计算机图象识别辅助菌种筛选及对甾体激素微生物酶反应的研究  

Studies on Seed Selection and Kinetic Model of Microorganism Enzymatic Steroid Transformation Using Computer Image Texture Analysis Technique

作　　者：王积分[1] 阳葵[1] 刘宏[1] 安中 段世铎[1] 

机构地区：[1]天津大学计算机系,天津大学化学系

出　　处：《高等学校化学学报》1994年第5期750-753,共4页Chemical Journal of Chinese Universities

基　　金：国家自然科学基金

摘　　要：将分形理论和计算机图象纹理分析技术移植到微生物发酵研究中，对菌落和菌丝体形态的定量描述进行了探讨。并利用菌种的分维特征作为筛选高产菌株的直观指标，有效地提高了菌种筛选工作的效率与准确度。探讨了沃氏底物霉菌转化的工艺过程，并以生物全息观念与分形理论为依托，在分形尺度上对生物体形态及其发育过程进行了描述，建立起菌体生长的形态动力学模型，并与生长（浓度）动力学模型相结合，设计出可判别理论最佳投料点的酶活描述模型。In this paper we employ Fractal Theory Organisms Full Information Law and Computer Texture Analysis Technique, and systematically investigate the characterization of morphologies of colonies and mycelium during the microorganism fermentation process. We demonstrate that using the fractal dimension characteristics to assist the mould seed selection can significantly improve the efficiency and accuracy. Moreover,we show that the morphology kinetic model and enzymatic activity model developed by us are capable of describing the microorganisms growing and enzymatic transformation process with a good degree of accuracy. When applied to predict the proper time of putting the substrates into the reaction system, the result is found to be in good agreement with the experimental data.

关 键 词：计算机 微生物酶反应 菌种筛选 

分 类 号：Q93－3[生物学—微生物学]