compared with the segmented images obtained using histogram thresholding, k-means, fuzzy c-means, intuitionistic fuzzy c-means, and interval type-2 fuzzy c-means algorithms and it has been proven that the proposed method is more effective.     

应用模糊增强及均值漂移实现红外目标分割

针对复杂环境下红外图像信噪比和对比度低,边缘模糊,目标分割困难的情况,提出一种基于模糊增强和均值漂移图像滤波的红外目标分割方法。首先定义新的隶属度函数,运用模糊集理论进行红外图像增强,避免了传统模糊增强算法的弊病,有效提高目标与背景的对比度;之后利用ICI（交叉置信区）规则确定均值漂移的带宽参数,提出一种新的自适应带宽均值漂移图像滤波方法,实现图像的进一步平滑和聚类;最后利用自适应阈值实现红外目标分割。实验结果表明,算法能够正确有效地分割出复杂环境下的红外目标,并且很好地保持了目标的轮廓细节。     

基于模糊散度理论的图像置乱程度评价研究

提出了基于模糊散度理论的图像置乱程度评价新方法。首先对置乱前后图像中任意像素所对应二阶邻域系统构造一个模糊集并定义其隶属度;其次计算图像中各像素所对应二阶邻域系统模糊集及其补集之间的模糊散度;最后根据置乱前后两图像的模糊散度来构造图像置乱程度评价函数。实验结果表明,提出的评价方法是能够较好地刻画图像的置乱程度,反映了加密次数与置乱程度之间的关系,与人的视觉基本相符。而且对于不同的图像,该评价方法能在一定程度上反映所用的置乱变换在各置乱阶段的效果。     

Strong fuzzy c-means in medical image data analysis

This paper presents a robust fuzzy c-means (FCM) for an automatic effective segmentation of breast and brain magnetic resonance images (MRI). This paper obtains novel objective functions for proposed robust fuzzy c-means by replacing original Euclidean distance with properties of kernel function on feature space and using Tsallis entropy. By minimizing the proposed effective objective functions, this paper gets membership partition matrices and equations for successive prototypes. In order to reduce the computational complexity and running time, center initialization algorithm is introduced for initializing the initial cluster center. The initial experimental works have done on synthetic image and benchmark dataset to investigate the effectiveness of proposed, and then the proposed method has been implemented to differentiate the different region of real breast and brain magnetic resonance images. In order to identify the validity of proposed fuzzy c-means methods, segmentation accuracy is computed by using silhouette method. The experimental results show that the proposed method is more capable in segmentation of medical images than existed methods.     

基于视觉注意和改进隶属度的FSVM彩色图像分割

针对SVM进行图像分割时存在对噪声和孤立点较敏感导致分割结果不佳和抗造性能低下等问题,提出一种基于视觉注意和改进隶属度的FSVM (Modified fuzzy SVM,MFSVM)彩色图像分割方法.该方法在考虑人类视觉显著性检测机制因素的同时,对标准的模糊SVM算法进行改进,新的隶属度函数综合考虑了样本点距离类中心的远近以及样本点的疏密程度,从而有效惩罚噪声点并增强了支持向量的作用.通过彩色图像分割进行验证,结果显示与标准的SVM及基于样本疏密程度隶属度的FSVM分割方法相比,本文方法能够对复杂场景下的彩色进行有效分割,同时呈现出良好的抗噪能力.