TWI697914B - Monitoring device and method thereof - Google Patents

Abstract

A monitoring system includes a camera, an identity recognition device, and a processing unit. The camera captures an image within a monitoring area. The identity identification device identifies an ID of a first tester from the image. The processing unit records a first walking path of the first tester in accordance with the image. When the processing unit detects that the first tester is masked or disappears from the image, and then detects a second tester that appears in the image, the processing unit start record a second walking path of the second tester. When the identity recognition device identifies that the second tester is the same as the first tester, the processing unit start records a third walking path. If the first, second and third walking path meet a specific criteria, the processing unit connects the first, second and third walking path.

Description

Monitoring system and method

The invention relates to a monitoring device, in particular to a walking path monitoring system and method.

According to research reports, if patients with early dementia can be treated with early intervention, the course of the disease can be delayed by about five years. If you ignore it or respond with a negative attitude, the patient's condition is likely to deteriorate rapidly in the next one or two years. At present, the very early dementia screening scale (AD8) is a simple, low-cost, convenient and fast screening method. However, if there is a method that can assist or replace the AD8 screening scale through an automatic judgment mechanism, such as the subjects’ usual walking behavior patterns, it may be able to further improve the accuracy, identify high-risk groups of dementia, and treat them earlier , To further control the patient's condition.

Existing personal track detection devices (such as smart watches, smart bracelets) can use GPS positioning to measure a person's moving path, but because GPS signals cannot penetrate indoors, they cannot be used for indoor tracking.

A monitoring system according to an embodiment of the present invention includes a camera for capturing an image in a monitoring area; a body recognition device for recognizing the identity of a first test subject from the image, And provide a first identifier according to the identity; and a processing unit; the processing unit records a first walking path of the first test subject according to the image, and the first walking path and the first identifier Correlation; when the processing unit detects that the first subject is occluded or disappears in the image, the recording of the first walking path is interrupted; when the processing unit detects a second subject Now in the image, start to record a second walking path of the second test subject, and the body identification device provides a second identifier to the second test subject appearing, the processing unit further performs the second test The walking path is associated with the second identifier; when the body identification device recognizes that the second subject is the first subject, the processing unit interrupts the recording of the second walking path and starts recording the first A third walking path of the test subject, and associate the third walking path with the first identifier; when the first, second, and third walking paths meet a specific condition, the processing unit applies the second The walking path is associated with the first identifier, and the first, second, and third walking paths are connected in time sequence to obtain a complete walking path of the first test subject.

The monitoring system as described above, wherein the specific conditions that the first, second and third walking paths meet, including the position and time of the first subject at the end of the first walking path, can be compared with The position and time at the starting point of the second walking path match; and the position and time of the first test subject at the end of the second walking path may be the same as at the beginning of the third walking path Match the location and time.

The monitoring system as described above, wherein, when the position difference between the end of the first walking path and the starting point of the second walking path is less than or equal to a first threshold, and the first walking path The time difference between the end point of and the start point of the second walking path is less than or equal to a second threshold, and the processing unit determines that the first walking path of the first subject can match the second walking path.

The monitoring system as described above, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path is less than or equal to a first threshold, and the second walking path The time difference between the end point of and the start point of the third walking path is less than or equal to a second threshold, then the processing unit determines that the second walking path of the first test subject can match the third walking path.

The monitoring system as described above, wherein, when the position difference between the end of the first walking path and the starting point of the second walking path is greater than the first threshold, or the end of the first walking path The time difference between the start point of the second walking path and the second threshold is greater than the second threshold, the processing unit determines that the first walking path of the first subject does not match the second walking path, and will not The second walking path is associated with the first identifier.

The monitoring system as described above, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is greater than the first threshold, or the second walking path The time difference between the end point and the start point of the third walking path is greater than the second threshold, then the processing unit determines that the second walking path of the first subject does not match the third walking path, and will not The second travel path is associated with the first identifier.

The monitoring method according to an embodiment of the present invention is suitable for a monitoring system that uses a camera to capture an image in a monitoring area; the monitoring system uses a body recognition device to identify a first person to be tested from the image The identity is changed here, and a first identifier is provided according to the identity; the monitoring method includes: recording a first walking path of the first subject according to the image, and the first walking path and the Is associated with the first identifier; when it is detected that the first person under test is obscured or disappeared in the image, the record of the first walking path is interrupted; when a second person is detected to appear in the image When the image is in the image, start to record a second walking path of the second test subject, and the body recognition device provides a second identifier to the second test subject that appears, and further the second walking path and the first test subject Two identifiers are associated; when it is recognized that the second subject is the first subject, the recording of the second walking path is interrupted, and a third walking path of the first subject is recorded, and the The third walking path is associated with the first identifier; when the first, second, and third walking paths meet a specific condition, the second walking path is associated with the first identifier, and the The first, second, and third walking paths are connected to obtain a complete walking path of the first subject.

The monitoring method as described above, wherein the specific conditions that the first, second and third walking paths meet, including the position and time of the first subject at the end of the first walking path, can be compared with The position and time at the start point of the second walking path match; and the position and time of the first test subject at the end of the second walking path may be the same as the position and time at the start point of the third walking path Match the location and time.

The monitoring method as described above, wherein, when the position difference between the end of the first walking path and the starting point of the second walking path is less than or equal to a first threshold, and the first walking path The time difference between the end point of and the start point of the second walking path is less than or equal to a second threshold, then it is determined that the first walking path of the first test subject can match the second walking path.

The monitoring method as described above, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path is less than or equal to a first threshold, and the second walking path The time difference between the end point of and the start point of the third walking path is less than or equal to a second threshold, then it is determined that the second walking path of the first test subject can match the third walking path.

The monitoring method as described above, wherein, when the position difference between the end of the first walking path and the starting point of the second walking path of the first subject is greater than the first threshold, or the end of the first walking path If the time difference between the start point of the second walking path and the second threshold is greater than the second threshold, it is determined that the first walking path of the first subject does not match the second walking path, and the second walking path will not be matched. The walking path is associated with the first identifier.

The monitoring method as described above, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is greater than the first threshold, or the second walking path The time difference between the end point and the start point of the third walking path is greater than the second threshold, it is determined that the second walking path of the first subject does not match the third walking path, and the second walking path will not be matched. Two walking paths are associated with the first identifier.

Figure 1 is a block diagram of a monitoring system according to an embodiment of the present invention. As shown in FIG. 1, the monitoring system 100 includes a camera 102, a body recognition device 104, and a processing unit 106. The camera 102 is used to capture an image 108 in a monitoring area. For example, the camera 102 may be various types of cameras, such as a charge-coupled device (CCD) sensor or a complementary metal-oxide-semiconductor (Complementary Metal-Oxide-Semiconductor: CMOS) sensor. Various types of cameras. In this embodiment, the number of cameras 102 is not limited to 1. Based on different monitoring applications, multiple cameras may also be used to capture images of the monitoring area, thereby expanding the scope of the monitoring area.

The identity recognition device 104 recognizes an identity of a person to be tested from the image 108 in the monitoring area, and provides an identifier 110 according to the identity. The body recognition device 104 includes various devices that can recognize the identity of the target person from the image, such as a face recognition device, a human body recognition device, a walking posture recognition device, etc. The processing unit 106 can be a processor of any desktop computer, notebook computer, server, or smart phone. By executing an application program stored in a storage device, the tester is placed in the monitoring area. A walking path is recorded in the storage device, and the walking path is associated with the identifier 110 provided by the body identification device 104.

Figure 2 is a schematic diagram of the walking path of a person under test monitored by the monitoring system in Figure 1 of the embodiment of the present invention. As shown in FIG. 2, in the image 108 captured by the camera 102, the body recognition device 104 recognizes a first subject 200 and a third subject 202. The first test subject 200 and the third test subject 202 respectively walk along the direction of the marked arrow, and in the process, the third test subject 202 obscures the first test subject 200 in the image 108, so that the first The subject 200 disappears in the image 108 for a short time. First, the processing unit 106 records a first walking path 210 of the first subject 200 according to the image 108. Since the body identification device 104 has recognized the first test subject 200, a first identifier A is provided to mark the first test subject 200; and a third identifier B is provided to mark the third test subject 200者202. Next, the processing unit 106 associates the first walking path 210 with the first identifier A, and confirms that the first walking path 210 is the walking path of the first subject 200.

When the processing unit 106 detects that the first subject 200 is hidden or disappears in the image 108 (for example, the third subject 202 is hidden or disappears due to other obstacles), the processing unit 106 interrupts the first walk Path 210 records. In another embodiment, the processing unit 106 may disappear in the image 108 because the first subject 200 has left the monitoring area, thereby interrupting the recording of the first walking path 210. After the first subject 200 and the third subject 202 are interlaced, the first subject 200 reappears in the image 108. However, at this time, the body discrimination device 104 has not recognized that the newly-appearing test subject is the first test subject 200, so when the first test subject 200 reappears in the image 108, the processing unit 106 regards it as a detective. A second test subject 204 (different from the first test subject 200) is detected, and a second walking path 212 of the second test subject 204 is recorded, and the body identification device 104 provides a second identifier (C ) To the second candidate 204 who appears. Wherein, the second identifier (C) represents an unknown identifier. The processing unit 106 further associates the second walking path 212 with the second identifier (C), and confirms that the second walking path 212 is the walking path of the second subject 204.

When the body identification device 104 recognizes that the second test subject 204 and the first test subject 200 are the same person, the processing unit 106 interrupts the recording of the second walking path 212 and starts to record a third test of the first test subject 200. The walking path 214 and the third walking path 214 are associated with the first identifier A. Then, the processing unit 106 starts to compare whether the first walking path 210, the second walking path 212, and the third walking path 214 meet a specific condition, which is used as whether to associate the second walking path 212 with the first identifier A Basis. For example, the position and time when the first subject 200 is at the end of the first walking path 210 can match the position and time at the beginning of the second walking path 212, and the first subject 200 200 (it is known that the second test subject 204 and the first test subject 200 are the same person) The position and time at the end of the second walking path can be the same as the position and time at the beginning of the third walking path 214 If they match, the processing unit 106 determines that the specific condition is met, and associates the second walking path 212 with the first identifier A, and confirms that the second walking path 212 is the walking path of the first subject 200.

Furthermore, when the position difference between the end of the first walking path 210 and the starting point of the second walking path 212 of the first subject 200 is less than or equal to a first threshold (for example, 50 cm or 1 meter), and The time difference between the end point of a walking path 210 and the start point of the second walking path 212 is less than or equal to a second threshold (for example, 1 second or 3 seconds), the processing unit 106 determines the first walking path of the first subject 200 210, can be matched with the second walking path 212. The first threshold and the second threshold can be adjusted according to different application conditions and are not limited to the numerical values exemplified above. When the position difference between the end point of the second walking path 212 and the start point of the third walking path 214 of the first subject 200 is less than or equal to the first threshold, and the end point of the second walking path 212 and the third walking path 214 The time difference between the starting points is less than or equal to the second threshold, the processing unit 106 determines that the second walking path 212 of the first subject 200 can match the third walking path 214.

When the first walking path 210 of the first subject 200 can be matched with the second walking path 212, and the second walking path 212 can be matched with the third walking path 214, the processing unit 106 determines that the specific condition is met, The second walking path 212 is associated with the first identifier A, and it is confirmed that the second walking path 212 is the walking path of the first subject 200. Then, the processing unit 106 connects the first walking path 210, the second walking path 212, and the third walking path 214 in time sequence to obtain a complete walking path of the first subject 200.

Conversely, when the position difference between the end point of the first walking path 210 and the start point of the second walking path 212 of the first subject 200 is greater than the first threshold, and the end point of the first walking path 210 and the second walking path The time difference between the starting point of 212 is greater than the second threshold, the processing unit 106 determines that the first walking path 210 of the first subject 200 does not match the second walking path 212. When the position difference between the end of the second walking path 212 and the start of the third walking path 214 of the first subject 200 is greater than the first threshold, and the end of the second walking path 212 and the beginning of the third walking path 214 The time difference between the points is greater than the second threshold, the processing unit 106 determines that the second walking path 212 of the first subject 200 does not match the third walking path 214. Therefore, the processing unit 106 will not associate the second walking path 212 with the first identifier A.

Figure 3 is a flowchart of the monitoring method of the monitoring system in Figure 1 according to the embodiment of the present invention. As shown in FIG. 3, the processing unit 106 records a first walking path 210 of the first subject 200 according to the image 108, and associates the first walking path 210 with the first identifier A (S300). Then, when the processing unit 106 detects that the first subject 200 is hidden or disappears in the image 108, it interrupts the recording of the first walking path 210 (S302). When the processing unit 106 detects that a second subject 204 appears in the image 108, it starts to record a second walking path 212 of the second subject 204, and the body identification device 104 provides a second identifier (C ) For the second test subject 204 that appears, the processing unit 106 further associates the second walking path 212 with the second identifier (C) (S304). When the body identification device 104 recognizes that the second test subject 204 is the first test subject 200, the processing unit 106 interrupts the recording of the second walking path 212 and starts recording a third walking path 214 of the first test subject 200. And the third traveling path 214 is associated with the first identifier A (S306). Finally, when the first walking path 210, the second walking path 212, and the third walking path 214 meet a specific condition, the processing unit 106 associates the second walking path 212 with the first identifier A, and the second walking path 212 is associated with the first identifier A according to the time sequence. The first walking path 210 and the second walking path 212 are connected with the third walking path 214 to obtain a complete walking path of the first subject 200 (S308). Among them, the specific condition of step S308 has been described in paragraphs [0021]-[0022] of this specification, so it will not be repeated.

According to the monitoring device and the monitoring method of the present invention, a complete walking path of a target person can be known, and it can be judged whether there is an abnormal behavior pattern, such as swaying steps, winding paths, etc. In the future, this complete walking path can be further correlated with dementia behavior.

Although the embodiments of the present invention are as described above, we should understand that what is presented above is only an example, not a limitation. According to this embodiment, many changes of the above exemplary embodiment can be implemented without violating the spirit and scope of the invention. Therefore, the breadth and scope of the present invention should not be limited by the embodiments described above. More precisely, the scope of the present invention should be defined by the following patented scope and its equivalents.

100 ~ Monitoring system 102 ~ Camera 104 ~ Body recognition device 106 ~ Processing unit 108 ~ Image 110 ~ Identifier 200 ~ First test subject 202 ~ Third test subject 204 ~ Second test subject 210 ~ First walking path

212: Second walking path

214: Third walking path

ID: A: first identifier

ID: B: third identifier

ID: C: second identifier

Figure 1 is a block diagram of a monitoring system according to an embodiment of the present invention. Figure 2 is a schematic diagram of the walking path of a person under test monitored by the monitoring system in Figure 1 of the embodiment of the present invention. Figure 3 is a flowchart of the monitoring method of the monitoring system in Figure 1 according to the embodiment of the present invention.

200: first candidate

202: The third candidate

204: The second candidate

210: First walking path

212: Second walking path

214: Third walking path

ID: A: first identifier

ID: B: third identifier

ID: C: second identifier

Claims (12)

A monitoring system includes: a camera for capturing an image in a monitoring area; a body recognition device for recognizing an identity of a first person to be tested from the image, and providing a first person according to the identity An identifier for marking the first subject; and a processing unit; the processing unit records a first walking path of the first subject according to the image, and the first walking path and the first Identifier associated; when the processing unit detects that the first test subject is occluded or disappears in the image, the recording of the first walking path is interrupted; when the processing unit detects a second test subject When the person appears in the image, a second walking path of the second testee is recorded, and the body identification device marks a second identifier for the second testee who appears, and the processing unit further The second walking path is associated with the second identifier; when the body identification device recognizes that the second subject is the first subject, the processing unit interrupts the recording of the second walking path and starts recording the A third walking path of the first test subject, and the third walking path is associated with the first identifier; when the first, second, and third walking paths meet a specific condition, the processing unit will The second walking path is associated with the first identifier, and the first, second, and third walking paths are connected in time sequence to obtain a complete walking path of the first subject. Such as the monitoring system described in item 1 of the scope of patent application, wherein the item 1. The specific conditions that the second and third walking paths meet, including the position and time of the first test subject at the end of the first walking path, can be compared with those at the beginning of the second walking path The position and time match; and the position and time of the first test subject at the end of the second walking path can match the position and time at the start point of the third walking path. The monitoring system described in item 2 of the scope of patent application, wherein, when the position difference between the end of the first walking path and the starting point of the second walking path of the first subject is less than or equal to a first threshold, And the time difference between the end point of the first walking path and the starting point of the second walking path is less than or equal to a second threshold, the processing unit determines that the first walking path of the first test subject can be compared with the first walking path Match the two walking paths. The monitoring system described in item 2 of the scope of patent application, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is less than or equal to a first threshold, And the time difference between the end point of the second walking path and the starting point of the third walking path is less than or equal to a second threshold, then the processing unit determines that the second walking path of the first subject can be compared with the first Match the three walking paths. The monitoring system described in item 3 of the scope of patent application, wherein when the position difference between the end of the first walking path and the starting point of the second walking path of the first subject is greater than the first threshold, or The time difference between the end point of the first walking path and the starting point of the second walking path is greater than the second threshold, the processing unit determines that the first walking path and the second walking path of the first subject do not correspond to each other Match without associating the second walking path with the first identifier. The monitoring system described in item 4 of the scope of patent application, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is greater than the first threshold, or The time difference between the end point of the second walking path and the start point of the third walking path is greater than the second threshold, and the processing unit determines that the second walking path of the first subject is different from the third walking path. Match without associating the second walking path with the first identifier. A monitoring method is applicable to a monitoring system that uses a camera to capture an image in a monitoring area, the monitoring system uses a body recognition device to recognize an identity of a first person to be tested from the image, and According to the identity, a first identifier is provided for marking the first test subject; the monitoring method includes: recording a first walking path of the first test subject according to the image, and the first walking path Is associated with the first identifier; when it is detected that the first person to be tested is obscured or disappeared in the image, the recording of the first walking path is interrupted; when a second person to be tested is detected Now in the image, start to record a second walking path of the second test subject, and mark a second identifier for the second test subject that appears, and further the second walking path and the second test subject Identifier associated; when the second test subject is identified as the first test subject, the recording of the second walking path is interrupted, and a third walking path of the first testee is recorded, and the The third walking path is associated with the first identifier; when the first, second and third walking paths meet a specific condition, the second walking path is associated with the first identifier, and the First, second and third lines The walking paths are connected to obtain a complete walking path of the first candidate. The monitoring method according to item 7 of the scope of patent application, wherein the specific conditions that the first, second, and third walking paths meet include the first test subject at the end of the first walking path The position and time can match the position and time at the beginning of the second walking path; and the position and time of the first test subject at the end of the second walking path can be the same as those at the third walking path. The location and time of the starting point of the path match. The monitoring method described in item 8 of the scope of patent application, wherein when the position difference between the end of the first walking path and the starting point of the second walking path of the first subject is less than or equal to a first threshold, And the time difference between the end point of the first walking path and the starting point of the second walking path is less than or equal to a second threshold, it is determined that the first walking path of the first subject can be compared with the second walking path Match. The monitoring method described in item 8 of the scope of patent application, wherein, when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is less than or equal to a first threshold, And the time difference between the end point of the second walking path and the start point of the third walking path is less than or equal to a second threshold, it is determined that the second walking path of the first subject is comparable to the third walking path Match. The monitoring method described in item 9 of the scope of patent application, wherein when the position difference between the end of the first walking path and the starting point of the second walking path of the first subject is greater than the first threshold, or The time difference between the end point of the first walking path and the starting point of the second walking path is greater than the second threshold, it is determined that the first walking path of the first subject does not match the second walking path, and Will not associate the second walking path with This first identifier is associated. The monitoring method described in item 10 of the scope of patent application, wherein when the position difference between the end of the second walking path and the starting point of the third walking path of the first subject is greater than the first threshold, or The time difference between the end point of the second walking path and the starting point of the third walking path is greater than the second threshold, then it is determined that the second walking path of the first subject does not match the third walking path, It does not associate the second walking path with the first identifier.

Images