Research On Information Acquisition Technology Of Glutamic Acid Crystallization Based On Image Analysis

In the process of chemical production and biopharmaceutical,as the chemical reaction continuous,the morphology of crystal will change constantly.To obtain the desired product,it is necessary to monitor the crystal crystallization process and obtain various growth information in the reaction process to control the reaction.With the development of science and technology,the emergence of hardware devices such as high-speed cameras have enable us to use image analysis to monitor the reaction process and extract various information.Image segmentation technology aims at separating the region of interest from the image background and removing various complex interferences in the background environment.It is a key step from image processing to image analysis.Texture analysis is a universal visual phenomenon,which can extract the feature parameters of the target by a certain image analysis technology,and then obtain the quantitative or qualitative analysis of the target.Feature extraction is an important step in the process of computer vision and image processing.It refers to the use of computer system to extract target information and determine whether each point in the image belongs to an image feature.The quality of feature has a crucial impact on the generalization performance.To effectively obtain various morphological information during crystal growth,an in-situ image analysis method based on invasive imaging system is proposed in this paper to monitor the morphologies of crystals so as to obtain various crystal information.The method includes image preprocessing,feature extraction and classification,agglomeration analysis.First,image preprocessing is carried out to eliminate the effects of water droplets,crystal shadows and uneven illumination.Secondly,the improved basic gray-level aura matrix is used to extract the texture features and two dimensionless image descriptors are used to extract the shape features.Then,the extracted features are used to identify different crystal morphologically.At last,the salient corners of the crystals are detected,and a segmentation algorithm for separating individual crystals from the agglomerates is constructed by clustering the same type of salient corners.Finally,the information in the process of crystal growth is counted.The case study and experimental results of L-glutamic acid cooling crystallization show that the image analysis method can be effectively used for morphology analysis of in-situ crystals with good precision.