US-20260126542-A1 - METHOD AND A DEVICE FOR DETERMINING OCCUPIED POSITIONS IN A SPACE

US20260126542A1US 20260126542 A1US20260126542 A1US 20260126542A1US-20260126542-A1

Abstract

A method for determining occupied positions in a space starts by receiving detections of object movement from a radar monitoring the space. Each detection is associated with a position in the space and a time point when the detection was made. The method accumulates detections that have a time point within a first time period of predefined duration, and identifies a set of clusters of detections in the space by analyzing similarity in position of the accumulated detections. Any cluster whose detections are temporally distributed within a proportion of the first time period which is below a predefined proportion threshold are then removed from the set of clusters. The positions in the space that correspond to the set of clusters are determined to be occupied during the first time period.

Inventors

Christoffer Kjellson

Assignees

AXIS AB

Dates

Publication Date: 20260507
Application Date: 20251031
Priority Date: 20241107

Claims (14)

1 . A method for determining occupied positions in a space, comprising: receiving detections of object movement from a radar monitoring the space, each detection being associated with a position in the space and a time point when the detection was made, wherein the received detections include detections from a plurality of time points corresponding to a plurality of frames of radar data, accumulating detections that have a time point within a first time period of predefined duration, wherein the first time period of predefined duration includes several frames of radar data, identifying a set of clusters of detections in the space by analyzing similarity in position of the accumulated detections, wherein each cluster includes a plurality of detections that were made at various time points during the first time period, removing, from the set of clusters, any cluster whose detections are temporally distributed within a proportion of the first time period which is below a predefined proportion threshold, and determining that positions in the space that correspond to the set of clusters were occupied during the first time period.
2 . The method of claim 1 , wherein each detection is further associated with a velocity of the object movement, and wherein in the step of accumulating only detections having a velocity below a first velocity threshold are accumulated.
3 . The method of claim 2 , wherein the first velocity threshold is lower than an average walking speed of a person.
4 . The method of claim 1 , wherein the step of removing further comprises removing, from the set of clusters, any cluster for which: a temporal correlation between the detections of the cluster and the detections of another cluster in the set of clusters is above a temporal correlation threshold, and a distance from the detections of the cluster to the radar is larger than a distance from the detections of said another cluster to the radar.
5 . The method of claim 1 , wherein the proportion of the first time period within which the detections of a cluster are temporally distributed is determined as a proportion of the first time period during which the detections of the cluster were present.
6 . The method of claim 1 , wherein the proportion of the first time period within which the detections of a cluster are temporally distributed is determined by comparing a dispersion measure of the time points of the detections of the cluster to the predefined duration of the first time period.
7 . The method of claim 1 , wherein the step of identifying a set of clusters of detections in the space includes applying a clustering algorithm to positions of the accumulated detections.
8 . The method of claim 1 , wherein the step of identifying a set of clusters of detections in the space includes: generating an occupancy map of the space by defining a plurality of grid cells in the space, and counting for each grid cell how many of the accumulated detections that have a position in the grid cell, and identifying the set of clusters by performing blob detection in the occupancy map of the space.
9 . The method of claim 8 , wherein the blob detection includes identifying local maxima in the occupancy map.
10 . The method of claim 1 , wherein the method is repeated for a subsequent time period of predefined duration.
11 . The method of claim 10 , further comprising: tracking the positions in the space that are determined to be occupied between the first time period and the subsequent time period.
12 . The method of claim 11 , further comprising: removing tracks which are outside of a region of interest in the space.
13 . A device for determining occupied positions in a space, wherein the device comprises circuitry configured to: receive detections of object movement from a radar monitoring the space, wherein each detection is associated with a position in the space and a time point when the detection was made, wherein the received detections include detections from a plurality of time points corresponding to a plurality of frames of radar data, accumulate detections that have a time point within a first time period of predefined duration, wherein the first time period of predefined duration includes several frames of radar data, identify a set of clusters of detections in the space by analyzing similarity in position of the accumulated detections, wherein each cluster includes a plurality of detections that were made at various time points during the first time period, remove, from the set of clusters, any cluster whose detections are temporally distributed within a proportion of the first time period which is below a predefined proportion threshold, and after the removal, determine that positions in the space that correspond to the set of clusters were occupied during the first time period.
14 . A non-transitory computer-readable medium comprising computer program code which, when executed by a device with processing capability, causes the device to carry out a method for determining occupied positions in a space, comprising: receiving detections of object movement from a radar monitoring the space, wherein each detection is associated with a position in the space and a time point when the detection was made, wherein the received detections include detections from a plurality of time points corresponding to a plurality of frames of radar data, accumulating detections that have a time point within a first time period of predefined duration, wherein the first time period of predefined duration includes several frames of radar data, identifying a set of clusters of detections in the space by analyzing similarity in position of the accumulated detections, wherein each cluster includes a plurality of detections that were made at various time points during the first time period, removing, from the set of clusters, any cluster whose detections are temporally distributed within a proportion of the first time period which is below a predefined proportion threshold, and after the removing, determining that positions in the space that correspond to the set of clusters were occupied during the first time period.

Description

TECHNICAL FIELD The present invention relates to the field occupancy detection in a space. In particular, it relates to a method and a system for determining occupied positions in a space by using radar technology. BACKGROUND In various environments, such as offices, public spaces, or industrial facilities, it is often valuable to know the number of individuals present and their spatial locations within the environment. Such information can be used for optimizing resource allocation and improving operational efficiency. For instance, in an office setting, understanding which desks are occupied and for how long can help facility managers optimize desk layouts, reduce energy consumption, and improve employee productivity. However, traditional methods for determining occupancy, such as analyzing images captured by surveillance cameras, may not always be feasible due to concerns about individual privacy. Radar technology offers a promising alternative for detecting occupancy without compromising individual privacy. By emitting radio waves and measuring the reflections, radar systems can detect the presence and location of objects within a given space. A challenge in occupancy detection is to distinguish between positions in the space that are occupied for a longer period of time, like desks occupied by individuals, and temporary presence in the space, such as individuals passing through the space. Thus, it is desirable to detect the locations of objects that remain at the same location over a period of time and disregard other temporarily present objects. To address these issues, there is hence a need for a reliable method that can accurately detect occupied positions within a scene using radar technology. Such a method would enable the identification of areas with prolonged occupation while still respecting individual privacy. SUMMARY OF THE INVENTION In view of the above, it is thus an object of the present invention to mitigate the above problems and provide a method and device that allows for determining occupied positions in a scene by using detections from a radar. According to a first aspect, the above object is achieved by a method for determining occupied positions in a space. The method comprises: receiving detections of object movement from a radar monitoring the space, wherein each detection is associated with a position in the space and a time point when the detection was made, accumulating detections that have a time point within a first time period of predefined duration, identifying a set of clusters of detections in the space by analyzing similarity in position of the accumulated detections, removing, from the set of clusters, any cluster whose detections are temporally distributed within a proportion of the first time period which is below a predefined proportion threshold, and determining that positions in the space that correspond to the set of clusters were occupied during the first time period. The invention relies on the idea that even if an object which is capable of moving remains at the same position within an environment it will still make small movements every now and then, referred to herein as micro-movements. For example, a person sitting at a desk will make some movements with the arms, torso, and head while working. These movements can be detected by the radar and will be located at roughly the same location in the space, but will typically not be detected so often by the radar. However, if radar detections are accumulated over a period of time, the detections stemming from movable objects that remain at the same position will form clusters in the space. This is in contrast to the movement of an object that moves around in the space, i.e., that changes its location. That movement will be distributed over a larger area in the space. Thus, when radar detections are accumulated over time, the detections stemming from objects that move around in the space will typically be spread out in the space and not form clusters. The accumulation of detections over time in combination with spatial clustering hence allows movable objects that remain at the same position for a prolonged period of time, i.e., during the first time period, to be distinguished from those that move around in the space. Moreover, even if the detections of an object that moves around in the space happen to form a cluster in the space, for instance due to the object making a short stop at some position, the detections in the cluster will be limited to a time window which is short in relation to the first time period. Therefore, any spatial cluster whose associated radar detections are temporally distributed during a limited proportion of the first time period is preferably removed. In this way, the risk that an object which makes short stops when passing through the space is mistaken for an object that remain at the same position for a prolonged period of time is reduced. By an occupied position in the space is general