Unattended object detection using machine learning

Patent metadata

Field	Value
Publication number	US-2022019854-A1
Application number	US-202016931098-A
Country	US
Kind code	A1
Filing date	Jul 16, 2020
Priority date	Jul 16, 2020
Publication date	Jan 20, 2022
Grant date	—

Abstract

Techniques for unattended object detection using machine learning are disclosed. A machine learning policy, for use in identifying unattended objects in a captured image depicting one or more objects, is generated. The generating includes determining a level of occlusion in the captured image relating to the objects, and determining the machine learning policy based on the determined level of occlusion. A machine learning model is selected, from among a plurality of pre-defined machine learning models, based on the generated machine learning policy. An unattended object is detected in the captured image using the selected machine learning model.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
generating a machine learning policy for use in identifying unattended objects in a captured image depicting one or more objects, comprising:
determining a level of occlusion in the captured image relating to the objects; and
determining the machine learning policy based on the determined level of occlusion;
selecting a machine learning model, from among a plurality of pre-defined machine learning models, based on the generated machine learning policy; and identifying an unattended object in the captured image using the selected machine learning model.

2 . The method of claim 1 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image does not exceed a first occlusion threshold, and in response selecting a simple feature analysis model for the determined machine learning policy, wherein the selected machine learning model comprises the simple feature analysis model.

3 . The method of claim 2 , wherein the simple feature analysis model uses intersection over union (IoU).

4 . The method of claim 1 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image exceeds a first occlusion threshold, and in response selecting a deep learning model for the determined machine learning policy, wherein the selected machine learning model comprises the deep learning model.

5 . The method of claim 4 , wherein generating the machine learning policy further comprises:
calculating an object size relating to the one or more objects; and determining that the object size exceeds an object size threshold, wherein the selecting the deep learning model is further based on the determining that the object size exceeds the object size threshold.

6 . The method of claim 5 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion exceeds a second occlusion threshold, wherein the selecting the deep learning model is further based on the determining that the level of occlusion exceeds the second occlusion threshold.

7 . The method of claim 1 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image exceeds a first occlusion threshold; calculating an object size relating to the one or more objects; and determining that the object size does not exceed an object size threshold, and in response selecting a simple feature analysis model for the determined machine learning policy, wherein the selected machine learning model comprises the simple feature analysis model.

8 . The method of claim 7 , wherein the simple feature analysis model comprises edge detection.

9 . The method of claim 7 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion does not exceed a second occlusion threshold, wherein the selecting the simple feature analysis model is further based on the determining that the level of occlusion does not exceed the second occlusion threshold.

10 . The method of claim 1 , wherein generating the machine learning policy further comprises:
determining that at least one of the objects is a person, wherein the determining the machine learning policy is further based on the determining that at least one of the objects is a person.

11 . A system, comprising:
a processor; and a memory storing a program, which, when executed on the processor, performs an operation, the operation comprising:
generating a machine learning policy for use in identifying unattended objects in a captured image depicting one or more objects, comprising:
determining a level of occlusion in the captured image relating to the objects; and
determining the machine learning policy based on the determined level of occlusion;

selecting a machine learning model, from among a plurality of pre-defined machine learning models, based on the generated machine learning policy; and
identifying an unattended object in the captured image using the selected machine learning model.

12 . The system of claim 11 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image does not exceed a first occlusion threshold, and in response selecting a simple feature analysis model for the determined machine learning policy, wherein the selected machine learning model comprises the simple feature analysis model.

13 . The system of claim 11 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image exceeds a first occlusion threshold, and in response selecting a deep learning model for the determined machine learning policy, wherein the selected machine learning model comprises the deep learning model.

14 . The system of claim 11 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image exceeds a first occlusion threshold; calculating an object size relating to the one or more objects; and determining that the object size does not exceed an object size threshold, and in response selecting a simple feature analysis model for the determined machine learning policy, wherein the selected machine learning model comprises the simple feature analysis model.

15 . The system of claim 11 , wherein generating the machine learning policy further comprises:
determining that at least one of the objects is a person, wherein the determining the machine learning policy is further based on the determining that at least one of the objects is a person.

16 . A non-transitory computer program product, the computer program product comprising:
a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation, the operation comprising:
generating a machine learning policy for use in identifying unattended objects in a captured image depicting one or more objects, comprising:
determining a level of occlusion in the captured image relating to the objects; and
determining the machine learning policy based on the determined level of occlusion;

selecting a machine learning model, from among a plurality of pre-defined machine learning models, based on the generated machine learning policy; and
identifying an unattended object in the captured image using the selected machine learning model.

17 . The computer program product of claim 16 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image does not exceed a first occlusion threshold, and in response selecting a simple feature analysis model for the determined machine learning policy, wherein the selected machine learning model comprises the simple feature analysis model.

18 . The computer program product of claim 16 , wherein generating the machine learning policy further comprises:
determining that the level of occlusion in the captured image exceeds a first occlusion threshold, and in response selecting a deep learning model for the determined machine learning policy, wherein the selected machine learning model comprises the deep learning model.

19 . The computer program product of claim 18 , wherein generating the machine l

Assignees

• IBM

Inventors

• Yan Rong
• Zhang shi lei
• Wang Junsong
• Sun Ke Wei

Classifications

• G06F18/2148 Primary

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06T2207/30232

  Surveillance · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title