Image classification method and device and equipment

Abstract

The invention relates to an image classification method and device and equipment. The method comprises the following steps: inputting a to-be-classified image data set; wherein the image data set comprises to-be-classified images; taking the center of the to-be-classified image as a reference center point, and intercepting an image which comprises a specific area and has a preset size as a standard image; selecting a pixel value array of any single channel in the standard image, and drawing a corresponding signal oscillogram based on the pixel value array; and determining the category of the to-be-classified image based on the signal oscillogram. The method comprises the following steps of: intercepting a specific area of an image to be classified, selecting a pixel value array of any single channel from a standard image obtained by intercepting, and then drawing a signal oscillogram according to the selected pixel value array of the single channel, so that when the image to be classified is classified, the image to be classified can be classified according to the selected pixel value array of the single channel. And division is carried out based on the drawn signal oscillogram, so that various images can be accurately and efficiently distinguished, and the method can be suitable for various image classification application scenes.

Description

technical field [0001] The present application relates to the technical field of image classification and recognition, and in particular to an image classification method, device and equipment. Background technique [0002] Usually, the storage and transmission of medical images, the doctor's diagnosis, and the tracking of the course of disease are all carried out with the patient as the dimension. The mixing of complex and different types of images is bound to cause great trouble for the subsequent sorting, archiving and research of different types of images. In related technologies, an image recognition model or a currently popular deep learning network model CNN can be used to implement image classification. However, when using an image recognition model for image classification, it is necessary to build a recognition model in advance. In the process of building a recognition model, several stages such as feature extraction, feature encoding, space constraints, classifi...

AI Technical Summary

Problems solved by technology

Using the deep learning network model CNN to realize image classification, although it can liberate manpower and realize fast and efficient automatic sorting, it takes a lot of time to build the neural network model, label the training data set, train the model parameters, tune, etc. in the early stage. The required technical threshold and hardware configuration are high, which makes the reproducibility and generalization ability of the traditional image classification method weak, and it is difficult to take it out of the box, quickly replicate it, and extend it to various complex and changeable clinical situations. in medical scene

Images