Image target detection method and device, terminal equipment and storage medium

Abstract

The invention discloses an image target detection method and device, terminal equipment and a computer readable storage medium, and the method comprises the steps: marking a label of a to-be-detectedtarget in a target sample image containing the to-be-detected target; constructing each training sample image according to the target sample image marked with the label and the scene sample image of each scene field, each scene sample image not containing the to-be-detected target; inputting each training sample image into a preset target detection model for model training, and detecting an imagetarget in the to-be-detected image by using the trained preset target detection model. According to the invention, there is no need to spend a lot of manpower resources to collect image data in different scene fields and mark image targets; training sample images for model training of the target detection model are enriched, false detection and missing detection of the target detection model during subsequent image target detection are avoided, and the image target detection efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a detection method, device, terminal equipment and computer-readable storage medium of an image target. Background technique [0002] The detection of image objects is a very important field in the development of computer vision. Different from the problem of image classification, the detection of image objects not only needs to judge whether there is an object in a picture, but also needs to locate the detected image objects. Generally speaking, the image data requirement for training the detection model for image targets is that the richer the scene where the image target is located, the better, and the more image data, the better. However, in real scenarios, the acquisition of image data is often not It is easy, and it faces the problem of heavy image data labeling. In this way, it leads to the lack of sample data required for training the target detection mod...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of the present invention is to provide an image target detection method, device, terminal equipment and computer-readable storage medium, aiming to solve the problem that the detection model cannot detect the image target due to the difficulty in obtaining a large amount of rich image data to train the detection model. , prone to technical problems such as false detection and missed detection

Images