Method and system for monitoring a safety area
Patent Information
- Application Number
- EP2024715107
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-03-28
- Filing Date
- 2024-03-22
- Publication Date
- 2026-02-11
AI Technical Summary
Existing security monitoring systems for areas accessible via secure locks are inefficient in quickly detecting and responding to intrusions, as they require evaluating camera images regardless of the presence of objects, wasting resources and delaying the securing of the area.
A method and system where a sensor detects object penetration, triggering the camera to record and evaluate images only when necessary, with automated evaluation using AI to quickly secure and release the area, and energy-efficient recording strategies.
Enhances the efficiency of security monitoring by ensuring rapid detection and response to intrusions, reducing resource waste and enabling quick resumption of operations by evaluating images only when necessary and using AI for automated assessment.
Smart Images
Figure EP2024057784_03102024_PF_FP_ABST
Abstract
Description
[0001] Method and system for monitoring a security area
[0002] The invention relates to a method for monitoring a security area.
[0003] The invention further relates to a system for monitoring a security area.
[0004] From DE 10 2009 36641 A1 a security device is known for securing a room area that is only accessible via a secure access lock.
[0005] The invention is based on the object of providing a method and a system which further develop the state of the art.
[0006] This object is achieved by a method according to claim 1 and a system according to claim 14.
[0007] The invention comprises a method for monitoring a safety area, in particular a machine tool, wherein the safety area has at least one access area and access to the safety area is only possible via the access area, wherein the access area is monitored by at least one camera and a further sensor, wherein the camera and the further sensor are connected to a computer system, wherein the computer system monitors by means of the further sensor whether an object is penetrating the safety area, wherein the camera takes at least one image of the access area when the computer system detects by means of the further sensor that an object is penetrating the safety area, wherein the image recorded by the camera is only evaluated by the computer system after it has been detected by means of the further sensor that an object has penetrated the safety area.
[0008] The intrusion of an object into the security area is therefore first detected by the additional sensor. Only then is the camera image evaluated. This image evaluation requires complex computational operations. The efficiency of the process is increased by evaluating the image only when necessary—namely, when the additional sensor has detected that an object has entered the security area.
[0009] The machine tool can, in particular, be a laser cutting machine, a laser welding machine, a punching machine, or a bending machine. Lasers and parts of the machine tool that move with great force pose a hazard that can potentially harm people or objects in the safety zone. On the other hand, objects in the safety zone can pose a hazard to the often complex and therefore valuable machine tool.
[0010] Preferably, the safety area is secured when the additional sensor has detected that an object has entered the safety area. Securing the safety area means that sources of danger in the safety area are deactivated. In the case of a machine tool's safety area, for example, the machine tool is brought into a safe operating state, e.g., by stopping the machine tool. Securing the safety area when the additional sensor has detected that an object has entered the safety area ensures particularly rapid security of the safety area. There is no need to wait until the image recorded by the camera has been evaluated.
[0011] Preferably, the camera records images repeatedly, especially at regular intervals, especially in video form. By recording multiple images, the status of the access area can be tracked over time.
[0012] Preferably, the images recorded by the camera are stored in a memory, in particular a ring buffer, for at least one second, preferably 2 seconds, particularly preferably 5 seconds. By storing the images, after the additional sensor has detected that an object has entered the security area, the object's entry into the security area can be reconstructed and / or verified in the images.
[0013] In a preferred embodiment, the camera continuously records images, particularly videos. Due to the longer storage time and, in particular, the use of a ring buffer, the condition of the access area prior to the object's entry can also be determined from the images.
[0014] In an alternative embodiment, the camera only starts recording one or more images after the additional sensor has detected that an object has entered the security area. By only recording images when needed, energy can be saved. Preferably, the camera starts recording within one second, particularly preferably within half a second, after the additional sensor has detected that an object has entered the security area.
[0015] Preferably, the camera continuously records images, at least after the additional sensor has detected that an object has entered the security area. Continuous recording allows the status of the security area to be reconstructed after the object has entered.
[0016] The computer system preferably evaluates one or more images recorded by the camera automatically, preferably using artificial intelligence, and automatically releases the safety area if the evaluation shows that the safety area is safe. The automated evaluation releases the safety area very quickly. For example, the machine tool can resume work very quickly. The automated evaluation is particularly advantageous if the evaluation of the images shows that the object triggered the additional sensor but immediately left the safety area again. The computer system preferably examines several consecutive images for at least one moving object. By examining the images for a moving object, the evaluation of the images can be accelerated because only moving objects can penetrate the safety area.
[0017] Particularly preferably, the computer system determines a direction of movement of the moving object, in particular whether the direction of movement of the moving object is directed into or out of the safety area. By determining the direction of movement, the danger posed by the object can be better assessed. Typically, an object moving into the safety area poses a greater danger than an object moving out of the safety area. In particular, the computer system can determine that an object first moved into the safety area and then out again. In this case, the danger only existed temporarily, and the safety area can be reopened.
[0018] Preferably, the computer system outputs at least one image for an operator to release the safety area, and optionally, the computer system outputs a warning signal to the operator. Using the image, the operator can assess the situation in the danger zone and, if necessary, release the safety zone. If a persistent danger exists, the operator can initiate countermeasures. The warning signal informs the operator that the additional sensor has detected that an object has entered the safety zone, and the operator should evaluate the image. The warning signal can be issued visually, e.g., with a flashing light, acoustically, e.g., with a warning tone, and / or haptically, e.g., with a vibration alarm from a mobile device worn by the operator.
[0019] Particularly preferably, the computer system outputs to the operator at least one image, in particular a video, recorded by the camera shortly after the additional sensor detected that an object had entered the security zone, and at least one further image, in particular a current video transmission, representing a current status. The operator can use the two outputs to assess both the situation when the additional sensor detected that an object had entered the security zone, as well as the current situation. The situation can be accurately assessed, in particular, by comparing the images or videos.
[0020] Preferably, the security area has multiple access areas, each access area being monitored by the computer system with at least one camera and at least one additional sensor. Monitoring all access areas ensures equal protection of the security area.
[0021] Preferably, the additional sensor is a light barrier, a contact sensor, an ultrasonic sensor, or a radar sensor. These types of sensors can be evaluated particularly quickly and easily. Therefore, it can be quickly detected that an object has entered the security area. The security area can then be secured quickly.
[0022] The following description of preferred embodiments, in conjunction with the drawings, serves to explain the invention in more detail.
[0023] The invention also includes a system according to claim 15. According to the invention, a system comprising a safety area, in particular a safety area of a machine tool, a camera, a further sensor and a computer system, wherein the safety area has at least one access area and access to the safety area is possible exclusively via the access area, wherein the camera and the further sensor are provided and configured to monitor the access area, wherein the camera and the further sensor are connected to the computer system, wherein the computer system is provided and configured to monitor by means of the further sensor whether an object is penetrating the safety area, wherein the camera is provided and configured to take at least one image of the access area if the computer system detects by means of the further sensor that an object is penetrating the safety area,The computer system is designed and configured to evaluate the image recorded by the camera only after the additional sensor has detected that an object has entered the security area. In other words, the system is designed and configured to carry out a method according to the invention.
[0024] Particularly preferably, the system for carrying out a method according to the invention is provided and configured according to a preferred embodiment described above.
[0025] They show:
[0026] Fig. 1 shows a system comprising a security area with an access area;
[0027] Fig. 2 a system comprising a security area with two access areas; and
[0028] Fig. 3 shows an exemplary method according to the invention.
[0029] Identical or functionally equivalent elements are designated by the same reference numerals in all embodiments.
[0030] Fig. 1 shows a system comprising a safety area 1. A machine tool 2 is located within the safety area 1. The safety area 1 has an access area 3. Access to the safety area 1 is only possible via the access area 3. In this example, the safety area 1, with the exception of the access area 3, is protected by a fence. The access area 3 is monitored by a camera 4 and another sensor 5, in this case a light barrier. The camera 4 and the other sensor 5 are connected to a computer system 6.
[0031] An object 7, in this case a ball, moving through access area 3 is detected by the additional sensor 5. The computer system 6 thus detects, using the additional sensor 5, that the object 7 has entered the safety area 1 and subsequently secures the safety area 1. For this purpose, the computer system 6 is connected to the machine tool 2, and the computer system 6 stops the machine tool 2.
[0032] Camera 4 generates an image 8 at regular intervals. These images 8 are stored in a memory 9 as a video stream. Memory 9 is used as a ring buffer. Ring buffering means that a video stream of a fixed length, in this case ten seconds, is permanently stored. Newly received images 8 or parts of the video stream overwrite the oldest images 8 or parts of the video stream in memory 9.
[0033] After the computer system 6 has determined, using the additional sensor 5, that the object 7 has entered the safety area 1, the computer system 6 evaluates the images 8 generated by the camera 4 and stored in the memory 9. The computer system 6 contains image processing algorithms for this purpose. Using the algorithms, for example, the direction of movement 10 of the object 7 can be determined. The algorithms can also be used to determine the type of object, for example. The algorithms can contain artificial intelligence, for example, in the form of a neural network. If the computer system 6 detects that there is no danger in the safety area 1, the computer system 6 releases the safety area 1 and starts the machine tool 2.
[0034] If the computer system 6 detects a potential hazard in the safety area 1 or if the computer system 6 cannot clearly assess the situation, the computer system 6 displays the images 8 to an operator 11. To alert the operator 11 to the potential hazard in the safety area 1, the computer system 6 can emit a warning signal 12, in this case a flashlight. The operator 11 can then evaluate the images 8 and either release the safety area 1 or initiate the hazard elimination process.
[0035] Fig. 2 shows a system with a security area 1 and two access areas 3. Each access area 3 is monitored by a camera 4 and an additional sensor 5. The cameras 4 and the additional sensors 5 are each connected to the computer system 6. No object is shown in this example. It is understood that if an object enters the security area 1, the same thing happens as described in Fig. 1, regardless of which access area 3 the object enters the security area 1 through.
[0036] Fig. 3 shows an exemplary method according to the invention. In a first step 110, the computer system 6 uses the additional sensor 5 to monitor whether an object 7 is entering the safety area 1. As soon as an object 7 is detected by the additional sensor 5 in a second step 120, the safety area 1 is secured in a third step 130. For this purpose, the computer system 6 stops the machine tool 2. In a fourth step 140, one or more images 8 from the camera 4 from the time of detection of the object 7 and preferably from the time thereafter are processed by the computer system 6. In this example, the camera 4 generates images 8 at regular intervals, which are stored in a memory 9 and can be loaded from there by the computer system 6. If the computer system 6 detects that there is no danger in the safety area 1, the computer system 6 releases the safety area 1 in a fifth step 150.If the computer system 6 detects a danger in the safety area 1 or cannot clearly assess the situation, an operator 11 is informed by the computer system 6 in a sixth step 160.
[0037] List of reference symbols
[0038] 1 security area
[0039] 2 machine tools
[0040] 3 Access area
[0041] 4 Camera
[0042] 5 Sensor
[0043] 6 Computer system
[0044] 7 Object
[0045] 8 Image
[0046] 9 memory
[0047] 10 Direction of movement
[0048] 11 operators
[0049] 12 Warning signal
[0050] 100 procedures
[0051] 110 first step
[0052] 120 second step
[0053] 130 third step
[0054] 140 fourth step
[0055] 150 fifth step
[0056] 160 sixth step
Claims
Patent claims 1. Method for monitoring a safety area (1), in particular a safety area (1) of a machine tool (2), wherein the safety area (1) has at least one access area (3) and access to the safety area (1) is possible exclusively via the access area (3), wherein the access area (3) is monitored by at least one camera (4) and a further sensor (5), wherein the camera (4) and the further sensor (5) are connected to a computer system (6), wherein the computer system (6) monitors by means of the further sensor (5) whether an object (7) is penetrating the safety area (1), wherein the camera (4) takes at least one image (8) of the access area (3) when the computer system (6) detects by means of the further sensor (5) that an object (7) is penetrating the safety area (1), characterized in that the image (8) recorded by the camera (4) is Computer system (6) is only evaluated after the further sensor (5) has determined that an object (7) has entered the security area (1).
2. Method according to claim 1, characterized in that the security area (1) is secured if it has been determined by means of the further sensor (5) that an object (7) has penetrated into the security area (1).
3. Method according to one of the preceding claims, characterized in that the camera (4) records images (8) repeatedly, in particular at regular intervals, especially in the form of a video.
4. Method according to one of the preceding claims, characterized in that the images (8) recorded by the camera (4) are a memory (9), in particular a ring memory, for at least one second, preferably 2 seconds, particularly preferably 5 seconds.
5. Method according to one of the preceding claims, characterized in that the camera (4) continuously records images (8), in particular at regular intervals, very particularly in the form of a video.
6. Method according to one of claims 1 to 4, characterized in that the camera (4) only starts to record one or more images (8) after it has been determined by means of the further sensor (5) that an object (7) has penetrated into the security area (1).
7. Method according to one of the preceding claims, characterized in that the camera (4) continuously records images (8), at least after it has been determined by means of the further sensor (5) that an object (7) has penetrated into the security area (1).
8. Method according to one of the preceding claims, characterized in that the computer system (6) automatically evaluates one or more images (8) recorded by the camera (4), preferably by means of artificial intelligence, and automatically releases the security area (1) if the evaluation shows that the security area (1) is secure.
9. Method according to claim 8, characterized in that the computer system (6) examines several successive images (8) for at least one moving object (7).
10. The method according to claim 9, characterized in that the computer system (6) determines a direction of movement (10) of the moving object (7), in particular whether the direction of movement of the moving object (7) is directed into the security area (1) or out of the security area (1).
11. Experience according to one of claims 1 to 7, characterized in that the computer system (6) outputs at least one image (8) for the release of the security area (1) by an operator (11) and optionally outputs a warning signal (12) to the operator (11).
12. Experience according to claim 11, characterized in that the computer system (6) outputs to the operator (11) at least one image (8), in particular a video, which was recorded by the camera (4) shortly after it was determined by means of the further sensor (5) that an object (7) had penetrated into the security area (1), and at least one further image (8), in particular a current video transmission, which represents a current state.
13. Method according to one of the preceding claims, characterized in that the security area (1) has a plurality of access areas (3), each access area (3) being monitored by the computer system (6) with at least one camera (4) and at least one further sensor (5).
14. Method according to one of the preceding claims, characterized in that the further sensor (5) is a light barrier, a contact sensor, an ultrasonic sensor or a radar sensor.
15. System comprising a safety area (1), in particular a safety area (1) of a machine tool (2), a camera (4), a further sensor (5) and a computer system (6), wherein the safety area (1) has at least one access area (3) and access to the safety area (1) is only possible via the access area (3), wherein the camera (4) and the further sensor (5) are provided and configured to monitor the access area (3), wherein the camera (4) and the further sensor (5) are connected to the computer system (6), wherein the computer system (6) is provided and configured to monitor by means of the further sensor (5) whether an object (7) in the Security area (1), wherein the camera (4) is provided and set up to take at least one image (8) of the access area (3) when the computer system (6) detects by means of the further sensor (5) that an object (7) is entering the security area (1), characterized in that the computer system (6) is provided and set up to evaluate the image (8) recorded by the camera (4) only after it has been detected by means of the further sensor (5) that an object (7) has entered the security area (1).
16. System according to claim 15, characterized in that the system (1) is provided and set up to carry out a method according to the invention according to one of claims 1 to 14.