EP-4738291-A1 - METHODS AND SYSTEMS FOR OBJECT DETECTION

EP4738291A1EP 4738291 A1EP4738291 A1EP 4738291A1EP-4738291-A1

Abstract

A computer implemented method for object detection comprises the following steps carried out by computer hardware components: acquiring an image of an environment; determining at least one reflection in the image, the at least one reflection being at least one reflection of a light source of an occluded object; determining enhanced reflection information based on the at least one reflection; and detecting the occluded objected based on the enhanced reflection information.

Inventors

HAHN, Lukas
Paus, Andre
Friedrichs, Klaus

Assignees

Aptiv Technologies AG

Dates

Publication Date: 20260506
Application Date: 20241029

Claims (15)

Computer implemented method for object detection, the method comprising the following steps carried out by computer hardware components: - acquiring (402) an image of an environment; - determining (404) at least one reflection in the image, the at least one reflection being at least one reflection of a light source of an occluded object; - determining (406) enhanced reflection information based on the at least one reflection; and - detecting (408) the occluded objected based on the enhanced reflection information.
The computer implemented method of claim 1, wherein the enhanced reflection information comprises a pattern of a plurality of reflections, and wherein the occluded object is detected based on the pattern.
The computer implemented method of at least one of claims 1 to 2, wherein the enhanced reflection information comprises information on a change of the reflection over time, and wherein the occluded object is detected based on the change.
The computer implemented method of at least one of claims 1 to 3, further comprising the following steps carried out by the computer hardware components: detecting a visible object in the environment; and determining whether the visible object is a static object; wherein the at least one reflection comprises at least one reflection on the determined visible object if it is determined that the visible object is a static object.
The computer implemented method of claim 4, wherein determining whether the visible object is a static object comprises determining an image of the visible object and/ or Lidar measurements related to the visible object and/ or radar measurements related to the visible object.
The computer implemented method of at least one of claims 4 or 5, wherein determining whether the visible object is a static object comprises determining an orientation of the visible object.
The computer implemented method of at least one of claims 1 to 6, further comprising the following steps carried out by the computer hardware components: determining situation information; determining whether the computer implemented method is to be carried out based on the situation information; and carrying out the computer implemented method only if it is determined that the computer implemented method is to be carried out.
The computer implemented method of claim 7, wherein the situation information comprises at least one of a daytime information, an environment traffic situation, a road width information, or a visibility information.
The computer implemented method of at least one of claims 1 to 8, further comprising the following steps carried out by the computer hardware components: determining tracking information of the at least one reflection; wherein the enhanced reflection information is determined based on the tracking information.
The computer implemented method of at least one of claims 1 to 9, wherein the method is carried out in relation to an ego-vehicle, the method further comprising at least one of the following steps carried out by the computer hardware components: outputting a warning to a driver of the ego-vehicle based on detecting the occluded object; or influencing a planning algorithm of an autonomous driving function of the ego-vehicle to reduce the risk of collision with the occluded object.
The computer implemented method of at least one of claims 1 to 10, wherein detecting (408) the occluded object comprises determining at least one of a distance of the occluded object, a velocity of the occluded object, or a direction of movement of the occluded object.
The computer implemented method of at least one of claims 1 to 11, wherein the method is carried out in relation to an ego-vehicle, the method further comprising the following step carried out by the computer hardware components: determining a risk of collision between the ego-vehicle and the occluded object based on detecting the occluded object.
Computer system (500), the computer system (500) comprising a plurality of computer hardware components configured to carry out the computer implemented method of at least one of claims 1 to 12.
Vehicle comprising the computer system (500) of claim 13 and a sensor (508) configured to acquire the image.
Non-transitory computer readable medium comprising instructions for carrying out the computer implemented method of at least one of claims 1 to 12.

Description

FIELD The present disclosure relates to methods and systems for object detection. BACKGROUND Advanced driver-assistance systems (ADAS) may be desired to be capable of detecting other traffic participants reliably and over larger distances. Occlusion of the direct line of sight however may become an issue, as it makes all state-of-the-art sensors, like lidar, radar and camera, become blind. This might lead to hazardous situations as objects might appear suddenly and users can generally be led to put too much trust in assistance systems when not being able to understand their (physical) capabilities. Accordingly, there is a need to provide enhanced methods and systems for object detection. SUMMARY The present disclosure provides a computer implemented method, a computer system, a vehicle and a non-transitory computer readable medium according to the independent claims. Embodiments are given in the subclaims, the description and the drawings. In one aspect, the present disclosure is directed at a computer implemented method for object detection, the method comprising the following steps carried out by computer hardware components: acquiring an image of an environment; determining at least one reflection in the image, the at least one reflection being at least one reflection of a light source of an occluded object; determining enhanced reflection information based on the at least one reflection; and detecting the occluded objected based on the enhanced reflection information. In other words, headlight reflections may be used to detect occluded vehicles. According to various embodiments, the occluded object may include or may be a vehicle. Reflection may be understood as the change in direction of an optical wavefront at an interface between two different media (for example between air and a parked vehicle) so that the wavefront returns into the medium from which it originated. According to an embodiment, the enhanced reflection information comprises a pattern of a plurality of reflections, and the occluded object is detected based on the pattern. The pattern may include features such as pixel brightness, and/ or reflection area, and/ or pixel color values, and/ or reflection motion relative to the reflecting body. According to an embodiment, the enhanced reflection information comprises information on a change of the reflection over time, and the occluded object is detected based on the change. For example, if the reflection over time increases in intensity, this may be considered as indicating that the reflection is related to an approaching object. According to an embodiment, the computer implemented method further comprises the following steps carried out by the computer hardware components: detecting a visible object in the environment; and determining whether the visible object is a static object; wherein the at least one reflection comprises at least one reflection on the determined visible object if it is determined that the visible object is a static object. Reflections on a static object may solely change due to a relative movement between the observer (for example an ego-vehicle) and an occluded object. According to an embodiment, determining whether the visible object is a static object comprises determining an image of the visible object and/ or Lidar measurements related to the visible object and/ or radar measurements related to the visible object. Based on the image and/or Lidar/ radar measurements, it may be determined whether the visible object is moving or not. According to an embodiment, determining whether the visible object is a static object comprises determining an orientation of the visible object. For example, for visible objects being a vehicle, it may be determined that the visible object is moving if the vehicle is oriented perpendicular to the forward direction of the ego-vehicle, since this may indicate that the visible object is turning onto the road on which the ego-vehicle is driving. According to an embodiment, the computer implemented method further comprises the following steps carried out by the computer hardware components: determining situation information; determining whether the computer implemented method is to be carried out based on the situation information; and carrying out the computer implemented method only if it is determined that the computer implemented method is to be carried out. It has been found that executing the method only under certain circumstances reduces computational effort, in particular in situations where the result of the object detection is not necessary. According to an embodiment, the situation information comprises at least one of a daytime information, an environment traffic situation, a road width information, or a visibility information. According to an embodiment, the computer implemented method further comprises the following steps carried out by the computer hardware components: determining tracking information of the at least one ref