数码舌象图像撷取的变因控制及其在脾胃虚实病证的运用观察
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
本研究主要目的是透过客观化控制数码舌象图像生成的影响变因,以获得能够真实还原被观察的舌象的图像,并与中医的“证”相联系。
     证是中医的一个特殊概念,简单来说包含了疾病的特质与机体对应当下所处的状态,也是对疾病发展过程中病位、病性的概括,是中医师处方用药的指导原则,可以说对证认识的深浅、判断的正确与否,就决定了中医的治疗效果。然而,长久以来,中医学的传承靠的是文字抽象的描述,能够有深刻体会到其中真意的不知几希!
     舌诊是中医辨证理论体系里的一种特色方法,是反映人体状态的重要观察指标之一。而舌象的客观化研究一直以来方兴未艾,近代的研究中也随着不同阶段的科学的发展而有所进展;而随着影像数字化(基于普及用语以下称为数码)的进程加速,因其带来的简便性、可操性,数码图像在舌象客观化的研究的应用中也越来越受到重视并渐渐占有重要的地位,过往很多难以分析的难题,在数码化的领域中,前人们慢慢摸索出一条应用之道。
     但是以往的研究对于影像生成的过程,也就是影像反射的光线经过镜头光学作用投射到感光组件经过一连串信号处理成数码图像文件格式输出的这个阶段对影像的影响,却鲜少论及。所以,如何在控制这些变因的基础上建立一个较客观的方法,并且可以使撷取得到的图像能够持续的利用,是本研究的重点工作。本文从以下几方面进行了研究。
     一、文献研究
     长期以来舌象客观化的研究先贤们一直通过各种方法将舌象客观化,在各领域也取得了一定的成绩;特别是近10多年,由于数码科技的兴起,因为其对于各种变因的掌控比以往简便以及操作更加容易等特性,逐渐在此领域取得了长远的进展。目前舌象客观化的研究涉及标准观察模式、色彩空间、舌象自动识别软件等多方面,研究范围较大,且涉及多学科、多专业,对舌象客观化的研究开展了较全面的探讨。
     二、舌象图像撷取过程的变因
     数码舌象图像撷取过程除了受到环境中光线或照明体色温的影响之外,在数码影像生成的过程中还受各种因素的影响,本文将图像生成的步骤分成两大部分详细探讨了会影响图像生成的各种变因。
     (一)舌象撷取过程变因分析
     拍摄环境中的变因主要影响的是光源及其色温,针对此一方面,本文介绍并沿用国际照明委员会(Commission International de l'Edairage,筒称CIE)建议的标准照明以及观测条件。
     作为撷取舌象图像的主要工具,相机功能影响图像是最关键的因素,因此有必要深入了解数码相机的分类、结构及其功能,还有对影像会产生偏差的的镜头。此外,由于本文使用的是数码图像,也探讨了对于不同的存储格式会舌象图像产生的影响
     (二)图像处理过程变因分析
     舌象图像后期处理主要是受到软件的影响,而软件对影像色彩表现所依附的最重要的即在色彩管理,色彩管理的重点则在色彩空间的选择,目前主要使用的色彩空间计有RGB、CMY、1931 CIE XYZ、CIE Lab、HSB、sRGB等色彩空间。
     图像的显示必须经过输入(相机、影像编辑软件)、输出(显示器、打印设备、投影仪器等)等设备,不同设备使用的色彩空间会影响色彩的一致性,色彩校正目的在使操作环境统一在同一个色彩空间,目的在于修正、减少这些原本存在环境、相机中的影响因素所带来对舌象图像真实性的干扰,以期使舌象图像能够更加客观化。
     三、数码舌象图像成像变因控制的选择
     (一)环境光源的选择
     1.D65、D55标准光源
     由前述可知环境光源会干扰舌色在色度学中的表现,也会影响人类视觉对颜色的辨识,因此本研究根据CIE的建议比较了D65、D55两种标准照明体,并分别进行照度测试、照明范围测定以及显色指数的比较。D65标准光源:采用飞利浦(PHILIPS)电子式螺旋型紧凑萤光灯装置在嘉美灯饰电器厂出产的嵌入式顶灯,并在灯具外缘加装遮光板;D55标准光源:采用冠军牌摄影器材厂出产的专业摄影光导纤维标准冷光灯(以下简称冠军牌冷光灯)作为光源。
     2.结果
     冠军牌专业摄影光导纤维标准冷光灯在相机快门速度、Ra、患者感觉舒适度三方面均优于飞利浦电子式螺旋型紧凑萤光灯,因此本课题选用前者作为拍摄数码舌象图像用途的标准照明体。
     (二)同厂牌不同相机白平衡与色彩还原的研究
     1.方法
     使用两种相机按照自动白平衡、手动设置白平衡的步骤分别拍摄一张在墙上的白纸以及GretagMacbeth24色校色板,Nikon Coolpix 4300直接使用所得的JPG图档,Nikon D70则先按照上述步骤拍摄,再以PHOTOSHOP影像编辑软件调校RAW格式,最后再比较拍摄结果。
     2.结果
     Nikon Coolpix 4300、Nikon D70在自动、手动白平衡在白平衡测试与Gretag-Macbeth24色校色板测试后,对照实际白色以及24色校色板均存在肉眼可分辨的色差,而Nikon D70经PHOTOSHOP调校后基本接近。
     (三)讨论
     根据结果提示,单纯依靠相机的白平衡设置,所形成的影像均存在着色差,而RAW格式文件PHOTOSHOP软件调校后的影像则相对能够满足此条件。
     四、临床研究
     (一)方法
     本研究选择中医辨病为胃痞、胃痛病,辩证为脾胃湿热证患者30例、脾气虚证患者28例以及肝胃不和证患者患者27例作为观察对象,于专门设立的暗室使用光纤冷光灯作为光源,拍摄前设定白平衡,以柯达灰卡(Kodak Graycard)做曝光矫正;要求患者当日早餐至拍摄前不要进食或至少拍摄前2hr不要进食,相机保持在患者法线上,并自订拍摄标准区域,拍摄后填写症状、体征调查表;所得图像后期以PhotoShop CS处理再在舌象图像上进行色度学分析,以比较三组不同证型患者舌象在CMY、HSB、Lab色彩空间的变化。
     (二)结果
     1.舌质部分
     舌质红色的表现由深至浅依次分别为脾胃湿热组、肝胃不和组、脾气虚组,说明在对舌质的观察中肉眼与色度学的分析结果基本相同。
     2.舌苔部分
     舌苔的表现脾胃湿热组要较肝胃不和组与脾气虚组朝黄色的向量偏移,而肝胃不和组与脾气虚组则无统计学差异,说明在对舌苔的观察中肉眼与色度学的分析结果基本相同。
     五、结论
     根据临床研究的结果显示本论文的图像生成干预方法能够基本还原真实状态的舌象,对于应用在临床治疗疗效观察中舌象变化的指标、教学应用等方面具备可行性。
The research is aimed at objectively controlling the influencing factors of digital tongue image to obtain the real image of the observed tongue and connect to TCM syndrome.
     Syndrome is a specific concept in TCM,which briefly includes the characteristic of a disease and corresponding state of the organism.It also concludes the location and nature of a disease in the course of its development and gives instruction to tcm doctor in prescribing medicine.The depth of recognition and the correctness of judgment towards syndrome decide the therapeutic effect of TCM.However,the inheritance of TCM relies on the abstract description of Chinese character,the profound meaning is only known by few people.
     Tongue diagnosis is a distinguishing feature in Syndrome differentiation system of TCM,and tongue is one of the significant observing indicators of one' s health state.The objectivising research of tongue never stops and has made progress in modern study with the development of science and technology through these years.Digital image technology(the universal phraseology "digital" is used in following paragraphs) plays a more and more important role in the application of tongue image objectifivising researches. Predecessors have gradually found an application method to solve many tough problems in digital area.
     The previous researches focus on the process of producing an image that is to output the image ESP processed from a series of signals by reflecting lights through camera lens.But the influence to the image itself is seldom mentioned.Therefore,how to establish a comparatively objective method on the basis of controlling these changeable values as well as obtain images for sustainable use is the key point of this research.The research contains aspects as follows.
     1.The Literature Research
     For a long time the predecessors have researched into tongue objectification in various ways in many fields and also gained certain achievements.Especially in the recent decade,with the rising up of the digital technology,the control of various changeable values as well as the operation in tongue image objectification becomes more convenient.Tongue objectification has gradually made long-term progress in this field.At present tongue image objectification research involves various fields,such as the standard observation pattern,color space,tongue image automatic diagnosis software.The research scope is wide and involves many subjects and specialties.The discussion of tongue image objectification research is comprehensive.
     2.The Changeable Values During the Digital Tongue Image Catching Process
     The light or the color temperature in the environment influences the digital tongue image catching process,besides,there are other factors which influences this process.This essay specifically discusses all kinds of changeable values in two divisions according to image catching process.
     2.1 The analysis of Changeable Values During the Digital Tongue Image Catching Process
     The main influence of shooting environment is light source and color temperature,so this essay introduces and uses the Commission International de l'Edairage(CIE) standard illumination and observation condition.
     As the main tool of catching image,a camera' s function is the key factor Of influencing image,so it is necessary to know the categorization,structure and function of a camera as well as the camera lens.Besides,because it is digital image,so the storage pattern,which also has some effect on,the tongue image will be discussed either.
     2.2 The analysis of Changeable Values in image Process
     The later stage of tongue image process is mainly influenced by software; however,the display of the image color mostly depends on color management in the software.Color management emphasizes the choice of color space and the present color space includes:RGB、CMY、1931 CIE XYZ、CIE Lab、HSB、sRGB etc.The display of the image should go through input equipment(camera,image editing software) and output equipment(monitor,print machine,projection machine,etc).For different color space used in different equipment will influence the consistency of color,so color rectification aims at unifying the same color space to rectify and reduce the interference of validity of the tongue image by those originally existing influencing factor in the environment and the camera itself.
     3.The choice of Changeable Values control in digital tongue image
     3.1 The choice of the environment light source
     3.1.1 D65、D55 standard light source
     As it is said above the light in environment can interfere the display of tongue color in eolourometry as well as the color differentiation of human vision.Therefore,according to the CIE this research compares D65、D55 in luminance test,illuminating scope test and chromomeric index D65 standard light source:use PHILIPS electronic spiral-styled compacted fluorescent light set which inserts the top light by Jiamei light decoration and electrical equipment factory and add an occluder on the edge of the light.D55 standard light source:use the professional optical fiber standard luminescence light, which is produced,by Guanjun photographic accessory factory as light source. (use the name of Guanjun uminescence light in following paragraphs )
     3.1.2 Result
     Guanjun uminescence light is prior to PHILIPS electronic spiral-styled compacted fluorescent light in shutter speed,Ra and patient' s degree of comfort.Therefore,the former one is chosen as the standard light source in this research.
     3.2 The White balance and color reducing research of different cameras within the same brand
     3.2.1 Method
     Using two cameras to respectively picture one piece of paper and GretagMacbeth24 color checker disk on the wall according to the step of auto white balance and hand-motion white balance,then directly use the JPG file by Nikon Coolpix 4300.When comes to Nikon D70,first picture according to the above steps and then edit images with PHOTOSHOP in RAW form,in the end compare the above two.
     3.2.2 Result
     After white balance test of both auto and hand-motion and GretagMacbetb24 colored checker disk test Nikon Coolpix 4300、Nikon D70,compare real white color and 24-colored color- rectifying disk,obvious chromatism is exist only by naked eye while in Nikon D70 there' s almost no chromatism after adjusted by PHOTOSHOP.
     3.3 Discussion
     According to the results,there' scbromatism in the images only set WHITE BALANCE in the camera while images adjusted by PHOTOSHOP with RAW form can relatively satisfy this condition.
     4.Clinic research
     4.1 Method
     In this research,we observe 30 patients diagnosed as stomach depress and stomachache with damp-heat in spleen and stomach syndrome,28 patients with spleen deficiency syndrome and 27 patient with in coordination between the liver and stomach syndrome.Choose optical fiber luminescence light as light source in specially made dark room,set WHITE BALANCE with Kodak Graycard as exposure rectification.Ask the patients not to eat anything before taking picture or at least 2hrs before taking picture that morning.The camera should stay on patient' s normal line and set the standard picturing area by themselves,fill the blanks in the form of syndrome and physical sign.Do colourometry analysis toward the later stage tongue images with PhotoShop CS and compare the changes of color space of CMY、HSB、Lab in patients' tongue images among three groups.
     4.2 Result
     4.2.1 Body of the tongue
     The redness of the body of the tongue from dark to light are:damp-heat in spleen and stomach group,incoordination between the liver and stomach group, spleen deficiency group.It shows that it is basically the same between colourometry analysis and naked-eye observation on the observation of the body of the tongue.
     4.2.2 The coating of the tongue
     The coating of the tongue shows that the damp-heat in spleen and stomach group appears offset to yellow vector compared to the other two groups and there' s no statistic difference between incoordination between the liver and stomach group and spleen deficiency group,it shows that it is basically the same between colourometry analysis and naked-eye observation on the observation of the coating of the tongue.
     5.Conclusion
     According to the results of clinical study the image-making intervention method in this essay can basically recover the real tongue state and it has feasibility both in observing index of tongue state changing in clinical therapeutic effect and teaching application.
引文
[1]清·陈萝雷等.古今图书集成医部全录(点校本)·第二册.北京:人民卫生出版社,2000年第一版:528-529
    [2]孙立友,程钊,谢虎臣,等.利用计算机图象识别技术进行舌诊客观化研究的探讨[J].安徽中医学院学报,1986,5(4):527.
    [2]李乃民.中国舌诊大全[M].北京:学苑出版社,1995.
    [4]侯浚,陈健民,罗东秀,等.4417例癌症患者舌象临床观察[J].浙江中医杂志,1992,27(8):368-369.
    [5]陈志峰,侯浚,王济民,等.1635例癌症病人红紫、淡青紫、青紫舌脉Ridit分析:附426例非癌症病人对照[J].河北中医,1992,14(5):324.
    [6]张伯礼,王新,姚远,等.中医舌诊检测仪器研制及应用[J].天津中医,1992,9(8):30231
    [7]徐顺潮,李月男,余永敏,等.光谱光度法舌色仪[J].中国医疗器械杂志,1993,17(3):1292132.
    [8]学村八一郎.用彩差计作舌色的客观表示[J].国外医学·中医中药分册,1989,11(3):425.
    [9]冈本芳幸.舌诊客观化的研究:应用显微镜观察舌表面形态[J].国外医学·中医中药分册,1995,17(3):52253.
    [10]冈本芳幸.舌诊客观化的研究:应用显微镜观察舌表面形态及与东洋医学各项问诊的相关性[J].国外医学·中医中药分册,1996,18(3):62.
    [11]中田熏.高、初中及小学生与成人的诊断和舌诊的关系[J].国外医学·中医中药分册,1997,19(6):27.
    [12]余兴龙,谭耀麟,竺子民,等.中医舌诊自动识别方法的研究[J].中国生物医学工程学报,1994,13(4):336-344.
    [13]余兴龙,竺子民,金国藩,等.中医舌诊自动识别系统的研究[J].华中理工大学学报,1995,23(8):55-59
    [14]翁维良,黄世敬.中医舌诊客观化研究.中国工程科学,2001,3(1):78-82,83
    [15]蒋依吾,陈建仲,张恒鸿,等.电脑化中医舌诊系统[J].中国中西医结合杂 志,2000,20(2):145-147.
    [16]张志枫,顾超,王志国,等,WZX舌色分析系统在胃病患者舌像信息处理中的应用.上海中医药大学学报,2004,18(3):52-56.
    [17]沈兰荪,赵忠旭,王爱民等.基于图像处理的中医舌像分析仪的研究.国外电子测量技术.1999,6:33-35
    [18]卫保国,沈兰荪,王艳清,等.数字化中医舌象分析仪[J].中国医疗器械杂志,2002,26(3):164-166.
    [19]沈兰荪,蔡轶珩,卫保国等.中医舌象分析技术的研究.世界科学技术-中医药现代化,2003,5(1):15-19
    [20]赵荣莱,许胜,危北海,等.舌质舌苔的计算机定量描述和分类[J].中医杂志,1989,(2):47-48.
    [21]朱洁华,阮邦志,励俊雄,等.舌诊客观化研究的一种图象处理方法[J].中国生物医学工程学报,2001,20(2):132-137.
    [22]周越,沉利,杨杰等.基于图像处理的中医舌象特征分析方法.红外与激光工程,2002,31(6):490-494
    [23]孙炀,罗瑜,周昌乐等.一种基于分裂-合并方法的中医舌像区域分割算法及其实现.钟国图象图形学报,2003,8(12):1395-1400
    [24]许家佗,孙炀,张志枫等.基于差分统计方法的舌象纹理特征的分析与识别.上海中医药大学学报,2003,17(3):55-58
    [25]许家佗,周昌乐,方肇勤等,舌像颜色特征的计算机分析与识别研究.上海中医药大学学报,2004,18(3):43-47
    [26]赵忠旭,王爱民,沈兰荪.舌像分析仪中彩色校正的研究.电子测量与仪器学报,1999,13(3):15
    [27]王爱民,沈兰荪,赵忠旭.监督FCM聚类算法及其在中医舌象自动分类中的应用[J].模式识别与人工智能,1999,12(4):4802485.
    [28]王永刚,王爱民,沈兰荪.舌象分析仪中舌色重现方法的研究.照明工程学报,2001,12(2):4 10
    [29]王爱民,赵忠旭,沈兰荪.中医舌象自动分析中舌色、苔色分类方法的研究.2000,19(3):136142
    [30]赵忠旭,王爱民,沈兰荪.基于数学形态学和HIS模型的彩色舌图像分割.北京工 业大学学报.1999,25(2):67 71
    [31]沈兰荪,王爱民,卫保国等.图像分析技术在舌诊客观化中的应用.2001,12A:17621765
    [32]卫保国,沉兰荪,王艳清等.数字化中医舌象分析仪.中国医疗器械杂志.2002,26(3):164 169
    [33]孟智宏,刘大功,梁兵,等.43例再生障碍性贫血舌色测定研究[J].天津中医,1992,9(4):41242.
    [34]韩新民,余永敏,孙天庭,等.200例人体舌象的定量化检测分析[J].中国中西医结合杂志,1993,13(2):110-111.
    [35]高秀梅,张伯礼,徐宗佩等.急性心肌梗塞的特殊舌象[J].中医杂志,1994,35(6):365.
    [36]张永涛,梁嵘,王召平等.884例体检人群舌色数字图像应用不同颜色模型的比较.中国中医基础医学杂志,2005,11(3):207-209
    [37]张永涛,梁嵘,王召平等.884例体检人群数字舌图的舌色特征探讨.北京中医药大学学报,2005,28(1):73-75
    [38]何国兴.颜色科学,上海:东华大学出版社,2004,第一版:61-118
    [39]薛朝华.颜色科学与计算器测色配色实用技术,北京:化学工业出版社,2004,第一版:41-44
    [40]黄爱民,安向京,骆力等.数字图像处理与分析基础,北京:中国水利水电出版社,2005,第一版:15-27
    [41]薛朝华.颜色科学与计算器测色配色实用技术,北京:化学工业出版社,2004,第一版:95-96
    [42]雅格布逊等着:杨词银等译.大不列颠摄影教程,长春:吉林摄影出版社,2002,第一版:69-180
    [43]何国兴.颜色科学,上海:东华大学出版社,2004,第一版:120-138
    [44]谢晓昱.色谱与色彩管理,南京:江苏科学技术出版社,2004,第一版:143-155
    [45]林福宗.多媒体技术基础,北京:清华大学出版社,2002,第二版第六章:13-32
    [46]劳绍贤,周正,林文良等.建立慢性浅表性胃炎脾胃湿热证诊断标准的探讨.广州中医药大学学报.2004,21(5):365-368
    [47]郑筱萸.中药新药临床研究指导原则(试行).北京:中国医药科技出版社,2002:361-368

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700