为探索基于光学相干层析成像(OCT)技术对古代青瓷釉层物理结构的分类, 综合应用OCT技术、X射线荧光光谱分析(XRF)技术、扫描电子显微镜-能谱分析(SEM-EDS)技术和激光拉曼光谱(LRS)技术对河南省宝丰县清凉寺窑址出土的金元时期青瓷和钧瓷样品残片进行了分析。根据获取的样品釉层物理结构OCT灰度图像特征对釉层进行定性分类, 利用图像纹理分析技术对釉层OCT图像进行量化表征, 并根据所确定的纹理特征参数进行主成分分析。对根据OCT图像对瓷釉的分类结果与根据XRF获得的釉层化学成分的分类结果进行比较, 结合SEM-EDS和LRS分析结果讨论了釉层材料学特征与OCT图像特征之间的内在联系。

Optical coherence tomography (OCT) is employed in the diagnosis of skin cancer. Particularly, quantitative image features extracted from OCT images might be used as indicators to classify the skin tumors. In the present paper, we investigated intensity-based, texture-based and fractalbased features for automatically classifying the melanomas, basal cell carcinomas and pigment nevi. Generalized estimating equations were used to test for differences between the skin tumors. A modified p value of <0.001 was considered statistically significant. Significant increase of mean and median of intensity and significant decrease of mean and median of absolute gradient were observed in basal cell carcinomas and pigment nevi as compared with melanomas. Significant decrease of contrast, entropy and fractal dimension was also observed in basal cell carcinomas and pigment nevi as compared with melanomas. Our results suggest that the selected quantitative image features of OCT images could provide useful information to differentiate basal cell carcinomas and pigment nevi from the melanomas. Further research is warranted to determine how this approach may be used to improve the classification of skin tumors.

A leukocyte recognition method for human peripheral blood smear based on island-clustering texture (ICT) is proposed. By analyzing the features of the five typical classes of leukocyte images, a new ICT model is established. Firstly, some feature points are extracted in a gray leukocyte image by mean-shift clustering to be the centers of islands. Secondly, the growing region is employed to create regions of the islands in which the seeds are just these feature points. These islands distribution can describe a new texture. Finally, a distinguished parameter vector of these islands is created as the ICT features by combining the ICT features with the geometric features of the leukocyte. Then the five typical classes of leukocytes can be recognized successfully at the correct recognition rate of more than 92.3% with a total sample of 1310 leukocytes. Experimental results show the feasibility of the proposed method. Further analysis reveals that the method is robust and results can provide important information for disease diagnosis.

The present work focuses on the development of a novel computer-based approach for tear ferning (TF) featuring. The original TF images of the recently developed five-point grading scale have been used to assign a grade for any TF image automatically. A vector characteristic (VC) representing each grade was built using the reference images. A weighted combination between features selected from textures analysis using gray level co-occurrence matrix (GLCM), power spectrum (PS) analysis and linear specificity of the image were used to build the VC of each grade. A total of 14 features from texture analysis were used. PS at different frequency points and number of line segments in each image were also used. Five features from GLCM have shown significant differences between the recently developed grading scale images which are: angular second moment at 0° and 45°, contrast, and correlation at 0° and 45°; these five features were all included in the characteristic vector. Three specific power frequencies were used in the VC because of the discrimination power. Number of line segments was also chosen because of dissimilarities between images. A VC for each grade of TF reference images was constructed and was found to be significantly different from each other's. This is a basic and fundamental step toward an automatic grading for computer-based diagnosis for dry eye.

为了对古代瓷釉的结构特征进行量化表征，基于图像的统计直方图以及灰度共生矩阵提取出了7 个纹理特征参数，用来对4 种不同类型典型素面瓷釉样品的光学相干层析(OCT)图像进行描述。对比分析发现4 种瓷釉的7个纹理特征参数差异明显。同时利用未知样品对基于K-邻近分类和7 个纹理特征参数的瓷釉自动识别方案进行了可行性验证，得到了未知样品的正确的分类结果。实验结果表明：7个特征参数可以很好地表征瓷釉的釉层结构特征，基于K-邻近分类和纹理特征参数量化分析的瓷釉快速自动识别方案是可行的，且具有广泛的应用前景。

研究皮肤纹理经过UVB光照射后的变化情况并对其进行识别。具体地, 采取图像纹理分析方法, 对经过光照射后不同时期的小鼠皮肤图像提取纹理特征, 进而建立一种新的皮肤纹理识别模型。采用空间灰度共生矩阵法提取图像纹理的4个主要特征, 即: 能量, 熵, 惯性矩, 相关度, 然后利用神经网络中的NNtool对皮肤纹理图像进行训练和分类识别。实验结果很好地证明了这种纹理分析和识别方法的可行性和有效性。

在分析针对局部二值模式进行降维方法的基础上, 提出了一种改进的局部二值模式描述符, 并用于图书文档图像分类。新方法首先将原局部邻域划分为多个 4-正交邻域, 然后统计 4-正交邻域二值化后所包含的“ 1”的个数作为特征, 最后融合所有 4-正交邻域特征进行图像分类。采用广泛应用的纹理图像库、前视红外目标图像库和图书文档图像库进行实验, 结果表明, 新方法的特征维数不但明显降低, 而且还取得了较高的分类准确率。

为了实现在包含部分变化信息的相同区域不同时相的高分辨率航空影像上获得准确一致的分割结果, 提出了两时相高分辨率航空影像联合分割方法。首先, 对两时相影像进行高斯平滑处理, 减小地物的内部差异以避免过分割; 然后将两时相影像进行波段组合, 采用主成分分析法剔除冗余数据, 取其第一主分量作为灰度分量; 最后对原始影像进行纹理分析, 获得两时相影像的纹理信息作为纹理分量并与灰度分量组合在一起进行MeanShift分割。实验对比结果表明, 该方法能够有效利用数据, 节省处理时间, 获得了较好的分割结果。

纸张等薄片的计数在工业生产中使用得非常普遍, 而传统的纸张计数方法大多是通过专门的机械装置甚至人工来完成的, 具有成本高、体积大、效率低、准确率低等各种突出问题, 迫切需要改进。针对彼此平行放置的纸张, 提出了一种沿其放置方向概率统计的算法, 在采样位置处统计正交于放置方向的纸张数目, 最后选取出现概率最大的一个数据作为统计结果。考虑到纸张之间彼此不平行且出现重叠的情况, 提出了一种全局轮廓的统计算法, 经过预处理, 对纸张边缘轮廓统计计数。实验结果表明, 该算法有效地克服了因为遮挡和重叠而引起的统计错误。

We report the results of a comparative study of Fourier domain analysis (FDA) and texture analysis (TA) of optical coherence tomography (OCT) images of resected human breast tissues for binary classification between normal-abnormal classes and benign-malignant classes. With the incorporation of Fisher linear discriminant analysis (FLDA) in TA for feature extraction, the TA-based algorithm provided improved diagnostic performance as compared to the FDAbased algorithm in discriminating OCT images corresponding to breast tissues with three different pathologies. The specificity and sensitivity values obtained for normal-abnormal classification were both 100%, whereas they were 90% and 85%, respectively for benign-malignant classification.