Image object detection method and device and computer readable storage medium

Abstract

The invention discloses an image object detection method and device, and a computer readable storage medium. The method comprises the steps of obtaining a sample image containing a detection object, wherein the sample image comprises position label information of the detection object; predicting a prediction object image area where the detection object is located in the sample image through an original detection model; obtaining prediction position distribution of the detection object in the prediction object image area; obtaining a position distribution difference between the prediction position distribution and the central position distribution, wherein the central position distribution is the position distribution of the detection object in the actual image area; and obtaining distribution difference loss according to the position distribution difference, and correcting model parameters of the original detection model based on the distribution difference loss, the prediction objectimage area where the detection object is located and the position label information so as to obtain a detection model used for identifying the position where the object is located in the image. According to the invention, the detection accuracy of the position of the object in the image can be improved.

Description

technical field [0001] The present application relates to the technical field of image processing, and in particular to an image object detection method, device and computer-readable storage medium based on artificial intelligence technology. Background technique [0002] With the continuous development of computer networks, image detection technology needs to be used in more and more life scenes. Among them, image detection can refer to the detection of specific objects in the image, so as to detect the specific objects in the image. The location of the object. [0003] For example, during the driving process of an unmanned car, the car can use the installed camera to capture the scene in front of the car's field of view, and obtain frame by frame images. The car can detect each captured frame of image, so as to detect the image area where the obstacle object in each frame of image is located, and then can calculate the position of the image area where the obstacle object ...

AI Technical Summary

Problems solved by technology

[0004] Among them, detecting the position of the image area where the obstacle object in the image is located is realized by the detection model. In the prior art, when using the detection model to detect the obstacle object in the image, if the background image in the image (that is, the image except for the obstacle The image other than the object) is more complex, the complex background image will cause greater interference to the detection process of the detection model for the obstacle object, resulting in the position of the image area where the obstacle object in the image finally detected by the detection model is not accurate. precise

Images