Research On The Method Of Laboratory Sheet Recognition Based On OCR Technology

With the development of medical Internet,health data from human health is growing rapidly.Health data is reflected in the laboratory sheet data after treatment.In China's medical industry,the difference between the number of doctors and patients lead to the barriers when interpreting laboratory sheet.Efficient and accurate interpretation of laboratory tests,the management of personal health data is the current health care industry challenges.In order to solve the problems in the application of OCR technology,this paper puts forward the basic flow method of laboratory sheet recognition based on OCR technology,including preprocessing,pattern recognition,content recognition and recognition result correction.The concrete work is as follows.First of all,the paper preprocesses the single image,including binarization and anti-skew.The global threshold method,the adaptive threshold method and the OTSU method are studied.The OTSU method is used as the basic method of binarization of the single image by comparing the different methods.Then,through the pattern recognition technique,the straight line detection method based on Hough transform is used to extract the single image feature,and the single image is classified according to the linear feature classification.Then use Tesseract open source engine,training data,identify the contents of the test book,parameter tuning.At the end of the paper,the error correction scheme is combined with the editing distance and the laboratory medical lexicon,and the final error correction option is determined by comparing the similarity of Chinese characters.The validity of the error correction scheme is verified by the comparison experiment before and after the error correction.