Image detection method and device, electronic equipment and medium

Abstract

The invention relates to an intelligent decision-making technology, and discloses an image detection method, which comprises the following steps: training an original object detection model by utilizing a training image sample set, testing a primary object detection model obtained by training the original object detection model by utilizing a test image sample set to obtain a detection result, and obtaining a misrecognition sample according to the detection result; obtaining a newly added training image sample set according to the misrecognition sample, the matting image sample set and the negative image sample set; retraining the primary object detection model by using the newly added training image sample set and the training image sample set, and detecting the to-be-detected image through a standard object detection model obtained by retraining to obtain an object detection result. In addition, the invention also relates to a block chain technology, and the object detection result can be stored in a node of a block chain. The invention further provides an image detection device, electronic equipment and a computer readable storage medium. According to the invention, the problem of low accuracy during image detection can be solved.

Description

technical field [0001] The invention relates to the technical field of intelligent decision-making, in particular to an image detection method, device, electronic equipment and medium. Background technique [0002] With the rapid development of artificial intelligence, the use of neural network models is becoming more and more extensive. When using neural network models to perform image detection (such as image detection) on objects, there will be "misjudgment" for some objects with similar characteristics. The situation is that the "non-target object" is misjudged as the "target object". In the prior art, the usual solution is to expand the sample set to allow the neural network model to extract more diverse object features, and the data of feature values ​​is more comprehensive and complete, thereby avoiding objects with similar features. However, in the existing technology, there is often no way to obtain a large number of training samples. Therefore, when the sample set...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides an image detection method, device and computer-readable storage medium, the main purpose of which is to solve the problem of low accuracy in image detection

Images