HK40020128B - A system and method for detecting inactive objects - Google Patents

Description

Systems and methods for detecting inactive objects

Technical Field

The present invention relates to a system and method for detecting inactive objects, and specifically, but not exclusively, to inactive or incapacitated human objects.

Background Technology

In the healthcare industry, significant resources are dedicated to monitoring individuals receiving care. This typically involves hiring additional staff to supervise those being cared for, including the elderly or those with medical conditions or physical disabilities.

Electronic devices have been invented to assist in the monitoring process, including the use of personal alarms and devices. More recently, with the development of wearable devices, users are now able to wear wearable devices to monitor their movements and issue alerts in conjunction with caregivers when the device detects any anomalies.

However, despite progress in this area, some people do not always want to carry wearable devices. This is especially true for older adults who are not accustomed to wearable technology or find such devices intrusive in their daily lives. In other cases, certain activities may also mean that wearable devices are unsuitable for the environments in which they are required to operate. Therefore, although these devices can often be helpful to users, older adults or others who require care may choose not to use them, resulting in gaps in the supervision and care of these users.

Summary of the Invention

According to a first aspect of the present invention, a system for detecting inactive objects is provided, comprising:

- A thermal imaging device configured to acquire the thermal characteristics of a moving object;

- A motion detection processor is configured to process the thermal features of the active object to monitor any movement of the active object, and to determine that the active object is inactive when the thermal features indicate that the movement of the active object is below a movement threshold.

In an embodiment of the first aspect, the motion detection processor continuously acquires and processes the thermal features of the active object during the detection phase.

In an embodiment of the first aspect, the acquired thermal features are divided into multiple parts, each of which is further processed to determine whether each part is dynamic or static.

In an embodiment of the first aspect, when the number of dynamic portions of the thermal feature is below a predetermined movement threshold, the active object associated with the thermal feature will be determined to be inactive.

In an embodiment of the first aspect, each of the plurality of portions is determined to be either dynamic or static by measuring any variance rate in the portions.

In an embodiment of the first aspect, if the variance rate exceeds a predetermined variance rate within a predetermined time period, then each of the plurality of portions is determined to be dynamic.

In an embodiment of the first aspect, the thermal feature is an image or image stream acquired by a thermal optical device.

In an embodiment of the first aspect, the plurality of portions includes one or more pixels of an image or image stream.

In an embodiment of the first aspect, the motion detection processor processes a thermal image stream to establish the presence of the active object.

In an embodiment of the first aspect, the motion detection processor tracks the active object within each image of the thermal image stream and acquires new thermal features of the active object over a time period.

In an embodiment of the first aspect, the detection phase begins when the active object is first acquired within the thermal image stream.

In an embodiment of the first aspect, the active object is first acquired within the thermal image stream by comparing earlier thermal images with subsequent thermal images.

In an embodiment of the first aspect, the detection phase is terminated when the active object is detected to have left the thermal image stream.

In an embodiment of the first aspect, the departure of the active object from the thermal image stream is detected by comparing an earlier thermal image with a subsequent thermal image within the thermal image stream.

In the first aspect of the embodiment, the thermal optical device is a low-resolution thermal imager.

In an embodiment of the first aspect, the low-resolution thermal imager is configured to acquire the outline of an object having minimal detail of the object.

In an embodiment of the first aspect, the thermal optical device is positioned above the working area.

In an embodiment of the first aspect, the working area includes one or more heat sources.

According to a second aspect of the present invention, a method for detecting inactive objects is provided, comprising the following steps:

- Obtain the thermal characteristics of the active object;

- Process the thermal characteristics of the active object to monitor any movement of the active object, and determine that the active object is inactive when the thermal characteristics show that the movement of the active object is below a movement threshold.

In an embodiment of the second aspect, the step of processing thermal features to monitor any movement of the moving object includes continuously acquiring and processing the thermal features of the moving object during a detection phase.

In an embodiment of the second aspect, the acquired thermal features are divided into multiple parts, each of which is further processed to determine whether each of the multiple parts is dynamic or static.

In a second aspect embodiment, when the number of dynamic portions of the thermal feature is below a predetermined threshold, the active object associated with the thermal feature will be determined as inactive.

In an embodiment of the second aspect, each of the plurality of portions is determined to be either dynamic or static by measuring any variance rate in the portions.

In an embodiment of the second aspect, if the variance rate exceeds a predetermined variance rate within a predetermined time period, then each of the plurality of portions is determined to be dynamic.

In an embodiment of the second aspect, the thermal feature is an image or image stream acquired by a thermal optical device.

In an embodiment of the second aspect, the plurality of portions includes one or more pixels of an image or image stream.

In a second aspect embodiment, the step of acquiring the thermal features includes processing a thermal image stream to establish the presence of the active object.

In a second aspect embodiment, the step of acquiring the thermal features further includes tracking the active object within each image of the thermal image stream and acquiring new thermal features of the active object over a time period.

In an embodiment of the second aspect, the detection phase begins when the active object is first acquired within the thermal image stream.

In a second aspect embodiment, the active object is first acquired within the thermal image stream by comparing earlier thermal images with subsequent thermal images.

In a second aspect embodiment, the detection phase is terminated when the active object is detected to have left the thermal image stream.

In a second aspect embodiment, the departure of the active object from the thermal image stream is detected by comparing earlier thermal images and subsequent thermal images within the thermal image stream.

In a second aspect embodiment, the thermal optical device is a low-resolution thermal imager.

In an embodiment of the second aspect, the low-resolution thermal imager is configured to acquire the outline of an object having minimal detail of the object.

In a second aspect embodiment, the thermo-optical device is positioned above the working area.

In an embodiment of the second aspect, the working area includes one or more heat sources.

Attached Figure Description

The embodiments of the invention will now be described by way of example with reference to the accompanying drawings, in which:

Figure 1 shows a block diagram of an example embodiment of a system for detecting inactive objects;

Figure 2A shows a schematic diagram of an example deployment of the system in Figure 1 in a shower room;

Figure 2B shows a schematic diagram of another example deployment of the system in Figure 1 in a shower room;

Figure 3 shows a timeline of an example operation of the detection window used to detect inactive events;

Figure 4A shows an example of obtaining the original thermal features of an active object using the system in Figure 1;

Figure 4B shows the processed original thermal characteristics of Figure 4A; and

Figure 5 shows a flowchart of the example operation of the system in Figure 1;

Detailed Implementation

Referring to Figure 1, a block diagram of a system for detecting an inactive object 100 is shown, including a thermal imaging device 104 configured to acquire thermal features 116 of an active object; and a motion detection processor 106 configured to process the thermal features of the active object to monitor any movement of the active object, and to determine that the active object is inactive when the thermal features indicate that the movement of the active object is below a motion threshold.

In this exemplary embodiment, the system for detecting inactive objects 100 can be operated by an operator as a fall detection system for various users such as the elderly, those with physical disabilities, or those with mobility difficulties. System 100 can be configured to operate to detect whether user 116 has fallen in a particular workspace and thereby alert any caregiver, first aide, or anyone else who can provide assistance.

As shown in this example embodiment, system 100 includes a thermal imaging device 104, which may be a thermal imager configured to acquire thermal images. The thermal imaging device can then communicate with an electronic or computing device 106 configured to process the thermal images acquired by the thermal imaging device to determine if there are any inactive users, and thus indicate that the user may have fallen or otherwise become inactive. By processing the thermal images, the electronic computing device 106 can also function as a motion detection processor 106 configured to detect whether an object acquired in the thermal image is moving or not.

As shown in the figure, the motion detection processor 106 can be integrated into a single device 102, which also includes a thermal imaging device 104. Alternatively, the motion detection processor 106 can be implemented as a separate device configured to communicate with each other and with one or more thermal imaging devices 104. A computer server 108 can also be connected to the thermal imaging device 104 or the motion detection processor 106 to collect data from the thermal imaging device 104 and/or the motion detection processor 106. As those skilled in the art will understand, the motion detection processor 106 can also be implemented as software of the server 108, wherein raw thermal images acquired by the thermal imaging device 104 are transmitted to the server 108 for processing, depending on specific implementation requirements.

Preferably, the thermal imaging device 104 is connected to and controlled by an embedded computing device 106, which utilizes the thermal imaging device 104 as a single unit 102 for deployment as a motion detection device. The thermal imaging device 104 is configured to acquire a stream of thermal images of a work area 114 in which the user 116 may be located. Such a work area may include, for example, a toilet, rest area, shower area, kitchen, laundry room, sauna, steam room, or any other area requiring personnel monitoring. As the thermal imaging device 104 acquires the thermal image stream, these images are then processed by a motion detection processor 106 implemented in hardware, software, or a combination of both within the embedded computing device 106. Conversely, the single device 102 may be referred to as a motion detector device 102 and may be used as a fall detection device 102, as explained below, configured to detect whether the person 116 has fallen in the work area 114 and requires assistance. This can be implemented based on the premise that a person 116 who has fallen has become inactive and needs assistance. Therefore, a user 116 who has fallen but can get up within a short period of time may not necessarily be considered a fall event by the system 100.

Once a thermal image is acquired and transmitted to the motion detection processor 106 for processing, the motion detection processor 106 can then determine whether a specific object of interest has entered the thermal image stream and continue monitoring the thermal characteristics of that specific object. In the example of the system 100 for monitoring elderly people, the specific object would be person 116, who would appear in the thermal image stream when person 116 enters the field of view of the thermal imaging device 104. Once person 116 enters the field of view of the thermal imaging device 104, and person 116 radiates heat, thus allowing the thermal imager 104 to acquire the characteristics of person 116, the motion detection processor 106 will determine the presence of the thermal characteristics of person 116.

Once the thermal features of person 116 are first detected in the first thermal image, the motion detection processor 106 can continue to track that feature in subsequent frames of the thermal image stream, where each frame shows some form of motion or movement of person 116. Once it is determined that person 116 or object is moving or in motion by comparing the changes between the current image and previous images within the thermal image stream, the thermal features associated with that person 116 or object will continue to be monitored or tracked by the motion detection processor 106 and considered an active object until the person leaves the field of view of the thermal imaging device 104. As an example, when the tracked object is no longer within the thermal image stream, it is determined that person 116 has left the work area 114. The process of tracking the thermal features of active person 116 or object is particularly useful in environments 114 where multiple heat sources 118 exist, such as in shower rooms, saunas, kitchens, and steam rooms. In these environments, inactive heat sources can also emit heat and thus have their own thermal characteristics, but by tracking only active characteristics, which may follow specific characteristics of people or other objects of interest, system 100 can issue only the thermal characteristics that are relevant to the need for an alert, thereby reducing false alarms and minimizing the effort of supervisors or caregivers.

During this monitoring or tracking process, the thermal characteristics of the active object 116 are continuously processed, with each thermal characteristic compared to the subsequent thermal characteristics. If the comparison shows a change between thermal characteristics at a specific time difference and these changes meet a minimum threshold, the active object is considered active when the object has moved over a period of time. However, if the comparison of two thermal characteristics within a specific time period shows no change or only a minimal change (below the threshold), it can indicate that the object 116 has become inactive, and therefore an alarm can be triggered, either by communicating with server 108 or other external device 112, to call for assistance from other parties.

As illustrated in this exemplary embodiment, a work area 114, such as various facilities within an aged care home, may have a central server 108 or computer system that can communicate with multiple motion detection devices 102 installed throughout the aged care facility. Each of these devices 102 can communicate with the server 108, which can record any images (preferably a low-resolution thermal imager as device 102 if privacy may not be a significant concern in some deployments, or if privacy may be an issue), and any data obtained from these devices into a database 110 for logging or alerting purposes. The server 108 may also be accessed by other computing devices 112, either near or remotely from the server 108, for access to information on the server 108 and for monitoring the operation of the motion detection devices 102.

In the event that person 116 has fallen, such as in a shower facility, motion detection device 102 will recognize that person 116 has suddenly become inactive because person 116 has fallen and cannot move. In this example, a signal can then be transmitted to server 108, which will record the signal and trigger an alarm on the caregiver's user interface 112 (speaker, light) or smart device 112 to alert them to the fall and call for assistance from the fallen user.

In another example embodiment of the system for detecting inactive objects 100, the system 100 is used to detect a user 116 who has fallen within one or more specific spaces. The system can be deployed to include one or more motion detection devices 102, each acting as a fall detector to detect whether the user 116 has fallen or become inactive in one or more spaces. Preferably, the fall detector 102 should use a thermal imager 104 with low resolution (e.g., no more than 40 × 30 pixels) and an embedded computing device 106 to process the acquired data by the thermal imager 104 before sending the signal to a server 108.

As illustrated in various examples, data processing can include inactivity identification to detect fall events experienced by a user. Data processing can also include an environmental perception detection process to distinguish people from other hot or heat-generating objects to reduce false alarms. This is particularly useful because thermal imagers can also capture other heat-generating non-human objects, such as shower heads, steam jets, or boilers, whose activity is not a concern; therefore, by using an environmental perception detection process, the inactivity of these objects will not trigger necessary alarms. An example of an environmental perception detection process is further described below with reference to Figure 3.

Conversely, the processing results, such as signals indicating a fall event, are sent to server 108 along with other information (such as the location, time, and other information of the fall detector) via a communication network such as a wired/wireless network.

In some examples, images acquired by the thermal imager may not be emitted by the fall detector 102 to prevent privacy-sensitive images from being stored or transmitted by any means. This is particularly advantageous in the deployment of system 100 for fall detection in a shower room, where user 116 may be naked or otherwise engaging in private activities. The low-resolution thermal imager 104, along with the proper handling and non-transmission of the images, helps ensure that the user's privacy or personal space is not violated.

Conversely, each fall detector 102 can also connect to a server 108 to further communicate with other user portable devices 112 (e.g., smartphones, tablets held by caregivers) or user interfaces 112 (which may include, but are not limited to, audio alarms, visual alarms, alarms sent to pagers, smartphones, or computers, security alarms, etc.). The server 108 can also be integrated into each fall detector 102, allowing each device 102 to operate as a server 108 to communicate with other fall detector devices 102 or user portable devices 112. Although preferably, in multiple deployments, the server 108 may be a separate network server 108 established and managed by a facility such as a hospital or nursing home.

Server 108 can also be configured to collect information sent from fall detectors and merge that information into database 110, whether local, remote, or cloud-based. Once a fall event is received from any fall detector 102, relevant information can be extracted from database 110 and sent to user interface 112 for further action.

Referring to Figures 2A and 2B, an example of a motion detection device installed in a suitable location within a shower room is shown. As previously described, a suitable deployment of a system for detecting inactive objects 100 is provided for shower facilities 200, where users such as the elderly or those with mobility impairments may suffer falls or become incapacitated during accidents or medical emergencies.

As shown in this exemplary embodiment, the motion detection device 202 can be placed in the top region, tilted downwards at an angle (0) 204, to primarily capture the minimum portion of the person's head and upper body. Preferably, the motion detection device 202 has two mounting positions, in front of the person 200F or behind the person (facing the person's back) 200R, or, if desired, in both positions as required.

These installation locations 200F and 200R are advantageous because placing the device 202 in the top area prevents water from spilling onto the device, thus eliminating the need for strong waterproofing. This ensures that the device can be constructed as inexpensively as possible.

Furthermore, limiting the acquisition of thermal images of users to only the head and minimal upper body, and displaying minimal visual detail of individuals when they are in their private space (e.g., naked while showering), helps to avoid intrusion into users' personal space and reduces any privacy issues they may experience.

Referring to Figure 3, a timeline 300 is shown, illustrating the activation of the detection phase 302 or detection window of the system for detecting inactive object 100, and thus illustrating an example of an environment-aware detection process designed to minimize any errors. In this embodiment, the detection phase 302 is a finite amount of time or window during which a warning or alarm 304 is triggered if the detection of an active object becomes inactive. In this embodiment, if the motion detection processor determines that an active object has become inactive 304, but this event occurs outside the detection phase or window 302, no alarm is issued, and the detection is ignored. This is because the environment of the alarm is not of system concern for detecting inactive object 100.

This exemplary embodiment is particularly advantageous in shower rooms, steam rooms, saunas, or any other room or area with more than one heat source. In these areas or rooms, hot objects such as hot water, hot showerheads, boilers, hot coal, steam vents, etc., are typically found. Conversely, because system 100 relies on processing thermal characteristics, these hot objects may trigger inactive events that are actually false alarms or unnecessary. This is especially true when these objects are moving objects (e.g., showerheads) or when there are temperature variations, such as in the case of steam vents, which may operate periodically to maintain the temperature of the steam room.

To filter out these false alarms in order to minimize the number of false alerts or alarms, an environment-aware detection process can be used to define and run a detection window 302 for detecting inactive objects 100 in the system when the detection environment 304 is relevant to the purpose of the system.

In one embodiment, detection phase 302 is initiated or activated when the motion detection processor 106 first detects that object 116 has entered work area 114 (e.g., a shower area or steam room). This can be performed by detecting the thermal characteristics of the moving person or object to initiate the detection phase. Detection of a person moving into work area 114 will be flagged as an environmental change, indicating that the active person now needs to be monitored. Conversely, detection phase 302 is deactivated when a moving person 116 or object is detected leaving work area 114. Therefore, other changes in heat sources that are part of work area 114 will not trigger any unnecessary alerts or alarms, as the environment will indicate that no relevant person or object is present in work area 114.

Referring to Figures 4A and 4B, the results show the effects of a motion detection processor processing the thermal features of an active object to detect the active object and monitor and identify any movement of the active object. In this example embodiment, a pixel-level monitoring process is used to perform inactivity identification to detect inactive persons who may have suffered a fall or become inactive.

In this example, a pixel-level monitoring process is used to detect inactivity events by tracking the values of each part of the thermal features associated with the active object, such as clusters of pixels associated with the active object, or preferably in the case of low-resolution images. If a pixel value remains stable for a certain period of time, the pixel is considered invalid. If the ratio of an inactive pixel to the total pixels of the foreground object is higher than a certain threshold, it is considered an inactive event. Therefore, mathematically, this process can be represented as follows:

1- Compare each pixel value of the foreground object (e.g., an active person) fg(x,y) at time t with its average value.

2. For a user-defined period T, if the difference between fg(x,y) and sfg(x,y) is less than the user-defined variance σ, the pixel value is considered a static pixel sfg(x,y). Therefore, no change is recorded for this pixel.

3- If the ratio of the total number of static pixels ∑sfg(x,y) to the total number of pixels of the foreground object ∑fg(x,y) is greater than the user-defined threshold R, then the foreground object (active person) is determined to be inactive.

As a result of this detection process, the foreground object 402 within the thermal image 400 will be processed for motion detection, meaning that all other objects in the background 404 that do not have substantial thermal features will be ignored. Furthermore, existing thermal objects within the working area can be dynamic or static, but since these may not be considered foreground objects 402 (as they may be users or objects that need to be monitored), their state will not affect the detection of the activity or inactivity state of the foreground object 402.

The way this example detection process is performed is advantageous because it can operate with low-resolution thermal images. Detecting other forms of inactivity requires analyzing motion trajectories. This is difficult in a shower room for several reasons, including:

1. In our design, from the perspective of the installed camera, the trajectory of the fall is unrelated to normal movement; and

2- Motion tracing is very noisy because shower areas are typically very small, thus potentially requiring high-resolution thermal imaging. However, obtaining high-resolution details of users in shower rooms or steam rooms where privacy is required is highly undesirable.

Referring to Figure 5, a flowchart 500 is shown for how a system for detecting inactive objects 100 is operated according to an operational example. In this example, a thermal imaging device (e.g., a thermal imager) will continuously acquire images (502) to identify any thermal features of active objects or people. This invokes an environmental awareness detection process (504), which can be expressed as the operation of a state machine. In this example implementation, the state machine may include four states. These states are as follows:

1- Idle State: If no foreground object is detected, the system will remain in the idle state.

2. Decision State: The system will observe the foreground object for a period of time using an inactive object recognition algorithm. If no foreground object is observed, it will jump to the idle state. If there is activity in the foreground object, it will jump to the active state. If there is no activity in the foreground object, it will jump to the inactive state.

3-Active State: A foreground object has been detected. The detection window remains open in this state, and any inactive event occurring in this state will be considered a fall event; and

4-Inactive State: A foreground object is detected, but the detection window is closed. Any inactive events occurring in this state will not be considered fall events.

In this example embodiment, to change between idle, active, and inactive states, the state can transition to each other simply through a decision state. Several factors (one or a combination thereof) can trigger the state machine to enter a decision state. These factors, for example, occur when a person enters or leaves the shower area, and include, but are not limited to:

- The area of a foreground object can be magnified or reduced by spot analysis of its thermal features;

-The center of mass of the foreground object moves along the direction in which a person enters or leaves the shower area;

- The foreground object is located at the edge of the image; in the real world, it would be outside the shower area.

Conversely, if a foreground object associated with an active person is detected, a detection phase or detection window (505) is initiated or activated, and the motion detection processor continues to monitor the movement or motion in the hot signature features of the active object or person.

At this stage, if any inactivity is detected (506, 507), an alarm signal or other information can be sent to the server (508). The server can then store and retrieve information from the database (509) and send it to the user interface of a supervisor or portable device held by the caregiver or supervisor. If no inactivity is detected (507), the detection window is closed (512) when the active person leaves the work area (510), and the entire process restarts while waiting for the next person to enter the work area.

Although not strictly necessary, the embodiments described with reference to the accompanying drawings can be implemented as an application programming interface (API) or a set of libraries used by developers, or can be included in another software application, such as the operating system of a terminal or personal computer or the operating system of a portable computing device. Typically, since program modules include routines, programs, objects, components, and data files that assist in performing specific functions, those skilled in the art will understand that the functionality of a software application can be distributed across multiple routines, objects, or components to achieve the same functionality required herein.

It should also be understood that any suitable computing system architecture can be used where the methods and systems of the present invention are implemented entirely or partially by a computing system. This will include stand-alone computers, network computers, and dedicated hardware devices. When using the terms "computing system" and "computing device," these terms are intended to cover any suitable configuration of computer hardware capable of implementing the described functions.

Those skilled in the art will understand that various changes and/or modifications can be made to the invention shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. Therefore, the embodiments of the invention are to be considered illustrative rather than restrictive in all respects.

Unless otherwise stated, any reference to prior art contained herein should not be construed as an admission that the information is common general knowledge.

Claims (28)

1. A system for detecting when an active object becomes an inactive object, comprising: - A thermal imaging device configured to acquire thermal features associated with each of a plurality of moving objects; and - A motion detection processor configured to process the thermal characteristics of the active object to monitor any movement of the active object. Each acquired thermal feature is divided into multiple pixels, and each pixel is further processed to determine whether it is dynamic or static, and this further processing includes: The pixel value of each pixel is compared with the average pixel value of the plurality of pixels of the acquired thermal features of the active object; During a predetermined period, if the difference between the pixel value and the average pixel value is less than a predefined variance, then the pixel is determined to be a static pixel; and If the ratio of the total number of static pixels to the total number of pixels in the acquired thermal features of an active object is greater than a predefined threshold, then the active object is determined to be inactive. Among them, the multiple objects of the activities are people. 2. The system of claim 1, wherein the thermal imaging device and the mobile detection processor continuously acquire and process the thermal features of the moving object during the detection phase. 3. The system of claim 1, wherein monitoring any movement of the moving object comprises: tracking each of the thermal features in subsequent frames of the thermal image stream acquired by the thermal imaging device. 4. The system of claim 1, wherein the thermal imaging device is a thermal optical device, and each thermal feature is an image or image stream acquired by the thermal optical device. 5. The system of claim 1, wherein the motion detection processor processes the thermal image stream to establish the presence of the active object. 6. The system of claim 5, wherein the motion detection processor tracks the active object within each image of the thermal image stream and acquires new thermal features of the active object over a time period. 7. The system of claim 6, wherein the detection phase begins when the active object is first acquired within the thermal image stream. 8. The system of claim 7, wherein the active object is first acquired within the thermal image stream by comparing an earlier thermal image with a subsequent thermal image within the thermal image stream. 9. The system of claim 7, wherein the detection phase is terminated when the active object is detected to have left the thermal image stream. 10. The system of claim 9, wherein the departure of the active object from the thermal image stream is detected by comparing an earlier thermal image with a subsequent thermal image within the thermal image stream. 11. The system of claim 4, wherein the thermal optical device is a low-resolution thermal imager. 12. The system of claim 11, wherein the low-resolution thermal imager is configured to acquire the outline of an object having minimal detail of the object. 13. The system of claim 4, wherein the thermal optical device is disposed above the working area. 14. The system of claim 13, wherein the working area includes one or more heat sources. 15. A method for detecting when an active object becomes an inactive object, the method comprising the following steps: - Obtain the thermal features associated with each of the multiple active objects; and - Process the thermal characteristics of the moving object to monitor any movement of the moving object. Each acquired thermal feature is divided into multiple pixels, and each pixel is further processed to determine whether it is dynamic or static. This further processing includes: The pixel value of each pixel is compared with the average pixel value of the plurality of pixels of the acquired thermal features of the active object; During a predetermined period, if the difference between the pixel value and the average pixel value is less than a predefined variance, then the pixel is determined to be a static pixel; and If the ratio of the total number of static pixels to the total number of pixels in the acquired thermal features of an active object is greater than a predefined threshold, then the active object is determined to be inactive. Among them, the multiple objects of the activities are people. 16. The method of claim 15, wherein the thermal features are continuously acquired and processed during the detection phase. 17. The method of claim 15, wherein monitoring any movement of the active object comprises: tracking each of the thermal features in subsequent frames of the thermal image stream. 18. The method of claim 15, wherein each thermal feature is an image or image stream acquired by a thermal optical device. 19. The method of claim 15, wherein the step of acquiring the thermal features includes processing a thermal image stream to establish the presence of the active object. 20. The method of claim 19, wherein the step of acquiring the thermal features further comprises tracking the active object within each image of the thermal image stream and acquiring new thermal features of the active object over a time period. 21. The method of claim 20, wherein the detection phase begins when the active object is first acquired within the thermal image stream. 22. The method of claim 21, wherein the active object is first acquired within the thermal image stream by comparing an earlier thermal image with a subsequent thermal image within the thermal image stream. 23. The method of claim 22, wherein the detection phase is terminated when the active object is detected to have left the thermal image stream. 24. The method of claim 23, wherein the departure of the active object from the thermal image stream is detected by comparing an earlier thermal image within the thermal image stream with a subsequent thermal image. 25. The method of claim 18, wherein the thermal optical device is a low-resolution thermal imager. 26. The method of claim 25, wherein the low-resolution thermal imager is configured to acquire the outline of an object having minimal detail of the object. 27. The method of claim 18, wherein the thermal optical device is disposed above the working area. 28. The method of claim 27, wherein the working area includes one or more heat sources.