Target detection model training method, classroom behavior detection method and related equipment

Abstract

The invention relates to the technical field of artificial intelligence, and provides a training method of a target detection model, a classroom behavior detection method and related equipment. The training method of the target detection model comprises the following steps: constructing an initial network model based on a central network Center Net; processing a sample image through the initial network model to obtain the category and the center point position of each target object in the sample image, and performing single-dimensional regression according to the center point position to obtain detection frame information of each target object; controlling the training of the initial network model by taking a multi-task loss function including classification loss, detection frame offset loss and center point offset loss as a constraint condition according to a sample image set including a plurality of sample images, and obtaining a target detection model. Based on the Center Net, by improving the regression process and the loss function, the training efficiency of the target detection model can be effectively improved, and the target detection precision is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a training method of a target detection model, a classroom behavior detection method and related equipment. Background technique [0002] Behavior detection based on target detection is an important application of artificial intelligence technology. With the popularization of information-based education, the use of artificial intelligence technology to detect classroom behaviors can be used to analyze the degree of student participation in the classroom, which is conducive to teaching evaluation and improving teaching quality. [0003] Existing target detection algorithms are divided into two categories based on traditional machine learning and based on deep learning. Taking the detection of hand-raising behavior in class as an example, the target detection algorithm based on traditional machine learning needs to use the designed feature template to extract hand-...

AI Technical Summary

Problems solved by technology

The human body key point detection algorithm first uses the human body key point detection algorithm to extract the human skeleton features, and then uses the classification model to judge whether it belongs to the behavior of raising hands; because the human body key point detection algorithm is easily affected by occlusion, and the occlusion phenomenon is more frequent in classroom teaching scenes , so the performance of the algorithm for human key point detection is seriously insufficient

[0005] The two-stage target detection algorithm can solve the challenges of different hand gestures in classroom teaching scenes, but the steps are cumbersome, the training is time-consuming, and the accuracy is not enough, so it cannot be widely used

Images