CN-121980427-A - Dynamic event sensing method, system and equipment

Abstract

The application relates to the technical field of intelligent perception, and provides a dynamic event perception method, a system and equipment, wherein the dynamic event perception method comprises the steps of obtaining an electric response signal output by a synaptic device in a preset time period to obtain an electric signal sequence; the method comprises the steps of calculating the change rate of an electric signal sequence in a preset time period according to the electric signal sequence, obtaining a preset event judging threshold value, comparing the change rate with the event judging threshold value to obtain a comparison result, recording event information corresponding to an electric response signal if the comparison result is an effective dynamic event, and packaging the event information to generate an event signal based on the electric signal sequence, wherein the self-triggering perception of the dynamic event is realized by utilizing the characteristic that an artificial synapse device responds to the signal to evolve along with time under external stimulus, and the robustness of the system to complex environment change is improved.

Inventors

• CHEN DAN
• HU DUNAN
• LV JIANGUO

Assignees

• 季华实验室

Dates

Publication Date

20260505

Application Date

20260403

Claims (10)

1. 1. A dynamic event sensing method is characterized by comprising the following steps: Acquiring an electric response signal output by the synaptic device in a preset time period to obtain an electric signal sequence; Calculating the change rate of the electric signal sequence in the preset time period according to the electric signal sequence; acquiring a preset event discrimination threshold, and comparing the change rate with the event discrimination threshold to obtain a comparison result; If the comparison result is an effective dynamic event, recording event information corresponding to the electric response signal; based on the sequence of electrical signals, the event information is encapsulated to generate an event signal.
2. 2. The method for sensing dynamic events according to claim 1, wherein the step of calculating the rate of change of the electrical signal sequence within the predetermined time period based on the electrical signal sequence comprises: extracting current values of adjacent sampling points from the electric signal sequence and corresponding time intervals; calculating the current values of the adjacent sampling points to obtain current difference values of the adjacent sampling points; based on the current difference and the time interval of the adjacent sampling points, the instantaneous change rate is calculated.
3. 3. The method for sensing dynamic events according to claim 1, wherein the step of obtaining a predetermined event discrimination threshold and comparing the rate of change with the event discrimination threshold to obtain a comparison result comprises: Acquiring a preset event discrimination threshold, wherein the event discrimination threshold comprises a stable threshold and a mutation threshold; when the absolute value of the change rate is smaller than or equal to the stability threshold value, judging that the signal is a static background signal; when the absolute value of the change rate is larger than the stable threshold value and smaller than or equal to the abrupt change threshold value, judging that the target disturbance is non-target weak; and when the absolute value of the change rate is larger than the mutation threshold value, judging that the dynamic event is effective.
4. 4. The method for sensing dynamic event according to claim 1, wherein if the comparison result is an effective dynamic event, recording event information corresponding to the electrical response signal, comprises: if the comparison result is an effective dynamic event, the space coordinates of the synaptic devices of the electric response signals in the array are read; And acquiring the current moment as event occurrence time and correlating the event occurrence time with the space coordinates to obtain event information.
5. 5. The method for dynamic event awareness according to claim 1, wherein the encapsulating the event information based on the sequence of electrical signals to generate the event signal comprises: Extracting the peak intensity and duration corresponding to the event from the electric signal sequence; Integrating the peak intensity, the duration and the event information to obtain event characteristic parameters; And packaging the event characteristic parameters according to a preset protocol to generate event signals conforming to the preset protocol.
6. 6. The method for dynamic event awareness according to claim 3, wherein the encapsulating the event information based on the sequence of electrical signals to generate the event signal further comprises: acquiring historical response data in a preset time period; analyzing the environmental noise level based on the historical response data to obtain an analysis result; Optimizing and adjusting the event discrimination threshold according to the analysis result; and taking the event discrimination threshold after optimization and adjustment as a preset event discrimination threshold.
7. 7. The method for dynamic event awareness according to claim 6, wherein the optimizing the event discrimination threshold according to the analysis result comprises: When the analysis result is that the environmental noise floor is increased, the stability threshold value is improved; When the analysis result indicates that the non-target weak disturbance frequently occurs, the mutation threshold is increased; And when the analysis result shows that the detection sensitivity of the target dynamic event needs to be improved, reducing the mutation threshold value.
8. 8. The method for sensing dynamic events according to claim 1, wherein the step of obtaining the electrical response signal outputted by the synaptic device within the predetermined time period to obtain the electrical signal sequence comprises: Acquiring an electric response signal output by the synaptic device in a preset time period; amplifying and analog-to-digital converting the electrical response signal to obtain a converted signal; And sampling the converted signal at a preset sampling rate to obtain a digitized electrical signal sequence.
9. 9. A dynamic event awareness system, comprising: the acquisition module is used for acquiring an electric response signal output by the synaptic device in a preset time period so as to obtain an electric signal sequence; the calculation module is used for calculating the change rate of the electric signal sequence in the preset time period according to the electric signal sequence; The comparison module is used for acquiring a preset event discrimination threshold, and comparing the change rate with the event discrimination threshold to obtain a comparison result; the recording module is used for recording event information corresponding to the electric response signal if the comparison result is an effective dynamic event; And the generation module is used for encoding the event information to generate an event signal.
10. 10. The dynamic event sensing device is characterized by comprising a memory and at least one processor, wherein the memory stores instructions; at least one of the processors invokes the instructions in the memory to cause the dynamic event aware device to perform the steps of the dynamic event aware method of any of claims 1-8.

Description

Dynamic event sensing method, system and equipment Technical Field The application relates to the technical field of intelligent perception, in particular to a dynamic event perception method, a system and equipment. Background The existing dynamic behavior sensing and motion detection technologies mainly rely on continuous signal sampling or frame data acquisition modes, for example, a CCD/CMOS image sensor is used for processing continuous frame images to identify a target motion state by combining inter-frame difference, optical flow calculation or a video analysis algorithm based on deep learning. The technology generally needs high-frequency sampling and continuous calculation, and has the problems of high power consumption, large data redundancy, high system complexity and the like. In addition, part of non-imaging motion detection schemes rely on infrared, millimeter wave or ultrasonic sensors to judge whether a target moves through fixed threshold values or simple time difference, but the schemes are sensitive to environmental noise, are difficult to distinguish real dynamic events from background disturbance, and lack resolving power on complex dynamic behaviors. Disclosure of Invention The present application aims to improve at least one technical problem in the background art. The application provides a dynamic event sensing method, which comprises the following steps: Acquiring an electric response signal output by the synaptic device in a preset time period to obtain an electric signal sequence; Calculating the change rate of the electric signal sequence in the preset time period according to the electric signal sequence; acquiring a preset event discrimination threshold, and comparing the change rate with the event discrimination threshold to obtain a comparison result; If the comparison result is an effective dynamic event, recording event information corresponding to the electric response signal; based on the sequence of electrical signals, the event information is encapsulated to generate an event signal. According to some embodiments of the present application, the calculating the rate of change of the electrical signal sequence within the preset time period according to the electrical signal sequence specifically includes: extracting current values of adjacent sampling points from the electric signal sequence and corresponding time intervals; calculating the current values of the adjacent sampling points to obtain current difference values of the adjacent sampling points; based on the current difference and the time interval of the adjacent sampling points, the instantaneous change rate is calculated. According to some technical schemes of the application, the method for obtaining the preset event discrimination threshold value, comparing the change rate with the event discrimination threshold value to obtain a comparison result, specifically comprises: Acquiring a preset event discrimination threshold, wherein the event discrimination threshold comprises a stable threshold and a mutation threshold; when the absolute value of the change rate is smaller than or equal to the stability threshold value, judging that the signal is a static background signal; when the absolute value of the change rate is larger than the stable threshold value and smaller than or equal to the abrupt change threshold value, judging that the target disturbance is non-target weak; And when the absolute value of the change rate is larger than the mutation threshold value, judging that the dynamic event is effective. According to some technical schemes of the application, if the comparison result is an effective dynamic event, recording event information corresponding to the electric response signal, specifically including: if the comparison result is an effective dynamic event, the space coordinates of the synaptic devices of the electric response signals in the array are read; And acquiring the current moment as event occurrence time and correlating the event occurrence time with the space coordinates to obtain event information. According to some embodiments of the present application, the packaging of event information based on the electrical signal sequence to generate an event signal specifically includes: Extracting the peak intensity and duration corresponding to the event from the electric signal sequence; Integrating the peak intensity, the duration and the event information to obtain event characteristic parameters; And packaging the event characteristic parameters according to a preset protocol to generate event signals conforming to the preset protocol. According to some aspects of the present application, the packaging the event information based on the electrical signal sequence to generate the event signal further includes: acquiring historical response data in a preset time period; analyzing the environmental noise level based on the historical response data to obtain an analysis result; Optimizing a