A tumble detection system, method and storage medium

Abstract

The application discloses a kind of rollover detection system, method and storage medium, comprising: image acquisition module, image processing module, rollover detection subsystem and alarm module;Image acquisition module is used to obtain the infrared light signal of the object to be monitored and the analog electric signal of infrared light signal;Image processing module is used to process infrared light signal and the analog electric signal of infrared light signal, obtains digitized infrared thermal image;Rollover detection subsystem is used to determine rollover detection result according to infrared thermal image to the object to be monitored and is rolled over detection;Alarm module is used to generate alarm information when rollover detection result is rollover event occurrence, alarm information is pushed to associated application, and alarm information and rollover image are synchronized to cloud end.The above technical scheme can realize the non-contact rollover detection of the object to be monitored, reduce the error of rollover detection, improve the accuracy of rollover detection, and ensure the safety of the object to be monitored.

Description

Technical Field

[0001] This invention relates to the field of home safety and protection technology, and in particular to a rollover detection system, method and storage medium. Background Technology

[0002] In the early stages of infant development, if an infant rolls or lies prone on the floor after drinking milk, it is easy for them to spit up. The vomit can cause airway obstruction and even lead to suffocation. Therefore, it is clear that infants need constant care from their guardians. However, guardians cannot be with their infants every moment. Thus, a method and system for monitoring infant rolling over is needed for home safety and protection.

[0003] Currently, existing methods and systems for detecting whether an infant is rolling over usually determine whether the infant is rolling over based on the coordinates of the infant's head position, mouth position, and nose position. However, the coordinates between the mouth and nose are very close, which can easily lead to misjudgment of whether the infant is rolling over, causing caregivers to be unable to detect dangerous situations in time, resulting in low protection for the infant's safety. Summary of the Invention

[0004] This invention provides a roll detection system, method, and storage medium that enables non-contact roll detection of the object to be monitored, reduces roll detection errors, improves roll detection accuracy, and ensures the safety of the object to be monitored.

[0005] In a first aspect, embodiments of this disclosure provide a rollover detection system, including: an image acquisition module, an image processing module, a rollover detection subsystem, and an alarm module;

[0006] The image acquisition module is used to acquire the infrared light signal of the object to be monitored and the analog electrical signal of the infrared light signal, and transmit the infrared light signal and the analog electrical signal of the infrared light signal to the image processing module.

[0007] The image processing module is used to process the infrared light signal and the analog electrical signal of the infrared light signal to obtain a digitized infrared thermal image, and transmit the digitized infrared thermal image to the roll detection subsystem.

[0008] The tumble detection subsystem is used to perform tumble detection on the object to be monitored based on the digitized infrared thermal image, and determine the tumble detection result of the object to be monitored.

[0009] The alarm module is used to generate a roll alarm message when the roll detection result indicates that a roll event has occurred, push the roll alarm message to the associated application, and synchronize the roll alarm message and the roll image to the cloud.

[0010] Secondly, embodiments of this disclosure provide a rollover detection method, including:

[0011] Acquire the infrared light signal of the object to be monitored and the analog electrical signal of the infrared light signal;

[0012] The infrared light signal and the analog electrical signal of the infrared light signal are processed to obtain a digitized infrared thermal image.

[0013] Based on the digitized infrared thermal image, roll detection is performed on the object to be monitored, and the roll detection result of the object to be monitored is determined.

[0014] When the roll detection result indicates that a roll event has occurred, a roll alarm message is generated, the roll alarm message is pushed to the associated application, and the roll alarm message and the roll video are synchronized to the cloud.

[0015] Thirdly, embodiments of this disclosure provide a computer-readable storage medium storing computer instructions that are used to cause the roll detection system to implement the roll detection method provided in the second aspect of the embodiments described above.

[0016] An embodiment of the present invention provides a roll detection system, method, and storage medium, comprising an image acquisition module, an image processing module, a roll detection subsystem, and an alarm module. The image acquisition module is used to acquire infrared light signals and analog electrical signals of the infrared light signals of a monitored object, and transmit the infrared light signals and analog electrical signals to the image processing module. The image processing module is used to process the infrared light signals and analog electrical signals to obtain a digitized infrared thermal image, and transmit the digitized infrared thermal image to the roll detection subsystem. The roll detection subsystem is used to perform roll detection on the monitored object based on the digitized infrared thermal image and determine the roll detection result of the monitored object. The alarm module is used to generate a roll alarm message when the roll detection result indicates that a roll event has occurred, push the roll alarm message to a related application, and synchronize the roll alarm message and the roll image to the cloud. The above technical solution enables non-contact tumble detection of the object to be monitored, reduces the error of tumble detection, improves the accuracy of tumble detection, and ensures the safety of the object to be monitored.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a structural block diagram of a roll detection system provided in Embodiment 1 of the present invention;

[0020] Figure 2 This is a structural block diagram of another roll detection system provided in Embodiment 1 of the present invention;

[0021] Figure 3 This is a flowchart of the body region cutting module involved in a rollover detection system provided in Embodiment 1 of the present invention;

[0022] Figure 4 This is an example illustration of the rectangular area corresponding to the body area of ​​the object to be monitored in a rollover detection system provided in Embodiment 1 of the present invention;

[0023] Figure 5 This is an example illustration of the rectangular area, head area, hand area, head position coordinates, hand position coordinates, and origin position coordinates involved in a rollover detection system provided in Embodiment 1 of the present invention.

[0024] Figure 6 This is a structural example block diagram of a roll detection system provided in Embodiment 1 of the present invention;

[0025] Figure 7 This is a flowchart of a tumble detection method provided in Embodiment 2 of the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," and "target," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] Example 1

[0029] Figure 1 This is a structural block diagram of a rollover detection system provided in Embodiment 1 of the present invention. This embodiment can be applied to situations where the object to be monitored is to detect whether it has rolled over. The rollover detection system is implemented in the form of a combination of hardware and software.

[0030] like Figure 1 As shown, the rollover detection system includes: an image acquisition module 11, an image processing module 12, a rollover detection subsystem 13, and an alarm module 14;

[0031] The image acquisition module 11 is used to acquire the infrared light signal and the analog electrical signal of the infrared light signal of the object to be monitored, and transmit the infrared light signal and the analog electrical signal of the infrared light signal to the image processing module 12.

[0032] Image processing module 12 is used to process infrared light signals and analog electrical signals of infrared light signals to obtain digital infrared thermal images and transmit the digital infrared thermal images to tumble detection subsystem 13.

[0033] The tumble detection subsystem 13 is used to perform tumble detection on the object to be monitored based on the digitized infrared thermal image and determine the tumble detection result of the object to be monitored.

[0034] The alarm module 14 is used to generate a roll alarm message when the roll detection result indicates that a roll event has occurred, push the roll alarm message to the associated application, and synchronize the roll alarm message and the roll image to the cloud.

[0035] In this embodiment, the image acquisition module 11 is connected to the image processing module 12, transmitting the acquired optical and electrical signals to the image processing module 12; the image processing module 12 is connected to the roll detection subsystem 13, transmitting the processed digital infrared thermal image to the roll detection subsystem 13 for roll detection; the roll detection subsystem is connected to the alarm module 14, sending the detection results of the roll detection of the monitored object to the alarm module 14; the alarm module 14 is connected to the roll detection subsystem 13, receiving the roll detection results sent by the roll detection subsystem 14, generating a roll alarm message when the roll detection result detected by the roll detection subsystem indicates that a roll event has occurred, pushing the roll alarm message to the associated application, and synchronizing the alarm message and the roll image to the cloud.

[0036] The image acquisition module 11 automatically acquires infrared light signals from the area where the monitored object is located when a roll detection requirement is generated. It converts the light signals into analog electrical signals and transmits both the infrared light signals and the converted analog electrical signals to the image processing module 12 for further image processing. The roll detection requirement can be a periodic cyclic acquisition requirement determined according to a pre-set cycle, or a manually triggered acquisition requirement by relevant personnel through a related application. The specific roll detection requirement is determined based on the actual situation, and this embodiment does not impose any limitations on it. It is understood that regardless of whether the ambient light is bright or dim, the image acquisition module 11 can obtain clear infrared light signals and converted analog electrical signals.

[0037] The image processing module 12 receives the infrared light signal and its analog electrical signal transmitted by the image acquisition module 11. Upon receiving the infrared light signal and analog electrical signal, to enable the roll detection subsystem 13 to more accurately detect the roll of the object under monitoring, it pre-processes the analog electrical signal of the infrared light signal with basic signal and image processing to obtain a digitized infrared thermal image with clearer features. Specifically, the analog electrical signal is converted into a digital electrical signal, and corresponding basic image processing, such as filtering, denoising, enhancement, or restoration, is performed based on the converted digital electrical signal to generate a digitized infrared thermal image. The processed digitized infrared thermal image is then transmitted to the roll detection subsystem 13.

[0038] The tumble detection subsystem 13 receives and stores the digitized infrared thermal image transmitted by the image processing module 12, which has undergone basic signal and image processing. It performs segmentation and feature extraction on the acquired infrared thermal image to determine the body region of the object to be monitored. Furthermore, it determines the head region, hand region, origin coordinates, head position coordinates, and hand position coordinates within the body region of the object to be monitored. Based on the positional changes of the head region and hand region of the object to be monitored in two adjacent cycles, it confirms the tumble detection result of the object to be monitored. After confirming the tumble detection result of the object to be monitored, it transmits the tumble detection result to the alarm module 14.

[0039] The alarm module 14 is used to receive the roll detection results sent by the roll detection subsystem 13 and perform corresponding operations based on the roll detection results. When the roll detection result indicates that no roll event has occurred, the alarm module 14 does not work; when the roll detection result indicates that a roll event has occurred, a roll alarm message is generated, the roll alarm message is pushed to the associated application, and the roll alarm message and roll image are synchronized to the cloud in a timely manner to remind the guardians and relevant care institutions of the monitored object, so that the guardians and relevant care institutions can respond to the roll event of the monitored object after receiving the roll alarm message and provide appropriate care for the monitored object in a timely manner.

[0040] This embodiment provides a roll detection system, including an image acquisition module, an image processing module, a roll detection subsystem, and an alarm module. The image acquisition module is used to acquire the infrared light signal and the analog electrical signal of the infrared light signal of the object to be monitored, and transmit the infrared light signal and the analog electrical signal of the infrared light signal to the image processing module. The image processing module is used to process the infrared light signal and the analog electrical signal of the infrared light signal to obtain a digitized infrared thermal image, and transmit the digitized infrared thermal image to the roll detection subsystem. The roll detection subsystem is used to perform roll detection on the object to be monitored based on the digitized infrared thermal image and determine the roll detection result of the object to be monitored. The alarm module is used to generate roll alarm information when the roll detection result indicates that a roll event has occurred, push the roll alarm information to related applications, and synchronize the roll alarm information and the roll image to the cloud. The above technical solution enables the image acquisition module to adaptively collect infrared light signals from the area where the monitored object is located under various environments. Before performing roll detection on the monitored object, the image processing module pre-processes the image, making the infrared thermal image input to the roll detection subsystem clearer. The roll detection subsystem monitors the position of the head and hand areas of the monitored object, determining the roll detection result based on the positional shift of the head and hands, effectively enhancing the accuracy and reducing the error of roll detection. When the roll detection result indicates a roll event has occurred, the alarm module can promptly generate an alarm message to notify the monitoring personnel of the monitored object, effectively improving the personal safety of the monitored object. This technical solution achieves non-contact roll detection of the monitored object, reducing roll detection errors, improving roll detection accuracy, and ensuring the safety of the monitored object.

[0041] As a first optional embodiment, based on the above embodiments, such as Figure 2 As shown, the image acquisition module 11 includes: a preprocessing module 111, a camera module 112, and an image sensor 113;

[0042] The preprocessing module 111 is used to assist the camera module 112 in acquiring light signals under dark conditions;

[0043] Camera module 112 is used to acquire light signals;

[0044] Image sensor 113 is used to converge light signals to form an infrared thermal image and convert the light signals into analog electrical signals.

[0045] In this embodiment, the preprocessing module 111 is connected to the camera module 112 to assist the camera module 112 in acquiring light signals under insufficient illumination conditions; the camera module 112 is connected to the image sensor 113 to transmit the acquired light signals to the image sensor 113; the image sensor 113 is connected to the camera module 112 to receive the light signals transmitted by the camera module 112 and complete the conversion between light signals and electrical signals.

[0046] The preprocessing module 111 is used to assist the camera module 112 in acquiring light signals under insufficient illumination conditions. It can be understood that the preprocessing module 111 can also be directly connected to the roll detection subsystem 13. The preprocessing module 111 senses the current ambient light conditions and sends the sensed ambient light conditions to the roll detection subsystem 13. The roll detection subsystem 13 controls the operation of the preprocessing module 111 and adjusts the camera module 112 so that the camera module 112 can work normally and collect light signals even in environments with insufficient illumination.

[0047] The camera module 112 is used to collect light signals from the area where the object to be monitored is located and to converge the collected light signals onto the image sensor 113.

[0048] Image sensor 113 is used to receive and aggregate the light signals collected by camera module 112, and convert the light signals into analog electrical signals. In traditional cameras, film is a photosensitive material that records the captured images after being treated with certain chemicals. In digital cameras, image sensor 113 replaces the film to form electronic images. Image sensor 113 can be a charge-coupled device (CCD) image sensor or a complementary metal-oxide-semiconductor (CMOS) sensor; this embodiment does not limit this.

[0049] Specifically, the image sensor 113 receives the light signal transmitted by the camera module 112, focuses it onto the CMOS / CCD, and converts the infrared light signal into an analog electrical signal. After completing the conversion between the infrared light signal and the analog electrical signal, the image sensor 113 transmits the infrared light signal and the analog electrical signal to the image processing module 12.

[0050] Furthermore, such as Figure 2 As shown, the preprocessing module 111 includes: an ambient light sensor, an infrared filter, and an infrared diode unit;

[0051] An ambient light sensor (ALS) is used to detect the illuminance of the environment. When the ambient light in the area where the object to be monitored is located dims, the ambient light sensor can detect the current illuminance of the environment and send a corresponding current signal to notify the tumble detection subsystem 13 according to the increase or decrease of the ambient light illuminance.

[0052] Infrared filters (IR-CUT) are used to automatically switch filters to reduce image noise in low-light conditions, thereby achieving the best image quality.

[0053] The infrared diode unit (IR-LED) is used to emit infrared light and receive infrared information in the infrared band emitted by the environment and objects illuminated by infrared light. Specifically, the infrared diode unit includes an infrared emitting diode (IR Transmitter LED) and an infrared receiving diode (IR Receiver LED). The infrared emitting diode is used to emit infrared light, and the infrared receiving diode is used to identify infrared information and transmit the identified infrared information to the camera module 112.

[0054] Specifically, since the image sensor 113 (CCD / CMOS) can detect most wavelengths of light, the colors output by the image sensor 113 differ from those seen by the naked eye due to the effects of various wavelengths of light. In dark or insufficient illumination conditions, the image sensor 113 cannot make full use of light, resulting in background snow noise in the sensed image information. Therefore, an ambient light sensor is needed to sense changes in illumination, convert the changes in illumination into current information, and transmit it to the roll detection subsystem 13 to determine whether the current environment is dark or insufficiently illuminated. The roll detection subsystem 13 then sends a command through the I2C interface to control the infrared filter to automatically switch filters to reduce noise caused by insufficient illumination, thereby achieving the best image effect.

[0055] When the ambient light sensor confirms that the illumination is very low or in the dark, the tumble detection subsystem 13 will activate the infrared emitting diode and the infrared receiving diode in the infrared diode unit. Since the human body emits heat, and heat in the microwave band is close to the color temperature of red, after the infrared emitting diode emits infrared light, the environment and objects in the area where the object to be monitored is located in the dark will also emit infrared bands accordingly. The infrared receiving diode receives the infrared information of these infrared bands and transmits the infrared information to the camera module 112.

[0056] As a second optional embodiment, based on the above embodiments, such as Figure 2 As shown, the image processing module 12 includes: an image capturing card 121, a filter 122, and a thermal image processor 123;

[0057] Image capture card 121 is used to convert analog electrical signals into digital electrical signals and perform digital image processing, and to store the converted digital electrical signals into roll detection subsystem 13.

[0058] Filter 122 is used to filter the analog electrical signal and transmit the filtered analog electrical signal to thermal image processor 123.

[0059] The thermal image processor 123 is used to convert the analog electrical signals of the infrared thermal image into digital electrical signals and transmit them to the roll detection subsystem 13.

[0060] In this embodiment, the image capturing card 121 is connected to the roll detection subsystem 13. After receiving the infrared light signal and the analog electrical signal of the infrared light signal from the image acquisition module 11, it stores the processed digital electrical signal (digital electrical signal) of the infrared light signal in the roll detection subsystem 13. The filter 122 is connected to the thermal image processor 123. After acquiring the analog electrical signal from the image acquisition module 11, it transmits the filtered analog electrical signal to the thermal image processor 123. The thermal image processor 123 is connected to the roll detection subsystem 13. After acquiring the analog electrical signal transmitted by the filter 122, it converts the analog electrical signal into a digital electrical signal and transmits it to the roll detection subsystem 13.

[0061] Image capture card 121 is used to acquire the analog electrical signal transmitted by image acquisition module 11, perform analog-to-digital conversion on the analog electrical signal to obtain a digital electrical signal, perform basic digital image processing on the digital electrical signal to generate a digitized infrared thermal image, and store the digitized digital information in the roll detection subsystem 13 so that the roll detection subsystem 13 can promptly call up the corresponding information and read the image for comparison when a roll detection requirement is generated.

[0062] Filter 122 is used to acquire the analog electrical signal transmitted by the image acquisition module 11, filter the analog electrical signal to remove unnecessary background noise, and transmit the filtered analog electrical signal to the thermal image processor 123.

[0063] The thermal image processor 123 is used to receive the filtered analog electrical signal transmitted by the filter, perform post-processing such as analog-to-digital conversion and color adjustment on the analog electrical signal, and then send the digital infrared thermal image generated by the processing to the roll detection subsystem 13.

[0064] As a third optional embodiment, based on the above embodiments, such as Figure 2As shown, the tumbling detection subsystem 13 includes: an image acquisition module 131, a body region segmentation module 132, a position information acquisition module 133, and a detection result determination module 134;

[0065] Image acquisition module 131 is used to acquire digitized infrared thermal images of the object to be monitored according to a preset period.

[0066] The body region cutting module 132 is used to cut out the body region of the object to be monitored based on the color temperature of the digitized infrared thermal image.

[0067] The location information acquisition module 133 is used to determine the head region and hand region of the object to be monitored from the body region of the object to be monitored, as well as the relevant location information of the object to be monitored; the relevant location information includes the origin position coordinates, head position coordinates and hand position coordinates;

[0068] The detection result determination module 134 is used to determine the tumbling detection result of the object to be monitored based on the changes in relevant location information in two adjacent cycles.

[0069] In this embodiment, the image acquisition module 131 is connected to the body region segmentation module 132. After acquiring the digitized infrared thermal image transmitted by the image processing module 12, it transmits it to the body region segmentation module 132. The body region segmentation module 132 is connected to the position information acquisition module 133. After the body region of the object to be monitored is segmented, the module locates the position coordinates of the object to be monitored based on the body region of the object to be monitored and obtains relevant position information. The detection result determination module 134 is connected to the position information acquisition module 133. It is used to perform roll detection on the object to be monitored based on the relevant position information of the object to be monitored.

[0070] The image acquisition module 131 is used to acquire digital electrical signals (digitized infrared thermal images) of the infrared light signals of the area where the monitored object is located from the image processing module 12 according to a preset periodic time. The preset periodic time can be one hour or one minute, and the specific time is determined according to actual needs; this embodiment does not impose any limitation on this. It is understood that the preset periodic time can be the same as the frequency of the analog electrical signals (analogous infrared thermal images) of the infrared light signals acquired by the image acquisition module 11, that is, the periodic frequency of the infrared thermal image information acquired by the image acquisition module 131 in the tumble detection subsystem 13 corresponds to that of the image acquisition module 11.

[0071] The image acquisition module 131 can also be understood as a module with a memory function, such as a memory. While acquiring the digitized infrared thermal image transmitted by the image processing module 12, it also stores it so that other modules in the roll detection subsystem 13 can call up the digitized infrared thermal image when performing roll detection.

[0072] The body region segmentation module 132 is used to segment the digitized infrared thermal image based on the color temperature of the digitized infrared thermal image acquired by the image acquisition module 131 to determine the body region of the object to be monitored. The method of segmenting the infrared thermal image can be a region growing algorithm or other methods, and this embodiment does not limit this.

[0073] Specifically, Figure 3 This is a flowchart illustrating the workflow of the body region segmentation module involved in a rollover detection system according to Embodiment 1 of the present invention. Figure 3 As shown, the body region segmentation module 132 processes the received digitized infrared thermal image after filtering and noise suppression according to a preset segmentation method (e.g., a region growing algorithm) to distinguish and segment the human-shaped region in the infrared thermal image from the surrounding environment. First, it identifies pixels that have not formed regions. Pixels with the same color temperature can be classified into the same region. It compares the largest group of pixels with the same color temperature in an NxN block, labeled T. This group of pixels with the same color temperature T is labeled as the average value Avg, and this group of pixels T is set as the initial region. It identifies the pixel groups around the initial region. Pixels with a color temperature close to the average value Avg and whose error is within a preset threshold of the largest pixel group T are added to region T. If a new region T' is formed when a preset number (e.g., 50) of pixels with the same color temperature are added to region T, the average value Avg is updated to the average value Avg' of all pixels in T'. Check if there are any more pixels that need to be added to the initial area. If there are no other pixels with a similar color temperature, they can be added to T'. At this point, the human-shaped area with a similar color temperature can be separated from the surrounding environment with a large color temperature difference. Update the initial image in memory every cycle (e.g., F minutes).

[0074] The location information acquisition module 133 is used to determine the head region and hand region from the body region of the object to be monitored, based on the color temperature, as well as the origin coordinates, head coordinates, and hand coordinates of the object to be monitored. The origin coordinates can be understood as the coordinates of the center position of the entire body region of the object to be monitored; the head coordinates can be understood as the coordinates of the center position of the head region of the object to be monitored; and the hand coordinates can be understood as the coordinates of the center position of the hand region of the object to be monitored. The center position can be the center of the region, the centroid, the orthocenter, etc., and this embodiment does not limit this.

[0075] It is understood that the head region and hand region and their related location information can be determined from the body region of the object to be monitored by a block comparison algorithm or other methods, and this embodiment does not limit this.

[0076] The detection result determination module 134 is used to confirm the roll detection result of the monitored object based on whether the changes in the origin position coordinates, head position coordinates, and hand position coordinates of the monitored object in two adjacent infrared thermal images are within a preset range. If the position changes of the relevant position information of the monitored object in two adjacent infrared thermal images are within the preset range, the roll detection result can be determined as no roll event has occurred; if the position changes of any relevant position information of the monitored object in two adjacent infrared thermal images exceed the preset range, the roll detection result can be determined as a roll event has occurred. After determining the roll detection result of the monitored object, the roll detection result is sent to the alarm module 14.

[0077] Optionally, the location information acquisition module 133 is specifically used for:

[0078] a1) Determine the rectangular area corresponding to the body area of ​​the object to be monitored.

[0079] Specifically, Figure 4 This is an example illustration of a rectangular area corresponding to the body region of the object to be monitored in a rollover detection system provided in Embodiment 1 of the present invention, as shown in the image. Figure 4 As shown, based on the human-shaped area cut out by the body area cutting module 132, a rectangular area with width (U) x length (E) corresponding to the body area of ​​the human-shaped figure is determined.

[0080] b1) Determine the center point of the object to be monitored based on the side length of the rectangular area, set the center point as the origin, and obtain the coordinates of the origin.

[0081] Specifically, based on the side lengths U and E of the rectangular region, the intersection of the line connecting the center points of the longer and shorter sides of the rectangular region is determined as the center point of the rectangular region, and this center point is also designated as the origin of the object to be monitored, with its coordinates set to (0,0). After determining the origin coordinates, coordinate axes are constructed in the horizontal and vertical directions to establish the four quadrants of the rectangular region.

[0082] c1) Determine the head and hand areas of the object to be monitored based on the color temperature in the rectangular area.

[0083] The object being monitored emits heat. From the perspective of infrared light, the heat emitted by the baby can be seen both day and night. The heat creates heat zones, and these heat zones have differences in temperature, thus forming a thermal image. The redder the thermal image seen by infrared light, the higher the temperature, such as the head, hands, and feet of the object being monitored. The bluer the thermal image seen by infrared light, the lower the temperature, such as background objects.

[0084] Specifically, based on the color temperature within the rectangular area, the red color temperature area is determined. Based on the size of each red color temperature area, the area corresponding to the head and the areas corresponding to the two hands are determined, i.e., the head area and the hand area.

[0085] d1) Determine the head position coordinates and hand position coordinates based on the origin position coordinates.

[0086] Specifically, Figure 5 This is an example illustration of the rectangular region, head region, hand region, head position coordinates, hand position coordinates, and origin position coordinates involved in a rollover detection system provided in Embodiment 1 of the present invention. Figure 5 As shown, based on the origin coordinates (0,0), the coordinates corresponding to the center positions of the head region and the hand region are determined, and they are respectively defined as the head position coordinates and the hand position coordinates. The hand position coordinates are HN(x1,y1) and HN(x2,y2); the head position coordinates are H(x3,y3).

[0087] Optional, such as Figure 2 As shown, the detection result determination module 134 includes: a displacement judgment unit, a first result determination unit, and a second result determination unit;

[0088] The displacement judgment unit is used to determine whether the displacements of the origin position coordinates and the head position coordinates in two adjacent cycles are both within the first preset range.

[0089] The first result determination unit is used to determine the roll detection result of the monitored object as a roll event when the displacement of the origin position coordinate and the head position coordinate in two adjacent cycles are both within a first preset range, and the hand position coordinates in two adjacent cycles are centrally symmetrical.

[0090] The second result determination unit is used to determine the roll detection result of the object to be monitored when at least one of the origin position coordinates and head position coordinates does not meet the first preset range in two adjacent cycles.

[0091] In this embodiment, the displacement judgment unit determines whether the displacements of the origin position coordinates and the head position coordinates in two adjacent cycles are both within a first preset range; if yes, the roll detection result is determined by the first result determination unit; if no, the roll detection result is determined by the second result determination unit.

[0092] The displacement judgment unit is used to determine whether the displacement of the origin position coordinates and the head position coordinates in the digitized infrared thermal images obtained in two adjacent cycles are within a first preset range. The first preset range can be understood as a preset offset range for the origin position coordinates and the head position coordinates, such as a distance range within half the width (U / 2) of a rectangular area.

[0093] For example, the first preset range is (U / 4~U / 2). It is determined whether the origin coordinates x,y = (0,0) of the rectangular area have moved to approximately half the body width, making the origin coordinates x,y = (U / 4~U / 2,0) or (-U / 2~-U / 4,0). If so, it can be determined whether the object being detected has rolled to the left or right. Simultaneously, it is also necessary to determine whether the head coordinates have moved to approximately half the body width, changing from H(x3,y3) to H(x3+U / 4~x3+U / 2,y3) or H(x3-U / 2~x3-U / 4,y3). Simultaneously determining whether the origin coordinates and head coordinates have moved effectively avoids misjudgments of rotation (but not rolling) where the head remains stationary while the body moves, thus improving the accuracy of roll detection.

[0094] The first result determination unit is used to determine that the roll detection result of the object to be monitored is a roll event when the displacement of the origin position coordinate and the head position coordinate in two adjacent cycles are both within a first preset range, and the hand position coordinates in two adjacent cycles are centrally symmetrical.

[0095] For example, if the coordinates of the origin change from x,y = (0,0) to (U / 4 ~ U / 2,0) or (-U / 2 ~ -U / 4,0), and the coordinates of the head change from H(x3,y3) to H(x3+U / 4 ~ x3+U / 2,y3) or H(x3-U / 2 ~ x3-U / 4,y3), then it is determined whether the coordinates of the hand have shifted, and whether the coordinates of the hand have changed from HN(x1,y1) to HN(-x1,-y1) or from HN(x2,y2) to HN(-x2,-y2). If the following conditions are met simultaneously: the origin coordinates change from x,y=(0,0) to (U / 4~U / 2,0) or (-U / 2~-U / 4,0), the head coordinates change from H(x3,y3) to H(x3+U / 4~x3+U / 2,y3) or H(x3-U / 2~x3-U / 4,y3), and the hand coordinates change from HN(x1,y1) to HN(-x1,-y1) or from HN(x2,y2) to HN(-x2,-y2), then the roll detection result of the monitored object is determined to be a roll event.

[0096] The second result determination unit is used to determine the roll detection result of the monitored object when the displacement of at least one of the origin position coordinates and the head position coordinates in two adjacent cycles does not meet the first preset range. Specifically, when the displacement of the origin position coordinates and the head position coordinates in two adjacent cycles of infrared thermal images cannot both meet the first preset range, it may be determined that the monitored object has undergone a rotation or displacement event. Therefore, the roll detection result of the monitored object is further determined by the far point position coordinates and the hand position coordinates.

[0097] Furthermore, the second result determination unit is specifically used for:

[0098] a2) If the displacement of at least one of the origin position coordinates and the head position coordinates in two adjacent cycles does not meet the first preset range, determine whether the origin position coordinates have moved within the second preset range and returned to the origin position.

[0099] In this embodiment, the second preset range can be understood as a preset offset range for the origin position coordinates, such as a distance range within a quarter width (U / 4) of a rectangular area.

[0100] For example, the second preset range is (U / 8~U / 4). When the displacement of at least one of the origin position coordinates and the head position coordinates in two adjacent cycles does not meet the first preset range, that is, when the displacement of the origin position coordinates and the head position coordinates cannot both meet the first preset range, in order to determine whether the monitored object rolls back to the origin after the body displacement, it is determined whether the origin position coordinates of the monitored object change from X,Y=(0,0) to (U / 8,0)~(U / 4,0) or (-U / 4,0)~(-U / 8,0) within a preset time (e.g., one hour), and then change back to X,Y=(0,0).

[0101] b2) If so, when the hand position coordinates are the same in two adjacent cycles, the roll detection result of the object to be monitored is determined to be a roll event.

[0102] Specifically, after determining that the origin position coordinates have moved within a second preset range and returned to the origin position within a preset time, it is determined whether the hand position coordinates have changed back to HN(x1,y1) or HN(x2,y2). If the hand position coordinates return to the original coordinates, that is, the hand position coordinates have changed back to HN(x1,y1) or HN(x2,y2), it can be determined that the roll detection result of the monitored object is a roll event; if the hand position coordinates have not changed back to HN(x1,y1) or HN(x2,y2), it can be determined that the roll event is not valid, and the roll detection process continues.

[0103] As a fourth optional embodiment, based on the above embodiments, such as Figure 2As shown, the alarm module 14 is connected to the roll detection subsystem 13 and includes: an emergency information call module 141, a cloud information transceiver module 142, and a speaker 143;

[0104] The cloud information transceiver module 141 is used to send and receive cloud information, which includes the roll image and roll alarm information of the object to be monitored.

[0105] Emergency information call module 142 is used to send cloud information to associated applications;

[0106] Speaker 143 is used to emit alarm sounds.

[0107] In this embodiment, the cloud information transceiver module 141, the emergency information call module 142, and the speaker 143 are all directly connected to the roll detection subsystem 13.

[0108] The cloud-based information transceiver module 141 is used to transmit and receive rollover images and rollover alarm information of the monitored object, and store them in the cloud. Among them, the rollover alarm information is an alert sent to the guardian of the monitored object when the rollover detection result indicates that a rollover event has occurred. It can be an alert message such as text and / or sound.

[0109] Emergency Information Call Module 142 is used to send cloud information to associated applications when the roll detection result indicates that a roll event has occurred, so as to issue a remote alarm prompt to the guardian of the monitored object.

[0110] Speaker 143 is used to emit an alarm sound when the rollover detection result indicates that a rollover event has occurred, so as to alert the guardian of the monitored object.

[0111] For example, to explain the roll detection system more clearly, the following is proposed: Figure 6 The diagram illustrates a structural example block diagram of a rollover detection system. The rollover detection subsystem 13 may include at least one processor and a memory communicatively connected to the at least one processor. The body region segmentation module, position information acquisition module, and detection result determination module can be understood as processors, and the image acquisition module can be understood as a memory. The memory also stores a computer program executable by the at least one processor, enabling the at least one processor to perform its corresponding functions. In the rollover detection subsystem 13, there may be three processors, each corresponding to one of the body region segmentation module, position information acquisition module, and detection result determination module, performing their respective functions; alternatively, there may be only one processor executing all the functions of the body region segmentation module, position information acquisition module, and detection result determination module. This embodiment does not impose any limitations on this approach.

[0112] Example 2

[0113] Figure 7 This is a flowchart of a roll detection method provided in Embodiment 2 of the present invention. It is applicable to situations where the object to be monitored is to detect whether it has rolled. The method can be executed by a roll detection system, which can be implemented in the form of a combination of hardware and software.

[0114] like Figure 7 As shown, the method includes:

[0115] S101. Acquire the infrared light signal and the analog electrical signal of the infrared light signal of the object to be monitored.

[0116] In this embodiment, when a roll detection requirement for a monitored object is generated, the image acquisition module of the roll detection system automatically collects infrared light signals from the area where the monitored object is located, converts the infrared light signals into analog electrical signals, and transmits the converted analog electrical signals and infrared light signals to the image processing module for further image processing. The roll detection requirement for the monitored object can be a periodic cyclical acquisition requirement determined according to a pre-set cycle, or it can be a acquisition requirement manually triggered by relevant personnel through a related application. The specific roll detection requirement is determined according to the actual situation, and this embodiment does not impose any limitations on it. It is understood that regardless of whether the ambient light is bright or dim, the image acquisition module can obtain clear infrared light signals.

[0117] S102. Process the infrared light signal and the analog electrical signal of the infrared light signal to obtain a digitized infrared thermal image.

[0118] In this embodiment, the image processing module receives the infrared light signal and its analog electrical signal transmitted by the image acquisition module. Upon receiving the infrared light signal and analog electrical signal, to enable the roll detection subsystem to more accurately detect the roll of the object under monitoring, the analog electrical signal of the infrared light signal undergoes basic signal and image processing to obtain a clearly defined digitized infrared thermal image. Specifically, the analog electrical signal is converted into a digital electrical signal, and the received infrared thermal image undergoes corresponding basic image processing, such as filtering, noise reduction, enhancement, or restoration. The processed digitized infrared thermal image (the digital information of the infrared thermal image) is then transmitted to the roll detection subsystem.

[0119] S103. Based on the digitized infrared thermal image, perform roll detection on the object to be monitored and determine the roll detection result of the object to be monitored.

[0120] In this embodiment, the roll detection subsystem receives and stores the digitized infrared thermal image transmitted by the image processing module, which has undergone basic signal and image processing. The digitized infrared thermal image is then processed through segmentation and feature extraction to determine the body region of the object to be monitored. Further, the head region, hand region, origin coordinates, head position coordinates, and hand position coordinates within the body region are determined. Based on the positional changes of the head and hand regions of the object in two adjacent cycles, the roll detection result is determined. After determining the roll detection result, it is transmitted to the alarm module.

[0121] S104. When the roll detection result indicates that a roll event has occurred, generate a roll alarm message, push the roll alarm message to the associated application, and synchronize the roll alarm message and roll image to the cloud.

[0122] In this embodiment, the alarm module receives the roll detection results sent by the roll detection subsystem and performs corresponding operations based on the roll detection results. When the roll detection result indicates that no roll event has occurred, the alarm module does not work; when the roll detection result indicates that a roll event has occurred, a roll alarm message is generated and the roll alarm message and roll image are synchronized to the cloud in a timely manner to remind the guardians and relevant care institutions of the monitored object, so that the guardians and relevant care institutions can respond to the roll event of the monitored object after receiving the roll alarm message and provide appropriate care for the monitored object in a timely manner.

[0123] In this embodiment, infrared thermal images and analog electrical signals of the infrared light signals of the object to be monitored are obtained by acquiring the infrared light signals and analog electrical signals of the infrared thermal images of the object to be monitored. The infrared light signals and analog electrical signals of the infrared light signals are processed to obtain digital infrared thermal images. Basic image processing is performed on the infrared thermal images to obtain processed infrared thermal images. Roll detection is performed on the object to be monitored based on the digital infrared thermal images to determine the roll detection result of the object to be monitored. When the roll detection result indicates that a roll event has occurred, a roll alarm message is generated, the roll alarm message is pushed to related applications, and the roll alarm message and the roll image are synchronized to the cloud. The above technical solution can adaptively acquire infrared light signals from the area where the monitored object is located in various environments. Before performing roll detection on the monitored object, corresponding image processing is performed in advance to make the digital infrared thermal image features input to the roll detection subsystem clearer. By monitoring the position of the head and hand areas of the monitored object, the roll detection result is determined based on the positional deviation of the head and hands, effectively enhancing the accuracy of roll detection and reducing the error. When the roll detection result indicates that a roll event has occurred, an alarm message can be generated in a timely manner to notify the monitoring personnel of the monitored object, effectively improving the personal safety of the monitored object. By adopting the above technical solution, non-contact roll detection of the monitored object is achieved, reducing the error of roll detection, improving the accuracy of roll detection, and ensuring the safety of the monitored object.

[0124] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0125] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A rollover detection system, characterized in that, include: Image acquisition module, image processing module, roll detection subsystem and alarm module; The image acquisition module is used to acquire the infrared light signal of the object to be monitored and the analog electrical signal of the infrared light signal, and transmit the infrared light signal and the analog electrical signal of the infrared light signal to the image processing module. The image processing module is used to process the infrared light signal and the analog electrical signal of the infrared light signal to obtain a digitized infrared thermal image, and transmit the digitized infrared thermal image to the tumble detection subsystem. The tumble detection subsystem is used to perform tumble detection on the object to be monitored based on the digitized infrared thermal image, and determine the tumble detection result of the object to be monitored. The alarm module is used to generate a roll alarm message when the roll detection result indicates that a roll event has occurred, push the roll alarm message to the associated application, and synchronize the roll alarm message and the roll image to the cloud. The tumbling detection subsystem includes: an image acquisition module, a body region segmentation module, a position information acquisition module, and a detection result determination module; The image acquisition module is used to acquire digitized infrared thermal images of the object to be monitored according to a preset period. The body region cutting module is used to cut out the body region of the object to be monitored based on the color temperature of the digitized infrared thermal image. The location information acquisition module is used to determine the head region and hand region of the object to be monitored, as well as the relevant location information of the object to be monitored, from the body region of the object to be monitored; the relevant location information includes the origin position coordinates, head position coordinates, and hand position coordinates; The detection result determination module is used to determine the tumbling detection result of the object to be monitored based on the changes in the relevant location information in two adjacent cycles. The detection result determination module includes: a displacement judgment unit and a second result determination unit; The displacement judgment unit is used to determine whether the displacements of the origin position coordinates and the head position coordinates in two adjacent cycles are both within a first preset range. The second result determination unit is used to determine the roll detection result of the object to be monitored when at least one of the origin position coordinates and the head position coordinates does not meet the first preset range in two adjacent cycles. The second result determination unit is specifically used for: If the displacement of at least one of the origin position coordinates and the head position coordinates in two adjacent cycles does not meet the first preset range, it is determined whether the origin position coordinates have moved within the second preset range and returned to the origin position. If so, when the hand position coordinates are the same in two adjacent cycles, the roll detection result of the object to be monitored is determined to be a roll event.

2. The system according to claim 1, characterized in that, The image acquisition module includes: a preprocessing module, a camera module, and an image sensor; The preprocessing module is used to assist the camera module in acquiring light signals under insufficient illumination conditions. The camera module is used to acquire the light signal; The image sensor is used to converge the optical signal and convert the optical signal into an analog electrical signal; The preprocessing module includes an ambient light sensor, an infrared filter, and an infrared diode unit. The ambient light sensor is used to detect the ambient illuminance and send a corresponding current signal to notify the roll detection subsystem based on the increase or decrease of the ambient light illuminance. The infrared filter is used to switch filters to achieve image noise reduction when the illumination is insufficient; The infrared diode unit is used to emit infrared light and receive infrared information in the infrared band emitted by the environment and objects illuminated by infrared light.

3. The system according to claim 1, characterized in that, The image processing module includes: an image capturing card, a filter, and a thermal image processor; The image capture card is used to convert the analog electrical signal into a digital electrical signal and perform digital image processing, and store the converted digital electrical signal into the roll detection subsystem. The filter is used to filter the analog electrical signal and transmit the filtered analog electrical signal to the thermal image processor. The thermal image processor is used to convert the analog electrical signal of the infrared light signal into a digital electrical signal and perform basic image processing to generate a digitized infrared thermal image, and transmit the digitized infrared thermal image to the roll detection subsystem.

4. The system according to claim 1, characterized in that, The location information acquisition module is specifically used for: Based on the body region of the object to be monitored, a rectangular region corresponding to the body region is determined; The center point of the object to be monitored is determined based on the side length of the rectangular area, and the center point is determined as the origin point, and the coordinates of the origin point are obtained. Based on the color temperature within the rectangular area, the head and hand areas of the object to be monitored are determined. The head position coordinates and hand position coordinates are determined based on the origin position coordinates.

5. The system according to claim 1, characterized in that, The detection result determination module further includes: a first result determination unit; The first result determination unit is used to determine that the roll detection result of the object to be monitored is a roll event when the displacement of the origin position coordinate and the head position coordinate in two adjacent cycles is within a first preset range and the hand position coordinate in two adjacent cycles is centrally symmetrical.

6. The system according to claim 1, characterized in that, The alarm module, connected to the roll detection subsystem, includes: an emergency information call module, a cloud information transceiver module, and a speaker; The cloud information transceiver module is used to send and receive cloud information, which includes the roll image of the object to be monitored and roll alarm information. The emergency information call module is used to send the cloud information to the associated application; The speaker is used to emit an alarm sound.

7. A rollover detection method, characterized in that, include: Acquire the infrared light signal of the object to be monitored and the analog electrical signal of the infrared light signal; The infrared light signal and the analog electrical signal of the infrared light signal are processed to obtain a digitized infrared thermal image. Based on the digitized infrared thermal image, roll detection is performed on the object to be monitored, and the roll detection result of the object to be monitored is determined. When the roll detection result indicates that a roll event has occurred, a roll alarm message is generated, the roll alarm message is pushed to the associated application, and the roll alarm message and roll image are synchronized to the cloud; The step of performing roll detection on the object to be monitored based on the digitized infrared thermal image and determining the roll detection result of the object to be monitored includes: The image acquisition module acquires digitized infrared thermal images of the object to be monitored according to a preset cycle. The body region segmentation module segments out the body region of the object to be monitored based on the color temperature of the digitized infrared thermal image. The location information acquisition module determines the head region and hand region of the object to be monitored, as well as the relevant location information of the object to be monitored, from the body region of the object to be monitored; the relevant location information includes the origin position coordinates, head position coordinates, and hand position coordinates; The detection result determination module determines the tumbling detection result of the object to be monitored based on the changes in the relevant location information in two adjacent cycles. The step of determining the roll detection result of the object to be monitored based on the changes in the relevant location information in two adjacent cycles includes: The displacement judgment unit determines whether the displacements of the origin position coordinates and the head position coordinates in two adjacent cycles are both within a first preset range. The second result determination unit determines the roll detection result of the object to be monitored when at least one of the origin position coordinates and the head position coordinates does not meet the first preset range in two adjacent cycles. Wherein, determining the rollover detection result of the object to be monitored when at least one of the origin position coordinates and the head position coordinates does not meet the first preset range in two adjacent cycles includes: If the displacement of at least one of the origin position coordinates and the head position coordinates in two adjacent cycles does not meet the first preset range, it is determined whether the origin position coordinates have moved within the second preset range and returned to the origin position; if so, when the hand position coordinates are the same in two adjacent cycles, it is determined that the roll detection result of the monitored object is a roll event.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause the roll detection system to implement the roll detection method of claim 7 when executed.

Images