Irregular object semantic segmentation rapid labeling method

Abstract

The invention discloses an irregular object semantic segmentation rapid labeling method, which comprises the following steps of: 1) carrying out image data acquisition and preprocessing to obtain a data set of an image; 2) carrying out grid division on the image in the data set, and labeling a plurality of obtained small grids; 3) converting the label into a segmentation mask and carrying out edge optimization to obtain a data set for model training; and 4) selecting a segmentation model structure to carry out model training. Compared with pixel-level labeling, the irregular object semantic segmentation rapid labeling method has the advantages that the labeling time is greatly saved while the precision is slightly reduced; compared with rectangular frame labeling, the method can obviously improve the recognition precision, can greatly improve the labeling efficiency, and reduces the interference of background information on model training.

Description

technical field [0001] The invention relates to the technical field of semantic segmentation, in particular to a fast labeling method for semantic segmentation of irregular objects. Background technique [0002] Semantic segmentation refers to classifying each pixel in an image to identify the region where the object is located in the image. Usually, the training of semantic segmentation algorithms needs to be based on a large number of pixel-level mask labels manually labeled, and the labeling process is time-consuming and laborious. The segmentation method based on weak annotation usually first labels the bounding rectangle of the target, and then converts the rectangle into a mask for model training. Although this rectangular bounding box labeling method improves the labeling speed, the labeling box contains background information other than the target. Especially when the target is a non-rectangular irregular object, the background information in the labeling will serio...

AI Technical Summary

Problems solved by technology

Although this rectangular bounding box labeling method improves the labeling speed, the labeling box contains background information other than the target. Especially when the target is a non-rectangular irregular object, the background information in the labeling will seriously interfere with the model training.

Images