小儿指纹采集与图像处理研究

发布时间：2018-04-10 17:26

本文选题：中医儿科 + 小儿指纹　；参考：《山东中医药大学》2013年硕士论文

【摘要】：目的：探索图像采集与处理技术在中医小儿指纹研究中的应用。通过研究设计图像采集硬件，并进行临床指纹图像采集以及指纹图像处理分析，以此实现小儿指纹图像的客观化、数字化的采集及研究，从而探讨中医理论的客观化以及进行特色诊疗方法的研究。论文分为5部分：1.绪论；2.小儿指纹采集硬件设计；3.小儿指纹临床样本采集；4.小儿指纹图像处理研究；5.总结与讨论。方法：采用维视高分辨率彩色数字摄像机，配合手动变焦镜头以及环形LED视觉光源搭建图像采集硬件设备，同时设计图像采集软件平台。临床采集不同年龄、性别、病症的小儿指纹图像并记录其临床症状、体征及病情变化等信息存档，同时采集正常小儿的指纹图像以作对照实验。研究当前图像处理、分割方法，并利用MATLAB软件平台对小儿指纹图像进行处理分析，，对比图像处理效果。结果：采用图像采集系统能够客观形象、实时地观察小儿指纹，同时采集的指纹图像可以长期保存，便于后期处理观察分析。各种图像处理算法能够明确清晰的分辨出小儿指纹三关位置以及指纹长短特征。处理后的图像结果显示与肉眼主观观测的指纹特征相吻合。通过对各方法的探讨，利用MATLAB环境进行实验，总结各传统图像分割算法的优缺点。图像边缘检测方法对于指纹特征提取分析具有显著效果，其中Canny图像分割算法较其他算法在指纹三关分析上优势明显，水平集算法对于处理分析小儿食指轮廓以及指纹长短特征优势明显。三维指纹图像能立体客观地反映出指纹的长短、浮沉等特征信息。结论：本文实现了利用彩色数字图像摄像机临床采集小儿指纹图像，数据收集做到客观、真实、准确、方便。同时利用MATLAB进行边缘检测、图像分割等操作，提取了我们所感兴趣的信息，以此实现小儿指纹图像的客观化的采集及研究，避免临床医生诊断时的主观偏差，为临床诊断和教学工作提供一定的参考。
[Abstract]:Objective: to explore the application of image acquisition and processing technology in the study of pediatric fingerprint of traditional Chinese medicine (TCM).Through the research and design of image acquisition hardware, clinical fingerprint image acquisition and fingerprint image processing and analysis, in order to achieve the objective of the child fingerprint image, digital acquisition and research,So as to explore the objectivity of the theory of traditional Chinese medicine and the study of characteristic diagnosis and treatment methods.The paper is divided into five parts: 1.Introduction 2.Hardware design of fingerprint collection in children.The clinical samples of pediatric fingerprint were collected.Study on fingerprint Image processing in ChildrenSummary and discussionMethods: the hardware equipment of image acquisition was built by using high resolution color digital camera, manual zoom lens and annular LED visual light source. At the same time, the image acquisition software platform was designed.The fingerprint images of children with different age, sex and disease were collected and the clinical symptoms, signs and changes of the disease were recorded and archived. The fingerprint images of normal children were collected for the control experiment at the same time.The current image processing and segmentation methods are studied, and the MATLAB software platform is used to process and analyze the fingerprint images of children, and the effect of image processing is compared.Results: the image acquisition system can be used to observe the fingerprint of children in real time, and the collected fingerprint image can be preserved for a long time, which is convenient for post-processing observation and analysis.All kinds of image processing algorithms can clearly distinguish the position and length of the child fingerprint.The results of the processed images are consistent with the fingerprint features observed by the naked eye.The advantages and disadvantages of the traditional image segmentation algorithms are summarized through the discussion of the methods and the experiments in the MATLAB environment.Image edge detection method has a significant effect on fingerprint feature extraction and analysis, and Canny image segmentation algorithm has obvious advantages over other algorithms in fingerprint three-level analysis.The level set algorithm has a significant advantage in processing and analyzing the forefinger contour and fingerprint length and length characteristics in children.Three-dimensional fingerprint image can reflect the characteristic information of fingerprint such as length, float and sink.Conclusion: the color digital image camera is used to collect the fingerprint images of children. The data collection is objective, true, accurate and convenient.At the same time, we use MATLAB to carry out edge detection, image segmentation and other operations to extract the information we are interested in, so as to realize the objective collection and research of the fingerprint image of children, and avoid the subjective deviation in the diagnosis of the clinician.To provide some reference for clinical diagnosis and teaching.
【学位授予单位】：山东中医药大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：R272;TP391.41

【参考文献】