Automatic labeling method and device for industrial product image, electronic equipment and storage medium

Abstract

The invention provides an automatic labeling method and device for an industrial product image, electronic equipment and a storage medium. The method comprises the steps of obtaining a to-be-labeled image and classification information of an object; carrying out segmentation processing on the to-be-labeled image so as to obtain independent images of each component of the to-be-labeled object through segmentation; according to the classification information, recognizing the independent images of all the components of the to-be-labeled object, so that type recognition results of all the components are obtained; according to the type identification result, adding a corresponding label to each component in the to-be-labeled image to obtain a preliminary labeled image; displaying the preliminary annotation image to obtain modification information so as to generate a final annotation image; therefore, each part in the object formed by assembling a plurality of parts can be automatically labeled, the marking efficiency is improved, and the cost is reduced.

Description

technical field [0001] The invention relates to the technical field of image labeling, in particular to an automatic labeling method, device, electronic equipment and storage medium for industrial product images. Background technique [0002] At present, industrial robots are widely used in industrial production to grasp and transfer objects. Some objects are assembled from multiple parts, and when grasping, it is required to only grasp at certain parts, otherwise the object may be damaged. Therefore, it is necessary to train the industrial robot so that the industrial robot can recognize the position of the graspable object. Therefore, it is necessary to train industrial robots to recognize each part of an assembled object. [0003] In order to realize the training of industrial robots, it is necessary to label the object pictures as training pictures. When labeling, it is necessary to add labels to the target positions in the images. The labeling process is generally don...

AI Technical Summary

Problems solved by technology

[0003] In order to realize the training of industrial robots, it is necessary to label the object pictures as training pictures. When labeling, it is necessary to add labels to the target positions in the images. The labeling process is generally done manually, and the labor and time costs are huge.

Images