Method and device for extracting patient information of film image

Abstract

The invention relates to the field of artificial intelligence, in particular to a film image patient information extraction method and device. The method for extracting the patient information of the film image comprises the following steps: calling a pre-trained image area detection model to carry out image area detection on a to-be-identified film image, and identifying position information of one or more image areas; obtaining a subimage set according to the identified position information of each image area; selecting a subimage from the subimage set, performing color filling on an image area in the subimage according to a background color of the subimage, and generating a text information image only containing patient information and the background color; and performing text detection on the text information image to obtain patient information. The method has the advantages of high recognition effect and high universality.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a method and device for extracting patient information from film images. Background technique [0002] After taking the video film in the hospital, it often takes a while to get the inspection image of the film, and then take the film to the corresponding doctor to make a diagnosis. It is very inconvenient for many patients who go to the hospital to see a doctor. In recent years, a self-service film pickup machine has emerged that allows patients to self-print films at any time. The patient scans the barcode on the receipt voucher, and the film virtual printing server in the background verifies the patient information on the barcode, and automatically prints the film after the matching is successful. Self-service printing mainly uses OCR (Optical Character Recognition) technology to identify patient information in film images / images. [0003] At present, before OCR recogn...

AI Technical Summary

Problems solved by technology

Due to the incompatibility of film size, film typesetting layout and patient information layout, the OCR recognition rate is poor, which leads to the mismatch between the patient information recognized by OCR and the barcode information on the receipt voucher, resulting in the inability to print automatically, resulting in self-service printing Low effective utilization

In addition, in the process of OCR recognition, the traditional method needs to segment the characters from the film image, and then combine them into patient information after recognition. When the font size of the preset information of the film is reset, it will also affect the OCR recognition rate.

[0004] Aiming at the above-mentioned problems existing in extracting film user information based on OCR technology, the prior art does not provide an effective solution

Images