Process for detecting events or elements in physical signal by implementing artificial neural network

• Main Authors: FOLLIOT LAURENT, DEMAJ PIERRE
• Format: Patent
• Language: Chinese; English
• Published: 28.04.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

According to one aspect, a method of detecting events or elements in a physical signal by implementing an artificial neural network is presented. The method includes assessing a probability of presence of an event or element through implementation of a neural network. Implementation of the neural network according to the nominal mode takes as input a physical signal having a first resolution, referred to as nominal resolution, when the probability of the event or element being present is greater than a threshold. Implementation of the neural network according to the low power mode takes as input a physical signal having a second resolution, referred to as reduced resolution, lower than the first resolution when the probability of the event or element being present is below a threshold. 根据一个方面，提出了一种通过实现人工神经网络来检测物理信号中的事件或元素的方法。该方法包括通过神经网络的实现来评估事件或元素存在的概率。当事件或元素存在的概率大于阈值时，根据标称模式的神经网络的实现将具有称为标称分辨率的第一分辨率的物理信号作为输入。当事件或元素存在的概率低于阈值时，根据低功率模式的神经网络的实现将具有称为降低分辨率的第二分辨率的物理信号作为输入，第二分辨率低于第一分辨率。