Improved HRnet based on attention mechanism

Abstract

An improved HRnet model based on an attention mechanism is characterized in that when an input F serves as an input feature map, an attention mechanism module is added, and the attention mechanism module is subjected to the following two operations: by adopting the technical scheme, the improved HRnet model based on the attention mechanism has the following beneficial effects that on the basis ofan original HRnet model, the attention mechanism model is added, so that the improved HRNet is used for human body posture detection in the cardio-pulmonary resuscitation pressing action process, an accurate backbone network is provided by the model for example segmentation models of the chest, the head and the like of a dummy in cardio-pulmonary resuscitation medical examination, and the detection precision of the model is improved.

Description

technical field [0001] The invention relates to an improved algorithm, in particular to an improved HRnet model based on attention mechanism. Background technique [0002] Sudden cardiac arrest is a serious threat to people's lives and health. Carrying out high-quality cardiopulmonary resuscitation (CPR, cardiopulmonary resuscitation) can significantly improve the survival rate of patients, and it is also an important means of saving patients' lives. The American Heart Association (AHA, American Heart Association) and the International Liaison Committee on Resuscitation (ILCOR, International Liaison Committee on Resuscitation) regard high-quality cardiopulmonary resuscitation as the core of resuscitation [1]. At present, the conventional cardiopulmonary resuscitation training and assessment methods are applied medical simulators and judged by referees. There are several disadvantages in doing so, for example, the examiner's judgment is highly subjective and not objective en...

AI Technical Summary

Problems solved by technology

There are several disadvantages in doing so, for example, the examiner's judgment is highly subjective and not objective enough; in the process of assessment and judgment, the specific compression depth and frequency of the candidate depend on the quality conditions of the simulator itself, which is difficult for the examiner to judge; during the training process, the trainee needs the examiner Always supervise and cooperate to correct and improve their own operations, which consumes a lot of manpower costs for training and assessment, etc.

[0003] In the prior art, after obtaining the candidate's pressing image, since the pressing action is a dynamic process, it is impossible to judge whether the candidate's pressing posture is qualified according to the pressing image, which brings difficulties to the automatic evaluation.

Images