Condition detection using image processing

Patent metadata

Field	Value
Publication number	US-10127449-B2
Application number	US-201515313787-A
Country	US
Kind code	B2
Filing date	Aug 6, 2015
Priority date	Aug 6, 2015
Publication date	Nov 13, 2018
Grant date	Nov 13, 2018

Abstract

Condition detection using image processing may include receiving a mask generated from images and telemetry data captured by a vehicle, an altitude map, and alignment data for the mask. The images may be related to movement of the vehicle along a vehicle path and non-infrastructure entities along an infrastructure entity position of a corresponding infrastructure entity, and the telemetry data may include movement log information related to the movement of the vehicle along the vehicle path. Condition detection using image processing may further include using the mask related to the vehicle path and the non-infrastructure entities, and an infrastructure rule to detect a risk related to the infrastructure entity by analyzing the mask related to the vehicle path and the non-infrastructure entities, and the infrastructure rule, and determining whether the infrastructure rule is violated.

First claim

What is claimed is:

1. A condition detection using image processing system comprising:
a risks analyzer, executed by at least one hardware processor, to
receive at least one mask generated from images and telemetry data captured by a vehicle, an altitude map, and alignment data for the at least one mask, wherein
the images are related to movement of the vehicle along a vehicle path and at least one of a plurality of non-infrastructure entities along an infrastructure entity position of a corresponding infrastructure entity, and
the telemetry data includes movement log information related to the movement of the vehicle along the vehicle path, and

use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, a height map, the alignment data for the at least one mask, and an infrastructure rule to detect a risk related to the infrastructure entity by
analyzing the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, and the infrastructure rule,
determining whether the infrastructure rule is violated, and
determining a likelihood of the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity of falling over the infrastructure entity, wherein the risk includes the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity falling over the infrastructure entity.

2. The condition detection using image processing system according to claim 1 , wherein the risks analyzer is to use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, the height map, the alignment data for the at least one mask, and the infrastructure rule to detect the risk related to the infrastructure entity by
identifying at least one infrastructure rule condition related to the infrastructure entity,
determining whether the infrastructure rule condition is satisfied or not satisfied by the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, and
in response to a determination that the infrastructure rule condition is satisfied or not satisfied by the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, determining the risk related to the infrastructure entity by the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity.

3. The condition detection using image processing system according to claim 2 , wherein the at least one infrastructure rule condition includes satisfying a height threshold, and wherein the risks analyzer is to further determine whether the height threshold is met or exceeded by the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity by
determining a ground level in a region of analysis by using a ground mask and the altitude map related to the region of analysis,
determining a height map by using the ground level and the altitude map related to the region of analysis, and
identifying, by using the height map, the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity that includes a height that exceeds the height threshold.

4. The condition detection using image processing system according to claim 1 , wherein the risks analyzer is to use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, the height map, the alignment data for the at least one mask, and the infrastructure rule to detect the risk related to the infrastructure entity by
determining a height for which the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity is to fall over the infrastructure entity, wherein the risk includes the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity falling over the infrastructure entity, and
generating a group by grouping the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity for which an actual height exceeds the height for which the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity is to fall over the infrastructure entity.

5. The condition detection using image processing system according to claim 1 , wherein the risks analyzer is to use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, the height map, the alignment data for the at least one mask, and the infrastructure rule to detect the risk related to the infrastructure entity by
determining a height for which the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity is to fall to a specified region, wherein the risk includes the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity falling to the specified region.

6. The condition detection using image processing system according to claim 1 , wherein the risks analyzer is to use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity, the height map, the alignment data for the at least one mask, and the infrastructure rule to detect the risk related to the infrastructure entity by
generating a group by grouping the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity for which the likelihood is more than a predetermined threshold, and
determining the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity in the group that includes a maximum likelihood of the risk of falling over the infrastructure entity.

7. The condition detection using image processing system according to claim 6 , wherein the risks analyzer is to further
determine a distance between the at least one of the plurality of non-infrastructure entities along the infrastructure entity position of the corresponding infrastructure entity in the group that includes the maximum likelihood of the risk of falling over the infrastructure entity.

8. The condition detection using image processing system according to claim 1 , wherein the risks analyzer is to use the at least one mask related to the vehicle path and the at least one of the plurality of non-infrastructure entities al

Assignees

• Accenture Global Services Ltd

Inventors

• Pestun Vadim
• Eremina Olga
• Tyulyaeva Ekaterina

Classifications

• G05D1/0094

  involving pointing a payload, e.g. camera, weapon, sensor, towards a fixed or moving target · CPC title

• G06K9/00637

  Physics · mapped topic

• G06K9/00657 Primary

  Physics · mapped topic

• G08G5/0086

  Physics · mapped topic

• G08G5/0069

  Physics · mapped topic