Histogram-Based Decision Support System for Extraction and Classification of Leukemia in Blood Smear Images

An abnormality that develops in white blood cells is called leukemia. The diagnosis of leukemia is made possible by microscopic investigation of the smear in the periphery. Prior training is necessary to complete the morphological examination of the blood smear for leukemia diagnosis. This paper proposes a Histogram Threshold Segmentation Classifier (HTsC) for a decision support system. The proposed HTsC is evaluated based on the color and brightness variation in the dataset of blood smear images. Arithmetic operations are used to crop the nucleus based on automated approximation. White Blood Cell (WBC) segmentation is calculated using the active contour model to determine the contrast between image regions using the color transfer approach. Through entropy-adaptive mask generation, WBCs accurately detect the circularity region for identification of the nucleus. The proposed HTsC addressed the cytoplasm region based on variations in size and shape concerning addition and rotation operations. Variation in WBC imaging characteristics depends on the cytoplasmic and nuclear regions. The computation of the variation between image features in the cytoplasm and nuclei regions of the WBCs is used to classify blood smear images. The classification of the blood smear is performed with conventional machine-learning techniques integrated with the features of the deep-learning regression classifier. The designed HTsC classifier comprises the binary classifier with the classification of the lymphocytes, monocytes, neutrophils, eosinophils, and abnormalities in the WBCs. The proposed HTsC identifies the abnormal activity in the WBC, considering the color and shape features. It exhibits a higher classification accuracy value of 99.6% when combined with the other classifiers. The comparative analysis expressed that the proposed HTsC model exhibits an overall accuracy value of 98%, which is approximately 3%–12% higher than the conventional technique.     

Textile fabric defect detection based on low-rank representation

In this paper, we propose a novel and robust fabric defect detection method based on the low-rank representation (LRR) technique. Due to the repeated texture structure we model a defects-free fabric image as a low-rank structure. In addition, because defects, if exist, change only the texture of fabric locally, we model them with a sparse structure. Based on the above idea, we represent a fabric image into the sum of a low-rank matrix which expresses fabric texture and a sparse matrix which expresses defects. Then, the LRR method is applied to obtain the corresponding decomposition. Especially, in order to make better use of low-rank structure characteristics we propose LRREB (low-rank representation based on eigenvalue decomposition and blocked matrix) method to improve LRR. LRREB is implemented by dividing a image into some corresponding blocked matrices to reduce dimensions and applying eigen-value decomposition (EVD) on blocked matrix instead of singular value decomposition (SVD) on original fabric image, which improves the accuracy and efficiency. No training samples are required in our methods. Experimental results show that the proposed fabric defect detection method is feasible, effective, and simple to be employed.

     

Nonlocal image denoising via adaptive tensor nuclear norm minimization

Nonlocal self-similarity shows great potential in image denoising. Therefore, the denoising performance can be attained by accurately exploiting the nonlocal prior. In this paper, we model nonlocal similar patches through the multi-linear approach and then propose two tensor-based methods for image denoising. Our methods are based on the study of low-rank tensor estimation (LRTE). By exploiting low-rank prior in the tensor presentation of similar patches, we devise two new adaptive tensor nuclear norms (i.e., ATNN-1 and ATNN-2) for the LRTE problem. Among them, ATNN-1 relaxes the general tensor N-rank in a weighting scheme, while ATNN-2 is defined based on a novel tensor singular-value decomposition (t-SVD). Both ATNN-1 and ATNN-2 construct the stronger spatial relationship between patches than the matrix nuclear norm. Regularized by ATNN-1 and ATNN-2 respectively, the derived two LRTE algorithms are implemented through the adaptive singular-value thresholding with global optimal guarantee. Then, we embed the two algorithms into a residual-based iterative framework to perform nonlocal image denoising. Experiments validate the rationality of our tensor low-rank assumption, and the denoising results demonstrate that our proposed two methods are exceeding the state-of-the-art methods, both visually and quantitatively.

     

结构化加权稀疏低秩恢复算法在人脸识别中的应用

针对训练样本或测试样本存在污损的情况，提出一种结构化加权稀疏低秩恢复算法（structured and weighted-sparse low rank representation，SWLRR）。SWLRR对低秩表示进行加权稀疏约束和结构化约束，使得低秩表示系数更加趋近于块对角结构，进而可获得具有判别性的低秩表示。SWLRR将训练样本恢复成干净训练样本后，再根据原始训练样本和恢复后的训练样本学习到低秩投影矩阵，把测试样本投影到相应的低秩子空间，即可有效地去除测试样本中的污损部分。在几个人脸数据库上的实验结果验证了SWLRR在不同情况下的有效性和鲁棒性。     

高效的白细胞分割算法

白细胞分割是细胞形态学研究中一个重要、富有挑战性的课题。提出一种基于目标检测的白细胞分割算法。具体地讲,首先根据目标检测方法检测出白细胞,并由白细胞的位置信息得到包含白细胞的子图,然后运用多项式拟合的方法得到子图的灰度直方图的波谷值,再在子图上运用直方图阈值算法分割出细胞核。该算法在分割细胞核的过程中,既可以有效地避免血小板和红细胞等干扰,又能较容易地估计出阈值。对于白细胞的细胞质分割,将白细胞位置信息作为GrabCut算法中人工交互部分,通过迭代法分割出白细胞的细胞质。实验结果表明,该算法能准确地定位白细胞,并根据白细胞的位置信息可以降低白细胞分割的难度,提高其分割的精度和分割效率。特别地,在Cellavision的白细胞图片数据库的实验结果表明,所提的白细胞分割算法对不同类别、不同染色剂和不同拍摄环境下得到的白细胞都能得到较好的分割效果,同时算法又还具有很好的泛化性。     

基于模板校正与低秩分解的纺织品瑕疵检测方法

针对周期性纺织品存在的拉伸变形问题,提出结合模板校正与低秩分解的纺织品瑕疵检测方法.首先对原图像进行模板校正,减少图像拉伸变形对检测结果的影响.然后提出低秩校正分解模型,包含低秩项、稀疏项和校正项,通过交替方向法优化求解,生成低秩矩阵和稀疏矩阵.最后利用最优阈值分割算法,分割由稀疏矩阵产生的显著图,完成瑕疵检测.在标准数据库上的实验表明,文中方法的查全率有所提高.     

一种基于遗传算法的最小交叉熵阈值选择方法

最小交叉熵阈值法(MCET) 在二级阈值中是有效的, 但在多极阈值的穷尽搜索中却要付出昂贵的时间代价. 鉴于此, 提出一种基于遗传算法(GA) 的MCET选择方法: 在执行图像分割(IS) 任务之前, 先将IS 转化为在一定约束 条件下待优化的问题; 在寻找待优化问题最优解的计算过程中引入一种回归设计技巧以存储中间结果; 使用这种回 归设计技巧, 在一组标准测试图像上利用GA搜索待优化问题的最优解. 实验结果表明, 利用所提出的方法获得的多 个阈值非常接近于穷尽搜索获得的结果.

     

Discriminative low-rank representation with Schatten-p norm for image recognition

Low-rank representation (LRR) has attracted much attention recently due to its efficacy in a rich variety of real world applications. Recently, the non-convex regularization has become widely used in the rank minimization problem. In this paper, we propose a discriminative low-rank representation with Schatten-p norm (DLRR-SPN) to learn a robust and discriminative affinity matrix for image recognition. To this end, we first impose the Schatten-p norm regularization on the representation matrix to learn the global structure of data. Moreover, the adaptive distance penalty is used to preserve the local neighbor relationship of data. The objective function is formulated as a Schatten-p norm minimization problem, which can be solved via alternating direction method of multipliers (ADMM). To enhance the separation ability of the discriminative affinity matrix for semi-supervised recognition problem, the angular information of the principal directions of the low-rank representation is further exploited. Finally, an effective semi-supervised classifier is utilized on the learned affinity matrix for final prediction. Extensive experimental results on image recognition demonstrate the effectiveness of the proposed method and its superiority in performance over the related state-of-the-art methods.

     

基于低秩表示的非负张量分解算法

为了提高图像分类准确率,提出了一种基于低秩表示的非负张量分解算法。作为压缩感知理论的推广和发展,低秩表示将矩阵的秩作为一种稀疏测度,由于矩阵的秩反映了矩阵的固有特性,所以低秩表示能有效的分析和处理矩阵数据,本文把低秩表示引入到张量模型中,即引入到非负张量分解算法中,进一步扩展非负张量分解算法。实验结果表明,本文所提算法与其他相关算法相比,分类结果较好。     

Color image segmentation using histogram thresholding - Fuzzy C-means hybrid approach

This paper presents a novel histogram thresholding - fuzzy C-means hybrid (HTFCM) approach that could find different application in pattern recognition as well as in computer vision, particularly in color image segmentation. The proposed approach applies the histogram thresholding technique to obtain all possible uniform regions in the color image. Then, the Fuzzy C-means (FCM) algorithm is utilized to improve the compactness of the clusters forming these uniform regions. Experimental results have demonstrated that the low complexity of the proposed HTFCM approach could obtain better cluster quality and segmentation results than other segmentation approaches that employing ant colony algorithm.     

Discriminative low-rank preserving projection for dimensionality reduction

As an effective image clustering tool, low-rank representation (LRR) can capture the intrinsic representation of the observed samples. However, firstly, the good representation does not mean good classification performance. Secondly, no projection matrix is obtained in the training stage, and it cannot deal with the new samples. By incorporating the discriminant analysis and the local neighborhood relationship of the original samples into the low-rank representation, a novel discriminative low-rank preserving projection (DLRPP) algorithm is presented for dimensionality reduction. In DLRPP, the global structure information can be captured by LRR, and the local geometricinformation is simultaneously preserved by the manifold regularization term. The constrained term is induced by the adaptive graph, which is obtained by low-rank representation coefficients. In addition, by introducing discriminant analysis constraint term, DLRPP can learn an optimal projection matrix for data dimensionality reduction. The numerous experiments on six public image datasets prove that the proposed DLRPP can obtain better recognition accuracy compared with the state-of-the-art feature extraction methods.     

基于低秩表示与矩阵填充的人脸识别方法

基于回归分析的人脸识别方法在处理不完备数据矩阵时,先对矩阵进行填充,再使用人脸识别方法,因此会降低分类性能.为了更有效地执行关于不完备数据的识别,文中将低秩矩阵填充和低秩表示学习整合在同一个模型,提出基于低秩表示和低秩矩阵填充的人脸识别方法.通过最小化表示系数和矩阵秩交替计算样本低秩表示系数矩阵和恢复矩阵缺失项,再使用最近邻分类器实现分类.在一些公开人脸数据集上的实验表明,在训练样本矩阵元素随机缺失时,文中方法可以有效提高识别精度及降低填充误差.     

基于lP范数的非凸低秩张量最小化

在低秩矩阵、张量最小化问题中,凸函数容易求得最优解,而非凸函数可以得到更低秩的局部解.文中基于非凸替换函数的低秩张量恢复问题,提出基于l_p范数的非凸张量模型.采用迭代加权核范数算法求解模型,实现低秩张量最小化.在合成数据和真实图像上的大量实验验证文中方法的恢复性能.     

结合区域和边界信息的图像显著度检测

目的 图像显著度检测是许多图像应用的核心问题,为了能够在复杂背景下准确提取图像中前景对象的位置和尺度信息,提出一种结合区域和边界信息的图像显著度检测方法。方法 对于图像区域信息,提出一种基于图像等照度线的方法检测显著区域信息。该方法针对不同的特征(颜色、亮度和方向)提出统一的计算方法,使得不同特征下获得的显著信息具有一致的度量标准,从而方便后续多特征显著度图的融合。对于图像边界信息,采用一种结合多尺度Beltrami过滤器的全局方法检测显著边界信息。多尺度Beltrami过滤器可以显著增强图像中的边界信息。利用全局显著度检测方法对经过过滤器处理过的图像可以准确地获取图像中最为显著的边界信息。最后,由于区域和边界分别代表图像中的不同类型信息,可以直接采用线性融合方式构建最终的图像显著度图。结果 与其他9种流行图像显著度检测算法相比,本文算法无论在简单还是复杂背景下均能够较为准确地检测出图像中的显著度信息(Precision、Recall、F测试中获得的平均值为0.5905,0.6554,0.7470的最高测试结果)。结论 提出一种结合区域和边界信息的图像显著度检测算法,通过区域和边界信息相结合的方式实现图像中显著对象的准确检测。实验结果表明本文算法具有良好的适用性和鲁棒性,为图像中复杂背景下对象检测打下坚实基础。     

Color image segmentation using adaptive unsupervised clustering approach

This paper presents the Region Splitting and Merging-Fuzzy C-means Hybrid Algorithm (RFHA), an adaptive unsupervised clustering approach for color image segmentation, which is important in image analysis and in understanding pattern recognition and computer vision field. Histogram thresholding technique is applied in the formation of all possible cells, used to split the image into multiple homogeneous regions. The merging technique is applied to merge perceptually close homogeneous regions and obtain better initialization for the Fuzzy C-means clustering approach. Experimental results have demonstrated that the proposed scheme could obtain promising segmentation results, with 12% average improvement in clustering quality and 63% reduction in classification error compared with other existing segmentation approaches.     

A multilevel image thresholding segmentation algorithm based on two-dimensional K–L divergence and modified particle swarm optimization

Multilevel image segmentation is a technique that divides images into multiple homogeneous regions. In order to improve the effectiveness and efficiency of multilevel image thresholding segmentation, we propose a segmentation algorithm based on two-dimensional (2D) Kullback–Leibler(K–L) divergence and modified Particle Swarm Optimization (MPSO). This approach calculates the 2D K–L divergence between an image and its segmented result by adopting 2D histogram as the distribution function, then employs the sum of divergences of different regions as the fitness function of MPSO to seek the optimal thresholds. The proposed 2D K–L divergence improves the accuracy of image segmentation; the MPSO overcomes the drawback of premature convergence of PSO by improving the location update formulation and the global best position of particles, and reduces drastically the time complexity of multilevel thresholding segmentation. Experiments were conducted extensively on the Berkeley Segmentation Dataset and Benchmark (BSDS300), and four performance indices of image segmentation – BDE, PRI, GCE and VOI – were tested. The results show the robustness and effectiveness of the proposed algorithm.     

灰度图像交叉熵阈值法的三维推广

针对现有的灰度图像交叉熵阈值化方法无法有效分割含有混合噪声图像的问题,在图像三维直方图的基础上提出三维交叉熵阈值化算法,同时给出三维交叉熵阈值法的快速递推公式.实验结果表明,三维方法结合了图像中像素的灰度及其局部空间的均值和中值信息,对于含有混合噪声的图像,具有比现有交叉熵阈值化算法更好的分割效果.     

Automatic segmentation of dermoscopy images using saliency combined with adaptive thresholding based on wavelet transform

Segmentation is the essential requirement in automated computer-aided diagnosis (CAD) of skin diseases. In this paper, we propose an unsupervised skin lesion segmentation method to challenge the difficulties existing in the dermoscopy images such as low contrast, border indistinct, and skin lesion is close to the boundary. The proposed method combines the enhanced fusion saliency with adaptive thresholding based on wavelet transform to get the lesion regions. Firstly, a fusion saliency map increases the contract of the skin lesion and healthy skin, and then an adaptive thresholding method based on wavelet transform is used to obtain more accurate lesion regions. We compare the proposed method with seven state-of-the-art approaches using a series of evaluation metrics on both PH2 and ISBI2016 datasets. The results demonstrate the effectiveness of the proposed method superior to the state-of-the-art approaches in accordance with quantitative results and visual effects.

     

工业检测图像灰度波动变换自适应阈值分割算法

工业检测图像经常受到不均光照的影响,对该类图像局部自适应分割算法比全局算法能产生更好的分割效果. 但局部算法中基于分块的算法对分块方法缺乏指导,而基于邻域的算法容易在背景或前景内部产生误分. 针对上述缺点,本文提出了一种多方向灰度波动变换的自适应阈值分割算法. 该算法先从多个方向依照灰度波动对图像进行转换,构造以多维向量为基础的灰度波动变换矩阵, 然后利用主成分分析法(Principal component analysis, PCA)将高维向量压缩至一维并生成变换图像,最后运用Otsu算法分割变换图像. 该算法无需分块,并且仅需波动幅度阈值和布尔型背景色两个参数. 实验结果表明,该算法能够有效减少不均光照对工业检测图像分割的影响, 与Niblack法、Sauvola法等几种局部算法相比,该法在分割效果上具有了明显的提升.     

基于广义调和均值距离的图像分割新算法

提出了基于广义调和均值距离的最小偏差图像阈值化分割新算法。Otsu阈值法是图像分割中最典型阈值法之一,因其计算简单、速度快和性能稳定等优点而在图像分割中得到广泛应用;但是,传统Otsu阈值法是基于欧式距离的最小偏差阈值法,由于欧式距离没有可调节参数而导致Otsu阈值法分割图像缺乏鲁棒性。首先将Otsu图像分割法中的欧式距离用广义调和均值距离代替并得到一种具有鲁棒性的图像分割新算法,其次给出该算法中参数选取办法。大量实验结果表明,新的图像分割算法相比Otsu法更有效。