基于热红外图像处理技术的农作物冠层识别方法研究

为解决农作物冠层热红外图像边缘灰度级分布不均且噪声较大,而传统图像分割方法难以实现其目标区域有效识别的难题,以苗期红小豆冠层热红外图像为研究对象,将模糊神经网络和仿射变换有机结合,提出了基于热红外图像处理技术的农作物冠层识别模型。首先利用五层线性归一化模糊神经网络的自适应特性,选取高斯隶属度函数,自动计算冠层可见光图像识别的推理规则,有效地分割了可见光图像中的冠层区域。通过分析3种分割指标和熵,定量评价可见光图像冠层分割质量。网络迭代38次时,误差精度为0.000 952,该算法平均有效识别率为96.13%,获取可见光冠层图像的像元信息熵值范围为2.454 4~5.198 7,与标准算法所得冠层图像的像元信息熵仅相差0.245 9。然后以取得可见光图像的冠层有效区域为参考图像,采用仿射变换算法,调整优选平移、旋转、缩放等图像变换因子,配准原始热红外图像,提出了基于仿射变换的冠层热红外图像识别方法。对于初始温度范围值在16.35~19.92 ℃的农作物热红外图像,计算选取旋转幅度为1.0和缩放因子为0.9时,作为异源图像的最优配准参数,获取目标图像的最大温差为3.17 ℃,相对于原图像的平均温度值由18.711 ℃下降至17.790 ℃,进而实现了基于热红外图像处理技术的农作物冠层识别。最后以熵的互信息作为监督指标,对农作物冠层热红外图像识别方法进行评价。提出的冠层热红外图像识别方法,所获取的目标图像与初始热红外图像的平均互信息为4.368 7,标准目标图像和初始热红外图像的平均互信息为3.981 8,二者仅相差0.486 9。同时,两种冠层热红外图像的平均温度差值为0.25 ℃,高效消除了原始热红外图像的背景噪声。结果表明本研究方法的有效性和实用性,能够为应用热红外图像反映农作物生理生态信息特征指标参数提供技术借鉴。     

基于偏振测量的雾天图像场景分割

现有场景分割方法主要依赖于图像亮度、颜色和纹理等特征,然而在雾天图像中提取这些特征将变得困难且不稳定.基于此本文提出了适用于雾天图像场景分割的特征矢量,以及相应的特征提取算法.特征矢量由目标偏振度、深度和颜色三部分组成.特征提取算法分别为:用去相关的方法从图像偏振度分离出大气偏振度和目标偏振度;根据雾天退化模型和雾天图像偏振表示形式推导出场景深度信息;利用两幅偏振图像求出非偏振彩色图像,从而得到场景的颜色信息.将这些特征构成的特征矢量用于基于图的分割算法中,并从两个方面比较了仅使用颜色特征和使用本文特征矢量的分割结果.最后得出结论:对于雾天图像而言,这些特征比通常的颜色特征更加有效和鲁棒.     

改进的多分辨纹理图像分割算法

提出了一种改进的有监督纹理图像的分割算法。基于实际纹理图像是分割图像叠加了不规则噪声的假设,用被污染的高斯分布描述待分割的图像,并且利用多分辨模型得到代分辨层上的模型参数,从而实现由粗到细直到纹理图像的每个像素的分割。另外在禽域关联为先验信息利用上更为合理。所以这种方法不仅计算量小,而且分割结果也较为精确。     

Texture based segmentation of epithelial layer from oral histological images

The objective of this paper is to provide a texture based segmentation algorithm for better delineation of the epithelial layer from histological images in discriminating normal and oral sub-mucous fibrosis (OSF). As per literature and oral clinicians, it is established that the OSF initially originates and propagates in the epithelial layer. So, more accurate segmentation of this layer is extremely important for a clinician to make a diagnostic decision. In doing this, Gabor based texture gradient is computed in gray scale images, followed by preprocessing of the microscopic images of oral histological sections. On the other hand, the color gradients of these images are obtained in the transformed Lab color space. Finally, the watershed segmentation is extended to segment the layer based on the combination of texture and color gradients. The segmented images are compared with the ground truth images provided by the oral experts. The segmentation results depict the superiority of the texture based segmentation in comparison to the Otsu's based segmentation in terms of misclassification error. Results are shown and discussed.     

大豆冠层多光谱图像提取方法

为解决大豆冠层在近地端的多光谱图像边缘灰度不均,目标与背景之间灰度差别小,难以准确高效地获取大豆冠层目标区域的难题,将多光谱成像处理技术与经典图像分割方法有机融合,提出基于多光谱图像处理技术的大豆冠层提取方法。以东北大豆为对象,通过Sequoia多光谱相机采集绿光、近红外、红光、红边和可见光五类大豆多光谱图像,采用高斯平滑滤波法对原始大豆多光谱图像进行预处理,分析多光谱图像中大豆冠层和背景的灰度直方图分布特性,在此基础上利用迭代法、Otsu法和局部阈值法提取原大豆多光谱图像中冠层区域,并以图像形态学开运算处理细化和扩张背景,避免图像区域内干扰噪声对大豆冠层识别效果的影响,同时以有效分割率、过分割率、欠分割率、信息熵以及运行时间等为监督指标,对大豆冠层多光谱图像识别模型进行效果评价。大豆冠层识别模型中迭代法可以有效分割近红外和可见光大豆冠层图像,有效分割率分别为97.81%和87.99%,对绿光、红光和红边大豆冠层图像分割效果较差,有效分割率低于70%;Otsu法和局部阈值法可以有效分割除红光波段的其余四种多光谱大豆冠层图像,且有效分割率均在82%以上;三种算法对红光大豆冠层图像的有效分割率均低于20%,未达到较好效果。在原始多光谱图像中应用迭代法、Otsu法和局部阈值法提取大豆冠层图像与标准图像的信息熵平均值波动幅度分别为：0.120 1,0.054 7和0.059 8,其中Otsu法和局部阈值法较小,表明了对于大豆冠层多光谱图像识别中两种算法的有效性。该算法中Otsu法和局部阈值法均可以有效提取绿光、近红外、红边和可见光等多光谱的大豆冠层图像,二者较为完整地保留了大豆冠层信息,其中Otsu法实时性能较局部阈值法更好。该成果为提取农作物冠层多光谱图像提供理论依据和技术借鉴。     

基于面向对象的QuickBird影像退耕地树冠信息提取

分割算法的改进和特征空间的优化足采用面向对象技术提高退耕地树冠信息提取精度的关键,也是利用高分辨率影像提取树冠信息急需解决的问题.文章采用光谱阈值对QuickBird多光谱影像进行一级分割,获得了植被区域,并采用改进的基于边缘的算法对非线性滤波处理后的全色影像进行二级分割,选取光谱、形状和纹理特征组成的特征空间对退耕还林地树冠信息进行提取.结果表明,提取总体精度为84.67%,较传统方法提高17%,KAPPA系数为0.795 3,较传统方法提高0.168.该研究方法能实现较为精确的树冠信息提取,可为管理部门实施准确的监测提供依据,对快速评价退耕还林效果具有重要意义.     

A novel content-based active contour model for brain tumor segmentation

Brain tumor segmentation is a crucial step in surgical and treatment planning. Intensity-based active contour models such as gradient vector flow (GVF), magneto static active contour (MAC) and fluid vector flow (FVF) have been proposed to segment homogeneous objects/tumors in medical images. In this study, extensive experiments are done to analyze the performance of intensity-based techniques for homogeneous tumors on brain magnetic resonance (MR) images. The analysis shows that the state-of-art methods fail to segment homogeneous tumors against similar background or when these tumors show partial diversity toward the background. They also have preconvergence problem in case of false edges/saddle points. However, the presence of weak edges and diffused edges (due to edema around the tumor) leads to oversegmentation by intensity-based techniques. Therefore, the proposed method content-based active contour (CBAC) uses both intensity and texture information present within the active contour to overcome above-stated problems capturing large range in an image. It also proposes a novel use of Gray-Level Co-occurrence Matrix to define texture space for tumor segmentation. The effectiveness of this method is tested on two different real data sets (55 patients - more than 600 images) containing five different types of homogeneous, heterogeneous, diffused tumors and synthetic images (non-MR benchmark images). Remarkable results are obtained in segmenting homogeneous tumors of uniform intensity, complex content heterogeneous, diffused tumors on MR images (T1-weighted, postcontrast T1-weighted and T2-weighted) and synthetic images (non-MR benchmark images of varying intensity, texture, noise content and false edges). Further, tumor volume is efficiently extracted from 2-dimensional slices and is named as 2.5-dimensional segmentation.     

Investigation of relations between skin cancer lesions’ images and their fluorescent spectra

This investigation is based on images obtained from healthy tissue and skin cancer lesions and their fluorescent spectra of cutaneous lesions derived after optical stimulation. Our analyses show that the lesions’ spectra of are different of those, obtained from normal tissue and the differences depend on the type of cancer. We use a comparison between these “healthy” and “unhealthy” spectra to define forms of variations and corresponding diseases. However, the value of the emitted light varies not only between the patients, but also depending on the position of the tested area inside of one lesion. These variations could be result from two reasons: different degree of damaging and different thickness of the suspicious lesion area. Regarded to the visible image of the lesion, it could be connected with the chroma of colour of the tested area and the lesion homogeneity that corresponds to particular disease. For our investigation, images and spectra of three non-melanoma cutanous malignant tumors are investigated, namely—basal cell carcinoma, squamous cell carcinoma, and keratoacanthoma. The images were processed obtaining the chroma by elimination of the background—healthy tissue, and applying it as a basic signal for transformation from RGB to Lab colorimetric model. The chroma of the areas of emission is compared with the relative value of fluorescence spectra. Specific spectral features are used to develop hybrid diagnostic algorithm (including image and spectral features) for differentiation of these three kinds of malignant cutaneous pathologies.     

Region Adaptive Single Image Dehazing

Image haze removal is essential in preprocessing for computer vision applications because outdoor images taken in adverse weather conditions such as fog or snow have poor visibility. This problem has been extensively studied in the literature, and the most popular technique is dark channel prior (DCP). However, dark channel prior tends to underestimate transmissions of bright areas or objects, which may cause color distortions during dehazing. This paper proposes a new single-image dehazing method that combines dark channel prior with bright channel prior in order to overcome the limitations of dark channel prior. A patch-based robust atmospheric light estimation was introduced in order to divide image into regions to which the DCP assumption and the BCP assumption are applied. Moreover, region adaptive haze control parameters are introduced in order to suppress the distortions in a flat and bright region and to increase the visibilities in a texture region. The flat and texture regions are expressed as probabilities by using local image entropy. The performance of the proposed method is evaluated by using synthetic and real data sets. Experimental results show that the proposed method outperforms the state-of-the-art image dehazing method both visually and numerically.     

A Texture Combined Multispectral Magnetic Resonance Imaging Segmentation for Nasopharyngeal Carcinoma

Tumor segmentation from magnetic resonance imaging (MRI) is important for volume estimation and visualization of nasopharyngeal carcinoma (NPC). In some cases, segmentation using the general multispectral (GM) method often obtained poor results due to the high false positives caused by complex anatomic structures and serious overlap in feature space. In this study, a texture combined multispectral fuzzy clustering (TCMFC) segmentation algorithm was proposed. A texture measure of T1-weighted (T1) MR image was introduced by calculating the two-order central statistical information of every pixel within a window after the window convolution operation. The texture measure and the intensities in T1 and contrast-enhanced T1 images formed the new 3-D feature vector for fuzzy clustering implemented by semi-supervised fuzzy c-means (SFCM). Testing showed that by reducing the false positives significantly, the TCMFC method achieved improved segmentation results, compared with the GM method.     

基于互信息改进型脉冲耦合神经网络多值图像分割

为了自动地进行图像的多值分割,从原始图像与分割图像之间的相互关系出发,以最大互信息为优化分割目标,以互信息熵差作为一种新的分类类数判据,在对传统脉冲耦合神经网络模型改进的基础上,提出了一种基于最大互信息改进型脉冲耦合神经网络图像多值分割算法.理论分析和实验结果表明,该方法能够自动确定最佳分割迭代次数及最佳分割灰度类数,对分割图像具有良好的特征划分能力,且在分割类数较少的情况下,能较好地保持图像细节、纹理及边缘等信息,对不同图像分割准确度高,具有较强的适用性.     

An effective noise removal technique for recovering the shapes of diamond abrasive grains in SLM images degraded by clustered spike noise

A new noise-removal technique is applied to scanning laser microscopic (SLM) images to remove clustered spike noise in the images and to recover the shapes of diamond abrasive grains degraded by the noise. For achievement of this purpose, noise points in the SLM image are accurately detected by taking advantage of their properties in the space and spatial-frequency regions. The noise points are removed by a method of smoothing that is based on linear interpolation; that is, their pixel values are replaced by the interpolated values of their non-noise neighboring points. Noise-point information in the space region is acquired from image segmentation based on pixel classification, while noise-point information in the frequency region is derived from redundant wavelet decomposition for the SLM image. Fisher's linear discriminant method is used to yield the two noise-point images. The degraded grain shapes in the SLM images at different noise levels are satisfactorily recovered with a single iteration of smoothing without losses in sharp edges although a single smoothing needed four interpolations. Thus, the present noise-removal technique is shown to be effective for recovering the original shapes of the grains in every SLM image.     

一种基于图像特征和神经网络的苹果图像分割算法

苹果识别是开发苹果采摘机器人的关键环节,利用图像处理技术和神经网络分类器探索苹果图像分割算法.从苹果树图片中选取苹果图像样本和背景网像样本.分别计算这两类图像样本的颜色特征和纹理特征.颜色特征的计算基于RGB色彩模型,纹理特征的计算基于灰度共生矩阵.选取适当的颜色特征(R/B值)和纹理特征(对比度值和相关性值)作为输入节点,利用反向传播神经网络分类器建模,输出值是一个O～1之间的计算值.通过阈值将输出结果分类为苹果或背景.试验结果表明,该算法正确率大于87.6%,对光照的影响不敏感,是一利较为实用的苹果分割算法.     

图像复原处理方法在显微测量LiNbO_3表面缺陷中的应用

研究了图像复原处理在显微测量中的应用。在实测系统点扩展函数的基础上,首先利用图像复原方法对显微图像进行复原处理,然后用二维处理方法实现对缝宽的精确显微测量。实现了在普通光学显微镜下铌酸锂(LiN- bO_3)极化畴结构样品缺陷大小的便捷测量。     

Automatic registration of serial sections of mouse lymph node by using Image-Reg

The first step towards the three-dimensional (3D) reconstruction of histological structures from serial sectioned tissue blocks is the proper alignment of microscope image sequences. We have accomplished an automatic rigid registration program, named Image-Reg, to align serial sections from mouse lymph node and Peyer's patch. Our approach is based on the calculation of the pixel-correlation of objects in adjacent images. The registration process is mainly divided into two steps. Once the foreground images have been segmented from the original images, the first step (primary alignment) is performed on the binary images of segmented objects; this process includes rotation by using the moments and translation through the X, Y axes by using the centroid. In the second step, the matching error of two binary images is calculated and the registration results are refined through multi-scale iterations. In order to test the registration performance, Image-Reg has been applied to an image and its transformed (rotated) version and subsequently to an image sequence of three serial sections of mouse lymph node. In addition, to compare our algorithm with other registration methods, three other approaches, viz. manual registration with Reconstruct, semi-automatic landmark registration with Image-Pro Plus and the automatic phase-correlation method with Image-Pro Plus, have also been applied to these three sections. The performance of our program has been also tested on other two-image data sets. These include: (a) two light microscopic images acquired by the automatic microscope (stitched with other software); (b) two images fluorescent images acquired by confocal microscopy (tiled with other software). Our proposed approach provides a fast and accurate linear alignment of serial image sequences for the 3D reconstruction of tissues and organs.     

基于邻域分割的空谱联合稀疏表示高光谱图像分类技术研究

传统的高光谱遥感影像分类算法侧重于光谱信息的应用。随着高光谱遥感影像的空间分辨率的增加,高光谱影像中相同类别的地物在空间分布上呈现聚类特性,将空间特性有效地应用于高光谱遥感影像分类算法对分类精度的提升非常关键。但是,高光谱影像的高分辨率提供空间聚类特性的同时,在不同地物边缘处表现出的差异性更加明显,若不对空间邻域像素进行甄选,直接将邻域光谱信息引入,设计空谱联合稀疏表示进行图像分割,则分类误差较大,收敛速度大大降低。将光谱角引入空谱联合稀疏表示图像分类理论中,提出了一种基于邻域分割的空谱联合稀疏表示分类算法。该算法利用光谱角计算相邻像素的空间相似度,剥离相似度较低的邻域像素,将相似度高的邻域像素定义为同类地物,引入空谱联合稀疏表示模型中,采用子联合空间追踪算子和联合正交匹配追踪算子对其优化求解,以最小重构误差为准则进行分类。选取AVIRIS及ROSIS典型光谱影像数据进行实验仿真,从中可以看出,随着光谱角分割阈值的提高,复杂的高光谱影像分类精度和平滑区域的高光谱影像分类精度均逐步提高,表明邻域分割在空谱联合稀疏表示分类中的必要性。     

Automatic analysis for biochip based on macromolecular compound material

This paper illustrates a way of quantifying fluorescent chromogenic information through the image processing and identification, and analyzes the correlations between fluorescent chromogenic reaction and a probe. This analytical method is an important reference for probe development, and also used for analyzing the biochip interaction. The relationship between the same type but differing concentrations of probe and fluorescent images was derived. With light field analysis of probe attachment, we performed numerical analysis of the fluorescent signal in accordance with the method of biological area analysis. Through this method, biochips can simultaneously provide many types of quantitative and qualitative figures for research reference.     

基于支持度变换和top-hat分解的双色中波红外图像融合

为了解决用多尺度top-hat分解法融合双色中波红外图像时经常存在对比度提升有限、边缘区域失真较重的问题,提出了基于支持度变换和top-hat分解相结合的融合方法。先用支持度变换法将双色中波图像分解为低频图像和支持度图像序列;再从最后一层低频图像中用多尺度top-hat分解法提取各自的亮信息和暗信息;用灰度值取大法分别融合亮信息和暗信息;通过灰度值归一化和高斯滤波分别增强亮、暗信息融合图像;然后融合两低频图像和亮、暗信息增强图像;将融合图像作为新的低频图像和用灰度值取大法融合得到的支持度融合图像序列进行支持度逆变换,得到最终融合图像。该方法的实验结果同采用单一的支持度变换法融合和多尺度top-hat分解法融合相比,融合图像的对比度提升了11.69%,失真度降低了63.42%,局部粗糙度提高了38.12%。说明提出的从低频图像提取亮暗信息,并经过分别融合、增强,再与低频图像进行融合,能有效破解红外融合图像对比度提升和边缘区域失真度降低之间的矛盾,为提高图像融合质量提供了新方法。     

Correction of Fluorescent Images of Biological Tissues for the Effects of Light Absorption and Scattering

A method for correcting the influence of light attenuation processes in biological tissues on their fluorescent images is proposed. The transfer function that takes into account the radiation losses in the medium at the wavelengths of excitation and emission of fluorescence is calculated analytically, depending on the transport scattering index and the hemoglobin tissue index. The latter is determined on the basis of a color image of the tissue in reflected visible light. Verification of the developed method was carried out on the basis of computer simulation of fluorescent images of biological tissue by the Monte Carlo method.     

基于标准温度法的电弧高温区自动判别研究

在电弧等离子体的光谱诊断中,标准温度法测温原理与目前先进的图像传感技术相结合,通过特征谱图像完成电弧全场温度信息采集,因其良好的时、空分辨率而被广泛应用于电弧温度测量。但是谱线的发射系数与等离子体温度不是单调变化关系,传统标准温度法选取一条ArⅠ谱线完成对电弧等离子体的测量,在电弧内部的高温电离区域产生谱线辐射强度降低的现象,需要人为判定电弧不同位置所处的温度区间才能完成温度的计算,整个过程无法通过软件自主完成。针对此问题,根据电弧等离子体的局部热力学平衡条件,探索一种基于双特征谱线的标准温度法测温原理,通过融合电弧在外层低温区域聚集的Ar原子发出的ArⅠ谱线发射系数场,和在高温区域的Ar一次电离离子所发出的ArⅡ特征谱线发射系数场,将达到ArⅠ谱线标准温度的位置处的ArⅡ谱线发射系数作为电弧不同温度区域的判定依据,完成电弧等离子体高温区域的自动判别,继而应用ArⅠ谱线发射系数与温度对应关系在电弧高、低温区域分别计算电弧温度,消除单一的ArⅠ谱线发射系数场暗区给计算带来的不利影响;设计并搭建了一种镜前分幅采集系统,其中分光镜将弧光等能量分成两束,利用两组反射镜和窄带滤光片建立起两路光学通道,使CMOS在一次曝光中完成两组电弧特征谱图像的采集,并且两幅图像的采集时刻、焦距、光圈等拍摄参数完全一致,达到良好的时间、空间一致性,从而减小谱线融合时误差的输出,满足了原位获取两组电弧特征谱图像的需求;为验证测量系统可行性以及后期的电弧图像提取,以黑白棋盘为标靶,用Harris算子对系统采集的图像进行扫描,根据角点坐标证明系统所采集的两幅图像具有良好的一致性,并且据此将两幅图像做归一化处理,以便后期的电弧特征谱图像的提取;通过假设所测电弧等离子具有轴对称属性,以CMOS所采集的特征谱图像亮度信息作为电弧发射系数场在不同角度下的投影依据,经过中值滤波降噪后,利用ML-EM迭代重建算法求解电弧的三维发射系数分布。实验中,选择受自吸收效应影响较小的ArⅠ696.5 nm谱线和ArⅡ480.6 nm谱线为测量目标,并且在696.5 nm谱线的光通路中加入OD0.4的中性减光片,使两幅特征谱图像的最高亮度值保持一致。选取150A焊接等离子弧为测量对象,经ML-EM法三维还原后,将两条谱线发射系数场等像素融合,在ArⅠ谱线发射系数达到最大值的像素点位置处,ArⅡ谱线发射系数达到ε_rp,判定ArⅡ谱线发射系数大于ε_rp的像素点位置为电弧高温区域,其余位置为低温区域,最终在不同温度区域自动完成焊接等离子弧的温度计算。实验结果表明696.5 nm谱线和480.6 nm谱线发射系数场融合后可以自动识别电弧高温区域,继而完成电弧等离子体的自动测量,为电弧温度实时监测的实现提供更多可能。