Door operating system
A door operating system including a control unit and at least one sensor device, the system configured to operate a door from a closed state into an opened state. The sensor device is configured to observe an observation area and provide image data to the control unit, which is configured to receive and process the observation image data. Processing the data includes detecting an individual inside the observation area, extracting pose information and/or viewing direction information corresponding to the individual, determining, based on the pose information and/or the viewing direction information, a probability value representing the probability that the individual intends to pass the door, and determining whether the probability value is higher than a probability threshold value, wherein the door operating system operates the door from a closed state in an opened state if the control unit determines that the probability value is higher than a probability threshold value.
1. A door operating system comprising a control unit and at least one sensor device, the door operating system being configured to operate a door from a closed state into an opened state,
wherein the sensor device is configured to observe an observation area
and provide respective observation image data to the control unit,
wherein the control unit is configured to receive and process the observation image data,
wherein processing the observation image data comprises at least:
detecting an individual among other objects inside the observation area including distinguishing between individuals and non-individuals, being carried out based on the shape of a detected object,
extracting pose information and/or viewing direction information corresponding to the respective individual,
determining, based on the pose information and/or the viewing direction information, a probability value representing the probability that the respective individual intends to pass the door,
determining whether the probability value is higher than a probability threshold value,
wherein the door operating system operates the door from a closed state in an opened state if the control unit determines that the probability value is higher than a probability threshold value,
wherein extracting pose information and/or viewing direction information corresponding to the respective individual is performed by an image analysis engine,
wherein the image analysis engine is configured to extract a feature vector corresponding to the respective individual, the feature vector comprising a first set of features indicative for a pose of the individual and/or a second set of features indicative for a viewing direction of the individual, wherein the first set of features and the second set of features are at least partially mutually associated, the image analysis engine being configured to perform an inconsistency detection loop on the basis of the associated features for detecting and rectifying the feature vector.
2. The door operating system of claim 1 , wherein the pose information includes motion of pose of the respective individual and/or viewing direction information includes motion of viewing direction of the respective individual.
3. The door operating system of claim 1 , wherein detecting an individual inside the observation area is performed by an image classification engine,
wherein the image classification engine is configured to detect individuals, among one or more objects which may be represented by the observation image data, the image classification engine being trained via machine learning.
4. The door operating system of claim 1 , wherein the first set of features at least comprises the orientation of one or more extremities of a skeleton of the respective individual, at least the direction in which toes of respective feet of the individual are pointing.
5. The door operating system of claim 1 , wherein the image analysis engine is trained by machine learning.
6. The door operating system of claim 1 , wherein the probability threshold value is dynamically adjusted by a probability adjustment engine,
wherein the probability adjustment engine validates, for a detected individual for whom it is determined that the probability value is higher than a probability threshold value, whether the respective individual actually passed the respective door.
7. The door operating system of claim 1 , wherein in determining the probability value representing the probability that the respective individual intends to pass the door, path information obtained by a path tracking engine is considered,
wherein the path tracking engine provides a set of most frequently used paths for passing a respective door and considers, based on motion of a detected individual, whether the respective individual is about to use one of the most frequently used paths, wherein the set of most frequently used paths is obtained or dynamically updated through machine learning.
8. The door operating system of claim 1 , wherein the observation area is allocated to the respective door and is sectioned into a plurality of zones, namely at least a tracking zone and an activation zone, the activation zone being located closer to the respective door than the tracking zone,
wherein observation image data from the tracking zone is used to wake up the control unit and/or respective engines if an object enters the tracking zone,
wherein observation image data from the activation zone is used to trigger operating the respective door from a closed state into an opened state.
9. The door operating system of claim 1 , wherein the observation area is allocated to the respective door and is sectioned into a plurality of zones, namely at least an activation zone and a safety zone,
the safety zone being located closer to the respective door than the activation zone,
wherein observation image data from the activation zone is used to trigger operating the respective door from a closed state into an opened state,
wherein observation image data from the safety zone is used to operate the respective door into a safety mode, wherein in the safety mode the respective door is operated from a closed state into an opened state irrespective of the probability value.
10. The automatic door operating system of claim 1 , wherein operating the respective door from a closed state into an opened state is performed using a door motor disposed at the door configured to open the door, wherein the opening speed of the door is set depending on motion speed of the respective individual and/or the opening width of the door is set depending on the quantity of detected respective individuals.
11. The door operating system of claim 1 , wherein the observation area is allocated to the respective door and is sectioned into a plurality of zones,
at least a tracking zone, an activation zone and a safety zone, the safety zone being located closer to the respective door than the activation zone, and the activation zone being located closer to the respective door than the tracking zone,
wherein observation image data from the plurality of zones is used to trigger operating the respective door from an opened state into a closed state if a predetermined closing condition is met,
wherein the closing condition at least comprises that a respective individual has left one of the plurality of zones.
12. A computer implemented method for operating a door operating system according to claim 1 , the method including the following steps:
detecting an individual inside an observation area,
extracting pose information and/or viewing direction information corresponding to the respective individual,
determining, based on the pose information and/or the viewing direction information, a probability value representing the probability that the respective individual intends to pass the door,
determining whether the probability value is higher than a probability threshold value, and
operating a door from a closed state in an opened state if it is determined that the probability value is higher than a probability threshold value.
13. An access control system comprising a door and a door operating system according to claim 1 .