System and method for detecting proximity between objects
A method is provided for detecting proximity between objects using threshold based clustering. The method includes receiving object data of a plurality of objects in a space to be monitored, from one or more data capturing devices disposed in a space, clustering the object data into one or more clusters, based on a predetermined threshold distance, thereby reducing the search space, and processing the one or more clusters using brute force to identify the plurality of objects which lie within the predetermined threshold distance of each other, thereby identifying the plurality of objects that are at risk of collision.
1. A method ( 200 ) for detecting proximity between objects using threshold-based clustering, the method ( 200 ) comprising:
receiving ( 210 ) object data of a plurality of objects in a space to be monitored, from one or more data capturing devices ( 102 ) disposed in a space, wherein the one or more data capturing devices include 3D sensors such as radars, LiDARs, Laser Detection and Ranging (LaDAR), Light Emitting Diode Detection and Ranging (LeDDAR) mmWave Radar, C or K Band Radar, laser scanners and Time of Flight (ToF) sensors;
clustering ( 220 ) the object data using Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) clustering into one or more clusters, based on a predetermined spatial threshold distance, thereby reducing the search space, wherein the object data is represented as 2D, 3D, or N-dimensional point clouds;
processing ( 230 ) the one or more clusters by performing brute force computation on the reduced search space to identify the plurality of objects which lie within the predetermined threshold distance of each other, thereby identifying the plurality of objects that are at risk of collision; and
providing real-time results of identified objects at risk of collision.
2. The method ( 200 ) as claimed in claim 1 , wherein the step of clustering is to reduce computational complexity and processing is independent of the dimensionality of the monitored space, allowing for detection in 2D, 3D, or higher-dimensional object data.
3. The method ( 200 ) as claimed in claim 1 , wherein the step of processing the one or more clusters using brute force computation comprises identifying & providing a list of data points which lie within the predetermined threshold distance of each other, and accordingly identifying the plurality of objects that are at risk of collision.
4. The method ( 200 ) as claimed in claim 1 , wherein the plurality of objects are all kinds of living and non-living objects selected from a group comprising humans of multiple age groups, animals, plants, furniture, vehicles, natural resources, eatables, crops, infrastructure, stationery, sign boards, wearables, musical instruments, sports equipment, mechanical tools, electrical equipment and electronic equipment.
5. A system ( 100 ) for detecting proximity between objects using threshold-based clustering, the system ( 100 ) comprising:
one or more data capturing devices ( 102 ) disposed in a space to be monitored, wherein the one or more data capturing devices include 3D sensors such as radars, LiDARs, Laser Detection and Ranging (LaDAR), Light Emitting Diode Detection and Ranging (LeDDAR) mmWave Radar, C or K Band Radar, laser scanners and Time of Flight (ToF) sensors; and
a processing module ( 104 ) connected with the one or more data capturing devices ( 102 ), the processing module ( 104 ) comprising:
a memory unit ( 1042 ) configured to store machine-readable instructions; and
a processor ( 1044 ) operably connected with the memory unit ( 1042 ), the processor ( 1044 ) obtaining the machine-readable instructions from the memory unit ( 1042 ), and being configured by the machine-readable instructions to:
receive object data represented as 2D, 3D, or N-dimensional point clouds of a plurality of objects from the one or more data capturing devices ( 102 ) disposed in the space;
cluster the object data using Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) clustering, into one or more clusters, based on a predetermined threshold distance to reduce the search space; and
process the one or more clusters by performing brute force computation on the reduced search space to identify the plurality of objects which lie within the predetermined threshold distance of each other, thereby identifying the plurality of objects that are at risk of collision;
a configurable processing unit configured to accelerate clustering and brute force processing through runtime creation of soft logic cores, enabling hardware threading for dynamic load handling and real-time output.
6. The system ( 100 ) as claimed in claim 5 , wherein the processing module ( 104 ) is configured to run
an operating system ( 1048 ) which is further configured to make the processing on the Software (SW) threads on processor ( 1044 ) and Hardware (HW) threads on configurable processing unit ( 1046 ) transparent to application.
7. The system ( 100 ) as claimed in claim 5 , wherein the processing module ( 104 ) is configured to facilitate clustering by using Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) clustering to reduce computational complexity by grouping object data into clusters based on a predetermined spatial threshold distance, and processing is independent of the dimensionality of the monitored space, allowing for detection in 2D, 3D, or higher-dimensional object data.
8. The system ( 100 ) as claimed in claim 5 , wherein the processing module ( 104 ) is configured to process the one or more clusters using brute force by identifying & providing a list of data points which lie within the predetermined threshold distance of each other, and accordingly identifying the plurality of objects that are at risk of collision.
9. The system ( 100 ) as claimed in claim 5 , wherein the plurality of objects are all kinds of living and non-living objects selected from a group comprising humans of multiple age groups, animals, plants, furniture, vehicles, natural resources, eatables, crops, infrastructure, stationery, sign boards, wearables, musical instruments, sports equipment, mechanical tools, electrical equipment and electronic equipment.