An image segmentation method and system based on a convolutional neural network

Abstract

The invention discloses an image segmentation method and system based on a convolutional neural network, and the method comprises the following steps of collecting a preset number of sample images, carrying out the normalization and data enhancement of the sample images, and obtaining the training sample data; training a U-Net convolutional neural network model with a residual block through the training sample data to obtain a trained U-Net convolutional neural network model; carrying out the pixel normalization processing on to-be-segmented images which are the same as the sample images; inputting the image to be segmented after pixel normalization processing into the trained U-Net convolutional neural network model, and finally obtaining a segmented image. The method provided by the invention has the relatively higher segmentation precision, and an end-to-end segmentation mode is adopted, so that the relatively higher segmentation efficiency is achieved.

Description

Technical field [0001] The invention belongs to the technical field of image segmentation, and particularly relates to an image segmentation method and system based on a convolutional neural network. Background technique [0002] At present, traditional image segmentation methods generally include: first, manually extract some graphics features, such as texture features and chroma, etc.; then, based on the above extracted features and then segment the image, its defects include: The ability to extract image features is limited. The traditional single image algorithm can only extract features from a single angle; the traditional algorithm based on threshold segmentation has a simple principle. Image segmentation is achieved by manually traversing and selecting the optimal threshold, but its calculation The process is complicated and easy to be interfered by noise, and the robustness is poor. Based on the edge detection algorithm, the edge points in the graph are first detected, an...

AI Technical Summary

Problems solved by technology

[0002] At present, traditional image segmentation methods generally include: first, artificially extract some graphic features, such as texture features and chromaticity; The ability to extract image features is limited, and the traditional single image algorithm can only extract features from a single angle; the traditional algorithm based on threshold segmentation is simple in principle, and the image segmentation is realized by manually traversing to select the best threshold, but its calculation The process is complicated, and it is easily disturbed by noise, and its robustness is poor; the algorithm based on edge detection first detects the edge points in the picture, and then connects them into contours according to a certain strategy to form a segmented area. The disadvantage lies in the contradiction between noise resistance and detection accuracy , so the obtained segmentation is often intermittent and incomplete structural information

Images