Visual monitoring-based home-based care automatic alarm communication method

Abstract

The invention discloses a home-based care automatic alarm communication method based on visual monitoring, and relates to the technical field of home visual monitoring, and the method comprises the following steps: carrying out the graying processing of a home image, and obtaining a home gray-scale map; obtaining a first screening threshold and a second screening threshold based on the first number of human body images; performing binarization processing on the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binarization image; obtaining a target contour based on the home binary image; constructing a real-time comparison value based on the target contour; acquiring a comparison threshold based on the images when the second number of old people fall down; generating a falling signal based on the real-time comparison value and the comparison threshold value, and automatically alarming; the method and the device are used for solving the problem that the accuracy of tumble judgment under the conditions of squatting and stooping is reduced due to the fact that squatting, stooping and tumble cannot be further distinguished in an existing tumble detection technology.

Description

Technical Field

[0001] The present invention relates to the technical field of home visual monitoring, and in particular to an automatic alarm communication method for home-based elderly care based on visual monitoring. Background Art

[0002] Home-based care has become the mainstream model of elderly care. However, elderly people living alone, in empty nests, or at advanced ages face numerous safety risks at home. Accidental falls are the leading cause of serious injury, disability, and even death among the elderly. If they fail to receive timely assistance after a fall, the consequences will be significantly worse. Therefore, fall detection for the elderly is necessary. Existing fall detection includes sensor detection and monitoring detection. Sensor detection requires real-time wearing. For elderly people with declining memory, they may forget to wear the detection equipment, resulting in the inability to detect. Existing monitoring-based detection uses model training to determine whether a fall has occurred. However, when the elderly squat or bend over to pick up something, the image is relatively similar to the image when they fall, which reduces the accuracy of fall judgment. For example, the patent application with publication number CN119723807A discloses a visual intelligent monitoring and alarm system for home-based elderly care. This solution fails to further distinguish the images when squatting or bending over, resulting in a decrease in the accuracy of fall judgment. The existing fall detection technology fails to further distinguish squatting and bending over from falling, resulting in a decrease in the accuracy of fall judgment when squatting and bending over. Summary of the Invention

[0003] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by constructing a real-time contrast value based on the target contour; obtaining a contrast threshold based on images of a second number of elderly people falling; generating a fall signal and automatically alarming based on the real-time contrast value and the contrast threshold, so as to solve the problem that the existing fall detection technology fails to further distinguish between squatting and bending over and falling, resulting in a decrease in the accuracy of fall judgment in the case of squatting and bending over.

[0004] To achieve the above objectives, the present application provides a method for automatic alarm communication for home-based elderly care based on visual monitoring, comprising the following steps: Obtain a bird's-eye view image of the home environment in the white-hot mode of visual infrared monitoring, and mark it as a home image; Grayscale processing is performed on the home image to obtain a home grayscale image; Obtaining a first screening threshold and a second screening threshold based on a first number of human body images; Binarizing the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binary image; Obtain the target outline based on the home binary image; Build real-time comparison values ​​based on target profiles; Acquire a contrast threshold based on the images of a second number of elderly people falling; A fall signal is generated based on the real-time comparison value and the comparison threshold and an automatic alarm is issued.

[0005] Furthermore, grayscale processing is performed on the home image to obtain a home grayscale image, which includes the following sub-steps: If the home image is an image with a single color channel, the home image is marked as a home binary image. If the home image is an image with three color channels of red, green, and blue, the red, green, and blue color channel values ​​of each pixel in the home image are averaged and merged into a grayscale channel value to obtain a home grayscale image.

[0006] Furthermore, obtaining a first screening threshold and a second screening threshold based on the first number of human body images includes the following sub-steps: Marking a first number of human body images in the incandescent mode of the visual infrared monitoring as human body images; gray-scaling the human body images to obtain human body grayscale images; obtaining grayscale values ​​of human body parts in the human body grayscale images, and marking them as human body grayscale values; Obtain the range of human grayscale values; divide the range of human grayscale values ​​into a equal ranges, marked as human body division ranges; Count the frequency of each human body division range and mark it as human body division frequency; The human body division frequencies are sorted from small to large according to the middle values ​​of the human body division range to obtain the human body serial number, which is a positive integer starting from 1.

[0007] Furthermore, obtaining the first screening threshold and the second screening threshold based on the first number of human body images further includes the following sub-steps: Get the frequency of human grayscale value, marked as Rp; The human body segmentation threshold is calculated as: Rz=b×(Rp÷a); where Rz is the human body segmentation threshold and b is the ratio of 0 to 1; According to the human body serial number from small to large, determine whether the human body division frequency is less than the human body division threshold. If so, delete the human body division frequency, continue to determine until the human body division frequency is greater than or equal to the human body division threshold, and stop determining; if not, stop determining directly; according to the human body serial number from large to small, determine whether the human body division frequency is less than the human body division threshold. If so, delete the human body division frequency, continue to determine until the human body division frequency is greater than or equal to the human body division threshold, and stop determining; if not, stop determining directly; obtain the minimum and maximum values ​​of the human body division ranges corresponding to all the remaining human body division frequencies, and mark them as the first screening threshold and the second screening threshold, respectively.

[0008] Furthermore, binarizing the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binary image includes the following sub-steps: The grayscale values ​​in the home grayscale image that are greater than or equal to the first screening threshold and less than or equal to the second screening threshold are set to 0, and the grayscale values ​​in the home grayscale image that are less than the first screening threshold and greater than the second screening threshold are set to 255 to obtain a home binary image.

[0009] Furthermore, obtaining the target outline based on the home binary image includes the following sub-steps: The binary image of the home is scanned from top to bottom and from left to right to obtain the first pixel whose grayscale value changes from 255 to 0, marked as the initial pixel, and determine whether the adjacent pixels around the initial pixel contain pixels with a grayscale value of 0. If no new initial pixel is obtained until the adjacent pixels around the initial pixel, the initial pixel at this time is marked as the starting pixel; starting with the adjacent pixel above the starting pixel and the starting pixel as the center, the pixels are scanned in a clockwise direction to obtain the first pixel whose grayscale value changes from 255 to 0, marked as the search pixel; repeat starting with the adjacent pixel above the search pixel and the search pixel as the center, scanning the pixels in a clockwise direction to obtain a new search pixel, until the new search pixel is the same as the starting pixel, and stop the operation to obtain the starting pixel and all the search pixels to form the target contour.

[0010] Furthermore, constructing a real-time contrast value based on the target profile includes the following sub-steps: Establish a plane rectangular coordinate system, mark it as the construction coordinate system, and place the home binary map in the construction coordinate system; The home binary image is rotated 360° along the center of the home binary image, and the maximum and minimum values ​​of the horizontal coordinate of a target contour are obtained in real time, marked as the horizontal maximum value and the horizontal minimum value respectively, and the difference between the horizontal maximum value and the horizontal minimum value is calculated and marked as the horizontal difference value; Obtain the state of the home binary image in the constructed coordinate system when the horizontal difference value is the largest during the rotation process, and mark it as the recognition state; In the recognition state, the maximum and minimum values ​​of the vertical coordinate of a target contour are obtained and marked as the vertical maximum value and the vertical minimum value respectively; the difference between the vertical maximum value and the vertical minimum value is calculated and marked as the vertical difference value; the horizontal difference value is calculated and divided by the vertical difference value and marked as the real-time comparison value.

[0011] Furthermore, obtaining a contrast threshold based on the second number of images of elderly people falling down includes the following sub-steps: Marking a second number of images of the elderly falling in the incandescent mode of the visual infrared monitoring as fall images; obtaining real-time contrast values ​​of the fall images and marking them as fall contrast values; Obtain the range of the fall comparison value; divide the range of the fall comparison value into c equal ranges, marking them as fall division ranges; Count the frequency of each fall classification range and mark it as fall classification frequency; The fall frequency is sorted from small to large according to the middle value of the fall range to obtain the fall sequence number, which is a positive integer starting from 1.

[0012] Furthermore, obtaining the contrast threshold based on the second number of images of elderly people falling down further includes the following sub-steps: Get the frequency of fall comparison value, marked as Sp; The fall classification threshold is calculated as: Sz=d×(Rp÷c); where Z is the fall classification threshold and d is the ratio from 0 to 1; Determine whether the fall classification frequency is less than the fall classification threshold according to the fall serial number from small to large. If so, delete the fall classification frequency and continue to judge until the fall classification frequency is greater than or equal to the fall classification threshold. If not, stop judging directly; obtain the minimum value of the fall classification range corresponding to all the remaining fall classification frequencies and mark it as the comparison threshold.

[0013] Furthermore, generating a fall signal and automatically alarming based on the real-time comparison value and the comparison threshold includes the following sub-steps: If the real-time comparison value is less than the comparison threshold, a normal behavior signal is generated; if the real-time comparison value is greater than or equal to the comparison threshold, a fall signal is generated and sent to the supervisor.

[0014] Beneficial effects of the present invention: The present invention constructs a real-time contrast value based on the target outline; obtains a contrast threshold based on images of a second number of elderly people falling; generates a fall signal based on the real-time contrast value and the contrast threshold and automatically alarms. The advantage is that it can further distinguish between squatting and bending over and falling, and improve the accuracy of fall judgment in the case of squatting and bending over. The present invention constructs a real-time contrast value through the target contour. The advantage is that the real-time contrast value can be used to further distinguish the images of squatting and bending in the area from those of falling, thereby improving the accuracy of fall judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of the steps of the method of the present invention; Figure 2 A schematic diagram of obtaining search pixels according to the present invention; Figure 3 A schematic diagram of obtaining an identification state of the present invention; Figure 4 It is a schematic diagram of the horizontal maximum value, horizontal minimum value, vertical minimum value and vertical maximum value of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 As shown, the present application provides an automatic alarm communication method for home-based elderly care based on visual monitoring, comprising the following steps: Step S1, obtaining a bird's-eye view image of the home environment in the incandescent mode of visual infrared monitoring, and marking it as a home image; because the human body has temperature, the infrared image in the incandescent mode is used to facilitate human body recognition.

[0018] Step S2: grayscale the home image to obtain a home grayscale image. Step S2 includes the following sub-steps: Step S201: If the home image is a single-color channel image, mark the home image as a home binary image. If the home image is a three-color channel image of red, green, and blue, average the red, green, and blue color channel values ​​of each pixel in the home image and merge them into a grayscale channel value to obtain a home grayscale image. In practical applications, because the images obtained in the white-hot mode under different monitoring devices may be grayscale images or pseudo-color images, pseudo-color images are similar to grayscale images, but are actually RGB images. Therefore, they need to be converted into grayscale images for subsequent operations. For example, if the home image is an image with a single color channel, it is a grayscale image, and the home image is marked as a home binary image.

[0019] Step S3, obtaining a first screening threshold and a second screening threshold based on the first number of human body images; Step S3 includes the following sub-steps: Step S301: Mark a first number of human body images in a white-hot mode of visual infrared monitoring as human body images; grayscale the human body images to obtain human body grayscale images; obtain grayscale values ​​of human body parts in the human body grayscale images and mark them as human body grayscale values; in order to obtain a range of human body grayscale values, multiple human body images are required, and the more human body images there are, the more accurate the range of human body grayscale values ​​is, so the first number is set as large as possible, for example, the first number is set to 100; Step S302, obtaining the range of human grayscale values; dividing the range of human grayscale values ​​into a equal ranges, marked as human body division ranges; a is set to observe the number of human grayscale value distributions in different ranges, because grayscale values ​​are integers, and at the same time ensure that the human body division range is an integer, for example, a is 14; Step S303, counting the frequency of each human body division range and marking it as human body division frequency; Step S304, sorting the human body division frequencies from small to large according to the median values ​​of the human body division ranges to obtain human body serial numbers, where the human body serial numbers are positive integers starting from 1; Step S305, obtaining the frequency of the gray value of the human body, marked as Rp; In step S306, the human body segmentation threshold is calculated as: Rz = b × (Rp ÷ a); wherein Rz is the human body segmentation threshold, and b is a ratio of 0 to 1; b is set to obtain a human body segmentation range with a relatively small human body segmentation frequency, so b is relatively small. At the same time, the larger Rp is, the smaller b is set for better accuracy, for example, b is 0.1.

[0020] Step S307: Determine whether the human body division frequency is less than the human body division threshold according to the human body serial number from small to large. If so, delete the human body division frequency and continue to determine until the human body division frequency is greater than or equal to the human body division threshold; if not, stop determining directly; determine whether the human body division frequency is less than the human body division threshold according to the human body serial number from large to small. If so, delete the human body division frequency and continue to determine until the human body division frequency is greater than or equal to the human body division threshold; if not, stop determining directly; obtain the minimum value and maximum value of the human body division range corresponding to all the remaining human body division frequencies, and mark them as the first screening threshold and the second screening threshold respectively; this process can filter out individual abnormal human grayscale values ​​that are too large or too small, thereby obtaining an accurate human grayscale value distribution range; In practical applications, for example, the detectable temperature image of visual infrared monitoring is between -50°C and 50°C, and the range of human grayscale values ​​is 178 to 220. The range of human grayscale values ​​is evenly divided into 14 equal ranges, namely 178 to 181, 181 to 184, ..., 217 to 220. The frequency of human grayscale values ​​is 12.21 million, and the human body segmentation threshold is calculated as: Rz = 0.1 × (1221 ÷ 14) = 8.7214. The result is rounded to four decimal places. If the human body segmentation frequencies corresponding to 178 to 181, 181 to 184, ..., 214 to 217, and 217 to 220 are 6.1231, 12.1221, ..., 17.2398, and 4.1811, respectively, then 178 to 181 and 217 to 220 are deleted, and the first and second screening thresholds are obtained as 181 and 217, respectively.

[0021] Step S4, binarizing the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binary image; Step S4 includes the following sub-steps: Step S401: Set the grayscale values ​​in the home grayscale image that are greater than or equal to the first screening threshold and less than or equal to the second screening threshold to 0, and set the grayscale values ​​in the home grayscale image that are less than the first screening threshold and greater than the second screening threshold to 255, thereby obtaining a home binary image; thus, a human body contour can be obtained; In practical applications, the grayscale values ​​greater than or equal to 181 and less than or equal to 217 in the home grayscale image are set to 0, and the grayscale values ​​less than 181 and greater than or equal to 217 in the home grayscale image are set to 255 to obtain a home binary image.

[0022] Step S5, obtaining the target outline based on the home binary image; Step S5 includes the following sub-steps: Step S501: Scan the home binary image from top to bottom and from left to right to obtain the first pixel whose grayscale value changes from 255 to 0, mark it as the initial pixel, determine whether the adjacent pixels around the initial pixel contain pixels with a grayscale value of 0, and if no new initial pixel is obtained until the adjacent pixels around the initial pixel are obtained, mark the initial pixel at this time as the starting pixel; start with the adjacent pixel above the starting pixel and the starting pixel as the center, scan the pixels in a clockwise direction to obtain the first pixel whose grayscale value changes from 255 to 0, mark it as the search pixel; repeat the process of starting with the adjacent pixel above the search pixel and the search pixel as the center, scan the pixels in a clockwise direction to obtain a new search pixel, and stop the operation when the new search pixel is the same as the starting pixel, and obtain the target contour composed of the starting pixel and all the search pixels; In practical applications, please refer to Figure 2 As shown in FIG, the scanning method can filter out isolated interfering pixels and obtain the boundary contour of the human body.

[0023] Step S6, constructing a real-time contrast value based on the target profile; Step S6 includes the following sub-steps: Step S601: Establish a plane rectangular coordinate system, marked as a construction coordinate system, and place the home binary image in the construction coordinate system; Step S602: Rotate the home binary image 360° around the center of the home binary image, obtain the maximum and minimum values ​​of the horizontal coordinate of a target contour in real time, mark them as horizontal maximum and horizontal minimum, respectively, and calculate the difference between the horizontal maximum and horizontal minimum, which is marked as horizontal difference. Step S603, obtaining the state of the home binary image in the constructed coordinate system when the horizontal difference value is the largest during the rotation process, and marking it as the recognition state; Step S604: In the recognition state, the maximum and minimum values ​​of the vertical coordinate of a target outline are obtained, marked as the vertical maximum value and the vertical minimum value, respectively; the difference between the vertical maximum value and the vertical minimum value is calculated, marked as the vertical difference value; the horizontal difference value is divided by the vertical difference value, marked as the real-time comparison value; because the image obtained is a top view, in the case of a fall, the real-time comparison value is equivalent to the height being greater than the width of the upper body, while when the human body squats or bends, the corresponding length is less than the height being greater than the width of the upper body. Therefore, the fall can be distinguished based on the size of the real-time comparison value; In practical applications, please refer to Figure 3 As shown, the recognition status process is obtained, please refer to Figure 4 As shown, the maximum and minimum vertical values ​​obtained are 182 and 108 respectively, and the maximum and minimum horizontal values ​​obtained are 248 and 80 respectively. The real-time comparison value is: (248-80) ÷ (182-108) = 2.27.

[0024] Step S7, obtaining a contrast threshold based on the images of the second number of elderly people falling down; Step S7 includes the following sub-steps: Step S701: Mark a second number of images of an elderly person falling in the incandescent mode of the visual infrared monitoring system as fall images; obtain real-time contrast values ​​of the fall images and mark them as fall contrast values. To obtain a range of the fall contrast values, multiple fall images are required. The more fall images there are, the more accurate the range of the fall contrast values ​​will be. Therefore, the second number is preferably set as large as possible, for example, the second number is set to 200. Step S702: Obtain the range of the fall comparison value; divide the range of the fall comparison value into c equal ranges, marked as fall division ranges; c is set to observe the number of fall comparison value distributions in different ranges, for example, c is 10; Step S703: Count the frequency of each fall classification range and mark it as the fall classification frequency; Step S704: sort the fall frequency classifications from small to large according to the median value of the fall classification range to obtain a fall sequence number, where the fall sequence number is a positive integer starting from 1; Step S705, obtaining the frequency of the fall comparison value, marked as Sp; In step S706, the fall classification threshold is calculated as: Sz = d × (Rp ÷ c); where Z is the fall classification threshold and d is a ratio of 0 to 1. d is set to obtain a fall classification range with a relatively small fall classification frequency, so b is relatively small, for example, b is 0.1.

[0025] Step S707: Determine whether the fall classification frequency is less than the fall classification threshold according to the fall sequence number from small to large. If so, delete the fall classification frequency and continue to determine whether the fall classification frequency is greater than or equal to the fall classification threshold. If not, stop determining immediately. Obtain the minimum value of the fall classification range corresponding to all remaining fall classification frequencies and mark it as the comparison threshold. In practical applications, for example, the range of fall contrast values ​​is 1.8 to 3.8, and the frequency of fall contrast values ​​is 200. The fall classification threshold is calculated as: Sz = 0.1 × (200 ÷ 10) = 2. If the fall classification frequency in the fall classification range of 1.8 to 2.0 is 1, it is deleted. If the fall classification frequency in the fall classification range of 1.8 to 2.0 is 3, the judgment is stopped, and the comparison threshold is 1.8.

[0026] Step S8, generating a fall signal based on the real-time comparison value and the comparison threshold and automatically alarming; Step S8 includes the following sub-steps: Step S801: If the real-time comparison value is less than the comparison threshold, a normal behavior signal is generated; if the real-time comparison value is greater than or equal to the comparison threshold, a fall signal is generated and sent to the supervisor. In the case of a fall, the real-time comparison value is equivalent to the height being greater than the width of the upper body. When a person squats or bends, the length is less than the height being greater than the width of the upper body. Therefore, a fall can be determined based on the real-time comparison value being greater than or equal to the comparison threshold. In actual applications, if the real-time comparison value 2.27 is greater than the comparison threshold 1.8, a fall signal is generated and sent to the supervisor.

[0027] In a second embodiment, the present application further provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the automatic alarm communication method for home-based elderly care based on visual monitoring are executed to implement the following functions: obtaining a top-view image of a home environment in the incandescent mode of visual infrared monitoring and marking it as a home image; grayscale processing the home image to obtain a home grayscale map; obtaining a first screening threshold and a second screening threshold based on a first number of human body images; binarizing the home grayscale map based on the first grayscale threshold and the second grayscale threshold to obtain a home binary map; obtaining a target outline based on the home binary map; constructing a real-time contrast value based on the target outline; obtaining a contrast threshold based on images of a second number of elderly people falling; generating a fall signal and automatically alarming based on the real-time contrast value and the contrast threshold.

[0028] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0029] Example 3. The present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the automatic alarm communication method for home-based elderly care based on visual monitoring provided by the above methods, the method including: obtaining a top-view image of the home environment in the incandescent mode of visual infrared monitoring, marking it as a home image; grayscale processing the home image to obtain a home grayscale map; obtaining a first screening threshold and a second screening threshold based on a first number of human body images; binarizing the home grayscale map based on the first grayscale threshold and the second grayscale threshold to obtain a home binary map; obtaining a target contour based on the home binary map; constructing a real-time contrast value based on the target contour; obtaining a contrast threshold based on images of a second number of elderly people falling; generating a fall signal based on the real-time contrast value and the contrast threshold and automatically alarming.

[0030] Example 4. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned automatic alarm communication method for home-based elderly care based on visual monitoring are executed to achieve the following functions: obtaining a top-view image of the home environment in the white-hot mode of visual infrared monitoring, marked as a home image; grayscale processing is performed on the home image to obtain a home grayscale map; a first screening threshold and a second screening threshold are obtained based on a first number of human body images; the home grayscale map is binarized based on the first grayscale threshold and the second grayscale threshold to obtain a home binary map; a target contour is obtained based on the home binary map; a real-time contrast value is constructed based on the target contour; a contrast threshold is obtained based on images of a second number of elderly people falling; a fall signal is generated based on the real-time contrast value and the contrast threshold and an automatic alarm is issued.

[0031] Through the description of the above embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the essence of the above technical solutions or the portion that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various embodiments or certain portions of the embodiments.

[0032] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An automatic alarm communication method for home-based elderly care based on visual monitoring, characterized in that: The steps include: Obtain a bird's-eye view image of the home environment in the white-hot mode of visual infrared monitoring, and mark it as a home image; Grayscale processing is performed on the home image to obtain a home grayscale image; Obtaining a first screening threshold and a second screening threshold based on a first number of human body images; Binarizing the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binary image; Obtain the target outline based on the home binary image; Build real-time comparison values ​​based on target profiles; Acquire a contrast threshold based on the images of a second number of elderly people falling; A fall signal is generated based on the real-time comparison value and the comparison threshold and an automatic alarm is issued.

2. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 1 is characterized in that: Grayscale processing of the home image to obtain a home grayscale image includes the following sub-steps: If the home image is an image with a single color channel, the home image is marked as a home binary image. If the home image is an image with three color channels of red, green, and blue, the red, green, and blue color channel values ​​of each pixel in the home image are averaged and merged into a grayscale channel value to obtain a home grayscale image.

3. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 2 is characterized in that: Acquiring a first screening threshold and a second screening threshold based on the first number of human body images includes the following sub-steps: Marking a first number of human body images in the incandescent mode of the visual infrared monitoring as human body images; performing grayscale processing on the human body images to obtain human body grayscale images; Obtain the grayscale value of the human body part in the human body grayscale image, and mark it as the human body grayscale value; Obtain the range of human grayscale values; divide the range of human grayscale values ​​into a equal ranges, marked as human body division ranges; Count the frequency of each human body division range and mark it as human body division frequency; The human body division frequencies are sorted from small to large according to the middle values ​​of the human body division range to obtain the human body serial number, which is a positive integer starting from 1.

4. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 3 is characterized in that: Acquiring the first screening threshold and the second screening threshold based on the first number of human body images further includes the following sub-steps: Get the frequency of human grayscale value, marked as Rp; The human body segmentation threshold is calculated as: Rz=b×(Rp÷a); where Rz is the human body segmentation threshold and b is the ratio of 0 to 1; According to the human body serial number from small to large, determine whether the human body division frequency is less than the human body division threshold. If so, delete the human body division frequency, continue to determine until the human body division frequency is greater than or equal to the human body division threshold, and stop determining; if not, stop determining directly; according to the human body serial number from large to small, determine whether the human body division frequency is less than the human body division threshold. If so, delete the human body division frequency, continue to determine until the human body division frequency is greater than or equal to the human body division threshold, and stop determining; if not, stop determining directly; obtain the minimum and maximum values ​​of the human body division ranges corresponding to all the remaining human body division frequencies, and mark them as the first screening threshold and the second screening threshold, respectively.

5. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 4 is characterized in that: Binarizing the home grayscale image based on the first grayscale threshold and the second grayscale threshold to obtain a home binary image includes the following sub-steps: The grayscale values ​​in the home grayscale image that are greater than or equal to the first screening threshold and less than or equal to the second screening threshold are set to 0, and the grayscale values ​​in the home grayscale image that are less than the first screening threshold and greater than the second screening threshold are set to 255 to obtain a home binary image.

6. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 5 is characterized in that: Obtaining the target outline based on the home binary image includes the following sub-steps: The binary image of the home is scanned from top to bottom and from left to right to obtain the first pixel whose grayscale value changes from 255 to 0, marked as the initial pixel, and determine whether the adjacent pixels around the initial pixel contain pixels with a grayscale value of 0. If no new initial pixel is obtained until the adjacent pixels around the initial pixel, the initial pixel at this time is marked as the starting pixel; starting with the adjacent pixel above the starting pixel and the starting pixel as the center, the pixels are scanned in a clockwise direction to obtain the first pixel whose grayscale value changes from 255 to 0, marked as the search pixel; repeat starting with the adjacent pixel above the search pixel and the search pixel as the center, scanning the pixels in a clockwise direction to obtain a new search pixel, until the new search pixel is the same as the starting pixel, and stop the operation to obtain the starting pixel and all the search pixels to form the target contour.

7. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 6 is characterized in that: Constructing a real-time contrast value based on the target profile includes the following sub-steps: Establish a plane rectangular coordinate system, mark it as the construction coordinate system, and place the home binary map in the construction coordinate system; The home binary image is rotated 360° along the center of the home binary image, and the maximum and minimum values ​​of the horizontal coordinate of a target contour are obtained in real time, marked as the horizontal maximum value and the horizontal minimum value respectively, and the difference between the horizontal maximum value and the horizontal minimum value is calculated and marked as the horizontal difference value; Obtain the state of the home binary image in the constructed coordinate system when the horizontal difference value is the largest during the rotation process, and mark it as the recognition state; In the recognition state, the maximum and minimum values ​​of the vertical coordinate of a target contour are obtained and marked as the vertical maximum value and the vertical minimum value respectively; Calculate the difference between the maximum vertical value and the minimum vertical value, and mark it as the vertical difference; Calculate the horizontal difference value divided by the vertical difference value and mark it as the real-time comparison value.

8. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 7 is characterized in that: Acquiring a contrast threshold based on the second number of images of elderly people falling down includes the following sub-steps: Marking a second number of images of the elderly falling in the incandescent mode of the visual infrared monitoring as fall images; obtaining real-time contrast values ​​of the fall images and marking them as fall contrast values; Obtain the range of the fall comparison value; divide the range of the fall comparison value into c equal ranges, marking them as fall division ranges; Count the frequency of each fall classification range and mark it as fall classification frequency; The fall frequency is sorted from small to large according to the middle value of the fall range to obtain the fall sequence number, which is a positive integer starting from 1.

9. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 8 is characterized in that: Acquiring a contrast threshold based on the second number of images of elderly people falling down further includes the following sub-steps: Get the frequency of fall comparison value, marked as Sp; The fall classification threshold is calculated as: Sz=d×(Rp÷c); where Z is the fall classification threshold and d is the ratio from 0 to 1; Determine whether the fall classification frequency is less than the fall classification threshold according to the fall serial number from small to large. If so, delete the fall classification frequency and continue to judge until the fall classification frequency is greater than or equal to the fall classification threshold. If not, stop judging directly; obtain the minimum value of the fall classification range corresponding to all the remaining fall classification frequencies and mark it as the comparison threshold.

10. The method for automatic alarm communication for home-based elderly care based on visual monitoring according to claim 9 is characterized in that: Generating a fall signal and automatically alarming based on the real-time comparison value and the comparison threshold includes the following sub-steps: If the real-time comparison value is less than the comparison threshold, a normal behavior signal is generated; if the real-time comparison value is greater than or equal to the comparison threshold, a fall signal is generated and sent to the supervisor.

Images