Detecting system and detecting method for movement trajectory

1 . A detecting system for movement trajectory, comprising:

a display device, displaying a graphical user interface;

a radar, transmitting a radar signal to a detection area to receive a corresponding reflected signal; and

a processor, coupled to the radar and the display device and configured to:

obtain a point cloud comprising a plurality of sampling points derived from the reflected signal, wherein each of the plurality of sampling points has an associated timestamp;

group the point cloud to obtain a plurality of groups, comprising:

grouping a plurality of sampling points having timestamps within a first time period from the point cloud to obtain a first group among the plurality of groups,

generate a plurality of bounding boxes respectively corresponding to the plurality of groups, wherein the plurality of bounding boxes comprise a first bounding box corresponding to the first group;

generate a movement trajectory based on the plurality of bounding boxes;

calculate a difference between a timestamp of an earliest sampled sampling point and a timestamp of the latest sampled sampling point in the first bounding box to determine a dwell time of a subject at a location of the first bounding box;

determine a color of the first bounding box according to the dwell time; and

display information associated with the movement trajectory through the graphical user interface,

wherein the information comprises the color of the first bounding box.

2 . The detecting system of claim 1 , wherein the processor is further configured to:

obtain a plurality of centers respectively corresponding to the plurality of bounding boxes; and

generate the movement trajectory according to the plurality of centers.

3 . The detecting system of claim 2 , wherein the plurality of groups further comprises a second group, and the processor is further configured to:

obtain a plurality of sampling points during a second time period from the point cloud to obtain the second group, wherein the second time period is different from the first time period.

4 . The detecting system of claim 3 , wherein the first time period and the second time period partially overlap.

5 . The detecting system of claim 3 , wherein the processor is further configured to:

determine the color of the first bounding box according to a distance between the first bounding box and a reference location.

6 . The detecting system of claim 1 , wherein the information further comprises at least one of the following: a displacement, an instantaneous velocity, an average velocity, an instantaneous acceleration, an average acceleration, a movement offset, and a dwell time.

7 . The detecting system of claim 1 , wherein the processor is further configured to:

obtain environmental information of the detection area; and

filter a sampling point in the point cloud according to the environmental information to update the point cloud.

8 . The detecting system of claim 1 , wherein the processor is further configured to:

obtain a first longitudinal displacement signal according to the point cloud;

perform a detrending operation, a normalization, and a smoothing on the first longitudinal displacement signal to generate a second longitudinal displacement signal;

perform a peak detection on the second longitudinal displacement signal to obtain a plurality of peaks of the second longitudinal displacement signal; and

determine a number of times of standing-sitting actions according to the plurality of peaks, wherein the information comprises the number of times of the standing-sitting actions.

9 . The detecting system of claim 1 , wherein the processor is further configured to:

obtain a first longitudinal velocity signal according to the point cloud;

perform a detrending operation, a normalization, and a smoothing on the first longitudinal velocity signal to generate a second longitudinal velocity signal;

perform a peak detection on the second longitudinal velocity signal to obtain a plurality of peaks of the second longitudinal velocity signal; and

determine a number of times of standing-sitting actions according to the plurality of peaks, wherein the information comprises the number of times of the standing-sitting actions.

10 . A detecting method for movement trajectory, comprising:

transmitting a radar signal to a detection area to receive a corresponding reflected signal;

obtaining a point cloud comprising a plurality of sampling points derived from the reflected signal, wherein each of the plurality of sampling points has an associated timestamp;

grouping the point cloud to obtain a plurality of groups, comprising:

grouping a plurality of sampling points having timestamps within a first time period from the point cloud to obtain a first group among the plurality of groups,

generating a plurality of bounding boxes respectively corresponding to the plurality of groups, wherein the plurality of bounding boxes comprise a first bounding box corresponding to the first group;

generating a movement trajectory according to the plurality of bounding boxes;

calculating a difference between a timestamp of an earliest sampled sampling point and a timestamp of the latest sampled sampling point in the first bounding box to determine a dwell time of a subject at a location of the first bounding box;

determining a color of the first bounding box according to the dwell time; and

displaying information associated with the movement trajectory through a graphical user interface,

wherein the information comprises the color of the first bounding box.

11 . The detecting method of claim 10 , wherein the step of generating the movement trajectory according to the point cloud comprises:

obtaining a plurality of centers respectively corresponding to the plurality of bounding boxes; and

generating the movement trajectory according to the plurality of centers.

12 . The detecting method of claim 11 , wherein the plurality of groups further comprises a second group, and the step of grouping the point cloud to obtain the plurality of groups comprises:

obtaining a plurality of sampling points during a second time period from the point cloud to obtain the second group, wherein the second time period is different from the first time period.

13 . The detecting method of claim 12 , wherein the first time period and the second time period partially overlap.

14 . The detecting method of claim 12 , wherein the step of displaying the information associated with the movement trajectory through the graphical user interface comprises:

determining the color of the first bounding box according to a distance between the first bounding box and a reference location.

15 . The detecting method of claim 10 , wherein the information further comprises at least one of the following: a displacement, an instantaneous velocity, an average velocity, an instantaneous acceleration, an average acceleration, a movement offset, and a dwell time.

16 . The detecting method of claim 10 , further comprising:

obtaining environmental information of the detection area; and

filtering a sampling point in the point cloud according to the environmental information to update the point cloud.

17 . The detecting method of claim 10 , further comprising:

obtaining a first longitudinal displacement signal according to the point cloud;

performing a detrending operation, a normalization, and a smoothing on the first longitudinal displacement signal to generate a second longitudinal displacement signal;

performing a peak detection on the second longitudinal displacement signal to obtain a plurality of peaks of the second longitudinal displacement signal; and

determining a number of times of standing-sitting actions according to the plurality of peaks, wherein the information comprises the number of times of the standing-sitting actions.

18 . The detecting method of claim 10 , further comprising:

obtaining a first longitudinal velocity signal according to the point cloud;

performing a detrending operation, a normalization, and a smoothing on the first longitudinal velocity signal to generate a second longitudinal velocity signal;

performing a peak detection on the second longitudinal velocity signal to obtain a plurality of peaks of the second longitudinal velocity signal; and

determining a number of times of standing-sitting actions according to the plurality of peaks, wherein the information comprises the number of times of the standing-sitting actions.