Off-bed monitoring alarm method and system based on infrared rays

Abstract

The invention discloses an off-bed monitoring alarm method and system based on infrared rays and computing equipment. The method comprises the steps that at least one infrared camera is installed above each sickbed in a neurology ward; extracting a bed fixed edge contour in infrared thermal imaging according to a Hough transform algorithm; separating a difference area between a patient and a background from infrared thermal imaging, and performing edge detection on the difference area to obtain a preliminary contour of the patient as an initial position of a Snake model; internal energy and external energy of an energy function in the Snake model are minimized; when the shape of the Snake model does not change any more or reaches a preset number of iterations, taking a convergent model contour as a patient dynamic contour; calculating the relative position of the center of the patient dynamic contour relative to the bed fixed edge contour; when the relative position moves out of the boundary of the preset bed area, alarm information is sent to a central system. According to the method, the contour of the patient is iteratively calculated through the Snake model, and the safety and efficiency of night monitoring of the patient are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical monitoring and image processing, and in particular to an infrared-based bed exit monitoring alarm method and system, and a computing device. Background Art

[0002] As the population ages, the number of patients in neurology departments is increasing, and these patients are often at high risk of falling and getting lost. Especially at night or when there is no one to monitor them, these risks may lead to serious consequences, including patient injury or even death. Currently, hospitals generally use manual monitoring and regular inspections to monitor the status of patients, but this method has obvious limitations, and the bed exit alarm has a single function.

[0003] The existing method of monitoring patients leaving the bed mainly uses pressure sensors, that is, by embedding pressure sensors in the mattress to monitor the patient's weight changes, so as to determine whether the patient has left the bed. However, this method has problems such as easy damage of sensors, high maintenance costs, and inability to provide specific location and dynamic information of the patient.

[0004] In order to solve the above problems, the present invention proposes an infrared-based bed exit monitoring alarm method to improve the safety and efficiency of nighttime monitoring of patients. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] The present invention provides an infrared-based bed-leaving monitoring alarm method and system, and a computing device for improving the safety and efficiency of nighttime monitoring of patients.

[0007] (II) Technical solution

[0008] According to one aspect of the present invention, there is provided an infrared-based bed exit monitoring alarm method, comprising: At least one infrared camera is installed above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor for collecting infrared thermal images in the ward; Extracting the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; Separating the difference area between the patient and the background from the infrared thermal imaging, performing edge detection on the difference area to obtain a preliminary contour of the patient as the initial position of the Snake model; minimizing the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, taking the converged model contour as the patient's dynamic contour; The relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed is calculated; when the relative position moves out of the preset bed area boundary, an alarm message is sent to the central system of the nurse station.

[0009] In an optional manner, the extracting the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm further includes: Convert the external thermal image into a grayscale image, extract the grayscale image using a Sobel edge detection algorithm, and obtain a preliminary bed fixing edge; Performing Hough transformation on the preliminary fixed edge of the bed to obtain a boundary line of the fixed edge of the bed; The boundary line is screened according to the geometric characteristics and position information of the bed to obtain the fixed edge contour of the bed.

[0010] In an optional manner, the internal energy includes elastic energy and smoothing energy; The elastic energy is used to control the continuity of the curve to ensure that the curve is not over-stretched or over-compressed, and the elastic energy is proportional to the square of the first-order derivative of the curve; The smoothing energy is used to control the smoothness of the curve to avoid sharp corners of the curve. The smoothing energy is proportional to the square of the second-order derivative of the curve. Minimizing the internal energy of the energy function in the Snake model further includes:

[0011] The elastic coefficient and the rigidity coefficient of the internal energy are adjusted by the gradient descent method to iteratively adjust the position of the curve control point of the preliminary contour so as to gradually reduce the internal energy.

[0012] In an optional manner, the external energy is used to attract the curve of the preliminary contour to the edge of the patient; Minimizing the external energy of the energy function in the Snake model further includes: guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information; wherein when the curve of the preliminary contour is close to the edge of the patient, reducing the external energy by increasing the image gradient; The positions of the curve control points of the preliminary contour are adjusted by the Newton iteration method so that the external energy is gradually reduced.

[0013] In an optional manner, the expression of the internal energy is: ; in, is the parameterized representation of the contour curve; is the arc length parameter of the curve; is the elastic coefficient; is the rigidity coefficient.

[0014] In an optional manner, the expression of the external energy is: ; in, is the external energy adjustment coefficient; is the sensitivity coefficient of external energy changing with gradient; For point The magnitude of the image gradient at .

[0015] In an optional manner, guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information further comprises: Calculate the gradient vector of each pixel in infrared thermal imaging; A gradient direction vector is defined, wherein the gradient direction vector points to the direction in which the brightness in the infrared thermal imaging image changes fastest to obtain an edge direction; For each control point on the preliminary contour, calculate the dot product of the gradient direction vector of all pixel points in its neighborhood and the vector pointing from the control point to the pixel point, and take the pixel point corresponding to the maximum value as the attraction point of the control; According to the attraction points of all control points, the positions of the control points are updated by Newton iteration method.

[0016] In an optional manner, the iteration formula of the Newton iteration method is: ; in, For the At the first iteration, the contour The new coordinates of the control points; For the At the first iteration, the contour The coordinates of the control points; is the iteration step length; For control points The neighborhood set of ; Pixel The gradient direction vector at ; is the gradient direction vector Model; is the gradient direction vector and arrive The projection length between the vectors.

[0017] According to another aspect of the present invention, a rapid identification system based on stroke symptoms is provided, comprising: A video acquisition module is used to install at least one infrared camera above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor for acquiring infrared thermal images in the ward; A bed edge contour extraction module, used to extract the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; A patient dynamic contour generation module is used to separate the difference area between the patient and the background from the infrared thermal imaging, perform edge detection on the difference area to obtain the patient's preliminary contour as the initial position of the Snake model; minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; The patient position monitoring and alarm module is used to calculate the relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, an alarm message is sent to the central system of the nurse station.

[0018] According to another aspect of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned infrared-based bed exit monitoring alarm method.

[0019] (III) Beneficial effects (1) Through the enhanced infrared sensor and Hough transform algorithm, the fixed edge contour of the bed can be accurately extracted, and combined with the Snake model to iteratively optimize the patient's dynamic contour, high-precision monitoring of the patient's position can be achieved, effectively reducing false alarms and missed alarms.

[0020] (2) The Snake model can automatically adapt to changes in the patient's contour by minimizing internal and external energy, and can accurately track the patient's position even if the patient moves or changes posture in bed. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 A schematic diagram showing a flow chart of an infrared-based bed exit monitoring alarm method according to an embodiment of the present invention; Figure 2 A structural block diagram of a system for rapid identification of stroke symptoms according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0022] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0023] Figure 1 The flowchart of the infrared-based bed exit monitoring alarm method according to an embodiment of the present invention is shown. Specifically, the method includes the following steps.

[0024] Step S101, at least one infrared camera is installed above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor for collecting infrared thermal imaging in the ward.

[0025] In this embodiment, an infrared camera with an enhanced infrared sensor is selected to ensure that it can still image clearly under low light conditions. The camera is installed above each bed to ensure that the patient's body area can be fully covered while avoiding excessive exposure of the patient's privacy.

[0026] Step S102: extracting the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm.

[0027] In this embodiment, Hough Transform is a feature extraction algorithm, especially used to detect simple shapes (such as straight lines, circles, etc.) in image processing. The algorithm converts points in the image space into curves in the parameter space through a voting mechanism, thereby simplifying the shape recognition process. Hough Transform has a strong resistance to noise and incomplete data in the image, and can accurately extract the fixed edge contour of the bed from complex infrared thermal imaging. Even if there is a certain degree of interference or blur in the image, Hough Transform can suppress noise interference to a certain extent through its parameter space mapping mechanism, ensuring the accuracy and stability of the extracted edge contour.

[0028] Specifically, the image edge information obtained by edge detection is mapped into parameter space and accumulator voted through the Hough transform algorithm to extract the fixed edge contour of the bed. According to the geometric characteristics and position information of the bed, the edge contour extracted by the Hough transform is screened and optimized to remove noise points and interference contours of non-bed edges, for example, to remove contour segments that are too short or whose directions do not conform to the edge characteristics of the bed, and finally obtain the fixed edge contour of the bed, and obtain the final fixed edge contour of the bed.

[0029] In an optional manner, the extracting the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm further includes: Convert the external thermal image into a grayscale image, extract the grayscale image using a Sobel edge detection algorithm, and obtain a preliminary bed fixing edge; Performing Hough transformation on the preliminary fixed edge of the bed to obtain a boundary line of the fixed edge of the bed; The boundary line is screened according to the geometric characteristics and position information of the bed to obtain the fixed edge contour of the bed.

[0030] In this embodiment, the boundary lines obtained by the Hough transform are screened according to the geometric characteristics and position information of the bed, and the falsely detected boundary lines are further eliminated to ensure the accuracy of the final extracted fixed edge contour of the bed. For example, the grayscale image is processed by the Sobel edge detection algorithm to obtain a binary image containing the preliminary fixed edge of the bed and other edges. The Hough transform algorithm is applied to the binary image to obtain multiple possible straight lines (boundary lines). According to the geometric characteristics (such as length, width) and position information (such as the approximate position of the bed in the image) of the hospital bed, the boundary lines obtained by the Hough transform are screened to obtain the fixed edge contour of the bed.

[0031] Step S103, separating the difference area between the patient and the background from the infrared thermal imaging, performing edge detection on the difference area to obtain a preliminary contour of the patient as the initial position of the Snake model; minimizing the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, the converged model contour is used as the patient's dynamic contour.

[0032] In this embodiment, the Snake model (active contour model) can dynamically adapt to changes in the patient's contour through iterative optimization of internal energy (controlling the smoothness and continuity of the contour) and external energy (guiding the contour to approach the edge of the image feature).

[0033] Specifically, the infrared thermal imaging is preprocessed by filtering, denoising, contrast enhancement, etc., and the patient area is separated from the background according to the temperature difference by using methods such as threshold segmentation or clustering algorithm to obtain the difference area image. The Canny edge detector is applied to the difference area image to extract the preliminary contour of the patient, and the preliminary contour is used as the initial position of the Snake model. The coordinate points of the preliminary contour are set as the initial control points of the Snake model, and the internal energy parameters (elastic coefficient, rigidity coefficient) and external energy parameters (edge ​​force coefficient) of the Snake model are set. According to the energy function of the Snake model, the control points are iteratively calculated by gradient descent. In each iteration, the internal energy and external energy of each control point are calculated, and the position of the control point is updated to minimize the energy function. During the iteration process, it is determined whether the shape of the Snake model no longer changes (the change in the position of the control point is less than the preset threshold) or the preset number of iterations is reached. If any condition is met, the model is considered to have converged and the iteration is stopped. The converged Snake model contour is used as the dynamic contour of the patient.

[0034] In an optional manner, the internal energy includes elastic energy and smoothing energy; The elastic energy is used to control the continuity of the curve to ensure that the curve is not over-stretched or over-compressed, and the elastic energy is proportional to the square of the first-order derivative of the curve; The smoothing energy is used to control the smoothness of the curve to avoid sharp corners of the curve. The smoothing energy is proportional to the square of the second-order derivative of the curve. Minimizing the internal energy of the energy function in the Snake model further includes: The elastic coefficient and the rigidity coefficient of the internal energy are adjusted by the gradient descent method to iteratively adjust the position of the curve control point of the preliminary contour so as to gradually reduce the internal energy.

[0035] In this embodiment, the introduction of elastic energy can ensure that the curve maintains a natural and continuous shape during the deformation process, avoiding unnatural stretching or compression. The smoothing energy makes the curve smoother and reduces sharp corners.

[0036] Specifically, a set of initial contour curve control points are automatically selected to roughly outline the boundary of the target area. An internal energy function containing elastic energy and smooth energy is defined, where the elastic energy term is obtained by calculating the square of the first-order derivative of the curve, and the smooth energy term is obtained by calculating the square of the second-order derivative of the curve. The initial values ​​of the elastic coefficient and the rigidity coefficient are set, and the total value of the current internal energy is calculated. For each control point, the gradient (i.e., partial derivative) of its position change to the internal energy is calculated. The position of the control point is updated according to the gradient direction and step size (learning rate). The above steps are repeated until the internal energy reaches the preset threshold or the number of iterations reaches the upper limit. The final control point position is the optimized curve, which maintains continuity and has a high degree of smoothness.

[0037] In an optional manner, the external energy is used to attract the curve of the preliminary contour to the edge of the patient; Minimizing the external energy of the energy function in the Snake model further includes: guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information; wherein when the curve of the preliminary contour is close to the edge of the patient, reducing the external energy by increasing the image gradient; The positions of the curve control points of the preliminary contour are adjusted by the Newton iteration method so that the external energy is gradually reduced.

[0038] The external energy function is defined based on the gradient map. Usually, the external energy is proportional to the negative logarithm or negative exponential of the gradient value to ensure that the external energy decreases when the contour approaches the edge.

[0039] In this embodiment, a set of initial contour control points are selected, and the total value of the current external energy and its gradient (first-order derivative) and Hessian matrix (second-order derivative) with respect to the control point position are calculated. The control point position is updated using the Newton iteration formula, that is, a set of linear equations is solved to find the optimal step size and direction to reduce the external energy. The above steps are repeated until the external energy reaches a preset threshold or the number of iterations reaches an upper limit. The control point position finally obtained defines the optimized contour, which closely fits the edge of the patient.

[0040] In an optional manner, guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information further comprises: Calculate the gradient vector of each pixel in infrared thermal imaging; A gradient direction vector is defined, wherein the gradient direction vector points to the direction in which the brightness in the infrared thermal imaging image changes fastest to obtain an edge direction; For each control point on the preliminary contour, calculate the dot product of the gradient direction vector of all pixel points in its neighborhood and the vector pointing from the control point to the pixel point, and take the pixel point corresponding to the maximum value as the attraction point of the control; According to the attraction points of all control points, the positions of the control points are updated by Newton iteration method.

[0041] In this embodiment, for each control point on the preliminary contour, the dot product of the gradient direction vector of the pixel point in its neighborhood and the vector of the pixel point pointed to by the control point is calculated, and the pixel point corresponding to the maximum value is selected as the attraction point to ensure that the contour can move along the most likely edge path. The gradient direction vector can be obtained by normalizing the gradient vector, pointing to the direction with the fastest brightness change, that is, the edge direction. For each control point on the preliminary contour, its neighborhood range is determined (such as the area within a certain radius centered on the control point). In the neighborhood, the dot product of the gradient direction vector of each pixel point and the vector pointing from the control point to the pixel point is calculated. The pixel point corresponding to the maximum dot product value is selected as the attraction point of the control point, because the gradient direction at this point is most consistent with the direction pointed to by the control point to this point, and it is most likely to be a point on the edge. According to the attraction points of all control points, an external energy function based on the distance between the control point and the attraction point is defined. The gradient (first-order derivative) and Hessian matrix (second-order derivative) of the external energy function with respect to the position of the control point are calculated. The control point positions are updated using the Newton iteration formula until the external energy reaches a preset threshold or the number of iterations reaches an upper limit, and an updated preliminary contour is obtained as the edge of the patient.

[0042] In this embodiment, the iteration formula of the Newton iteration method is: ; in, For the At the first iteration, the contour The new coordinates of the control points; For the At the first iteration, the contour The coordinates of the control points; is the iteration step length; For control points The neighborhood set of ; Pixel The gradient direction vector at ; is the gradient direction vector Model; is the gradient direction vector and arrive The projection length between the vectors.

[0043] Select a control point , and determine the pixels in its neighborhood. For each pixel in the neighborhood , calculate the gradient direction vector With from arrive The projection length of the vector. According to the iterative formula, calculate The new coordinates of the control point will move one step toward the edge. Repeat the above steps to iteratively update all control points. After multiple iterations, the contour shape gradually approaches the true circular edge. When the iteration converges (that is, the contour shape no longer changes significantly or reaches the preset number of iterations), the final contour control point coordinates are output to depict the edge of the circular area according to the coordinate points.

[0044] In an optional manner, the expression of the internal energy is: ; in, is the parameterized representation of the contour curve; is the arc length parameter of the curve; is the elastic coefficient; is the rigidity coefficient.

[0045] The internal energy provides stable constraints for the contour optimization process, which can gradually approach the optimal solution during the iteration process without getting into trouble due to the interference of local optimal solutions or noise.

[0046] In an optional manner, the expression of the external energy is: ; in, is the external energy adjustment coefficient; is the sensitivity coefficient of external energy changing with gradient; For point The magnitude of the image gradient at .

[0047] External energy is calculated based on image gradients, which can prevent the contour curve from falling into the local optimal solution to a certain extent. When the contour curve is close to the real edge, the image gradient will increase significantly, thereby enhancing the attraction of external energy and helping the contour curve jump out of the local optimal area.

[0048] Step S104, calculating the relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, sending an alarm message to the central system of the nurse station.

[0049] By dynamically calculating the relative position of the center of the patient's contour and the fixed edge contour of the bed, the patient's position change can be monitored in real time. Once the patient's position exceeds the preset safety area, an alarm message is immediately sent to the nurse station to ensure that medical staff can respond quickly. In addition, it helps prevent patients from falling out of the bed due to unconscious movements (such as turning over, sitting up, etc.), especially at night or when medical staff are not present, reducing the risk of patient injury. The entire monitoring process reduces the workload of medical staff while improving the accuracy and timeliness of monitoring. Optionally, the preset bed area boundaries can be flexibly configured according to the needs of different patients and bed sizes to adapt to various clinical scenarios.

[0050] The solution provided by the above embodiment of the present invention is to install at least one infrared camera above each bed in the neurology ward, and the camera is equipped with an enhanced infrared sensor for collecting infrared thermal imaging in the ward; extract the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; separate the difference area between the patient and the background from the infrared thermal imaging, and perform edge detection on the difference area to obtain the preliminary contour of the patient as the initial position of the Snake model; minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; calculate the relative position of the center of the patient's dynamic contour relative to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, send an alarm message to the central system of the nurse station. The present invention can accurately extract the fixed edge contour of the bed through the enhanced infrared sensor and the Hough transform algorithm, and iteratively optimize the patient's dynamic contour in combination with the Snake model, so as to achieve high-precision monitoring of the patient's position and effectively reduce false alarms and missed alarms. The Snake model automatically adapts to changes in the patient's contour by minimizing internal and external energy, and accurately tracks the patient's position even if the patient moves or changes position in the bed.

[0051] Figure 2 The structural block diagram of the system for rapid identification of stroke symptoms according to an embodiment of the present invention is shown, comprising: The video acquisition module 210 is used to install at least one infrared camera above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor to collect infrared thermal images in the ward; A bed edge contour extraction module 220 is used to extract the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; The patient dynamic contour generation module 230 is used to separate the difference area between the patient and the background from the infrared thermal imaging, perform edge detection on the difference area to obtain the patient's preliminary contour as the initial position of the Snake model; minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; The patient position monitoring and alarm module 240 is used to calculate the relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, an alarm message is sent to the central system of the nurse station.

[0052] Figure 3 The schematic diagram of the structure of the computing device embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the computing device.

[0053] like Figure 3 As shown, the computing device may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .

[0054] The processor 302, the communication interface 304, and the memory 306 communicate with each other via the communication bus 308. The communication interface 304 is used to communicate with other devices such as a client or other server network elements. The processor 302 is used to execute the program 310, which can specifically execute the relevant steps in the above-mentioned infrared-based bed exit monitoring alarm method embodiment.

[0055] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0056] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0057] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0058] The solution provided by the above embodiment of the present invention is to install at least one infrared camera above each bed in the neurology ward, and the camera is equipped with an enhanced infrared sensor for collecting infrared thermal imaging in the ward; extract the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; separate the difference area between the patient and the background from the infrared thermal imaging, and perform edge detection on the difference area to obtain the preliminary contour of the patient as the initial position of the Snake model; minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; calculate the relative position of the center of the patient's dynamic contour relative to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, send an alarm message to the central system of the nurse station. The present invention can accurately extract the fixed edge contour of the bed through the enhanced infrared sensor and the Hough transform algorithm, and iteratively optimize the patient's dynamic contour in combination with the Snake model, so as to achieve high-precision monitoring of the patient's position and effectively reduce false alarms and missed alarms. The Snake model automatically adapts to changes in the patient's contour by minimizing internal and external energy, and accurately tracks the patient's position even if the patient moves or changes position in the bed.

[0059] The algorithm or display provided herein is not inherently related to any particular computer, virtual system or other equipment. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the embodiment of the present invention is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description made to specific languages ​​above is for disclosing the best mode of the present invention.

[0060] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0061] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0062] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and further may be divided into a plurality of submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device so disclosed may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0063] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0064] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention may also be implemented as a device or system program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0065] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The present invention may be implemented with the aid of hardware comprising several different elements and with the aid of appropriately programmed computers. In a unit claim that lists several systems, several of these systems may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A bed leaving monitoring alarm method based on infrared, characterized in that: include: At least one infrared camera is installed above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor for collecting infrared thermal images in the ward; Extracting the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; Separating the difference area between the patient and the background from the infrared thermal imaging, performing edge detection on the difference area to obtain a preliminary outline of the patient as the initial position of the Snake model; Minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; The relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed is calculated; when the relative position moves out of the preset bed area boundary, an alarm message is sent to the central system of the nurse station.

2. The infrared-based bed exit monitoring alarm method according to claim 1 is characterized in that: The extracting of the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm further comprises: Convert the external thermal image into a grayscale image, extract the grayscale image using a Sobel edge detection algorithm, and obtain a preliminary bed fixing edge; Performing Hough transformation on the preliminary fixed edge of the bed to obtain a boundary line of the fixed edge of the bed; The boundary line is screened according to the geometric characteristics and position information of the bed to obtain the fixed edge contour of the bed.

3. The infrared-based bed exit monitoring alarm method according to claim 1 or 2, characterized in that: The internal energy includes elastic energy and smooth energy; The elastic energy is used to control the continuity of the curve to ensure that the curve is not over-stretched or over-compressed, and the elastic energy is proportional to the square of the first-order derivative of the curve; The smoothing energy is used to control the smoothness of the curve to avoid sharp corners of the curve. The smoothing energy is proportional to the square of the second-order derivative of the curve. Minimizing the internal energy of the energy function in the Snake model further includes: The elastic coefficient and the rigidity coefficient of the internal energy are adjusted by the gradient descent method to iteratively adjust the position of the curve control point of the preliminary contour so as to gradually reduce the internal energy.

4. The infrared-based bed exit monitoring alarm method according to claim 1 or 2, characterized in that: The external energy is used to attract the curve of the preliminary contour to the edge of the patient; Minimizing the external energy of the energy function in the Snake model further includes: guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information; wherein when the curve of the preliminary contour is close to the edge of the patient, reducing the external energy by increasing the image gradient; The positions of the curve control points of the preliminary contour are adjusted by the Newton iteration method so that the external energy is gradually reduced.

5. The infrared-based bed exit monitoring alarm method according to claim 3 is characterized in that: The expression of the internal energy is: ； in, is the parameterized representation of the contour curve; is the arc length parameter of the curve; is the elastic coefficient; is the rigidity coefficient.

6. The infrared-based bed exit monitoring alarm method according to claim 4, characterized in that: The expression of the external energy is: ； in, is the external energy adjustment coefficient; is the sensitivity coefficient of external energy changing with gradient; For point The magnitude of the image gradient at .

7. The infrared-based bed exit monitoring alarm method according to claim 4 is characterized in that: The step of guiding the curve of the preliminary contour to move toward the edge of the patient by using image gradient information further comprises: Calculate the gradient vector of each pixel in infrared thermal imaging; Define a gradient direction vector, wherein the gradient direction vector points to the direction in which the brightness in the infrared thermal imaging changes fastest to obtain an edge direction; For each control point on the preliminary contour, calculate the dot product of the gradient direction vector of all pixel points in its neighborhood and the vector pointing from the control point to the pixel point, and take the pixel point corresponding to the maximum value as the attraction point of the control; According to the attraction points of all control points, the positions of the control points are updated by Newton iteration method.

8. The infrared-based bed exit monitoring alarm method according to claim 7, characterized in that: The iteration formula of the Newton iteration method is: ； in, For the At the first iteration, the contour The new coordinates of the control points; For the At the first iteration, the contour The coordinates of the control points; is the iteration step length; For control points The neighborhood set of ; Pixel The gradient direction vector at ; is the gradient direction vector Model; is the gradient direction vector and arrive The projection length between the vectors.

9. An infrared-based bed exit monitoring alarm system, comprising: A video acquisition module is used to install at least one infrared camera above each bed in the neurology ward, wherein the camera is equipped with an enhanced infrared sensor for acquiring infrared thermal images in the ward; A bed edge contour extraction module, used to extract the fixed edge contour of the bed in the infrared thermal imaging according to the Hough transform algorithm; A patient dynamic contour generation module is used to separate the difference area between the patient and the background from the infrared thermal imaging, and perform edge detection on the difference area to obtain a preliminary contour of the patient as the initial position of the Snake model; Minimize the internal energy and external energy of the energy function in the Snake model through multiple iterative calculations; when the shape of the Snake model no longer changes or reaches a preset number of iterations, use the converged model contour as the patient's dynamic contour; The patient position monitoring and alarm module is used to calculate the relative position of the center of the patient's dynamic contour with respect to the fixed edge contour of the bed; when the relative position moves out of the preset bed area boundary, an alarm message is sent to the central system of the nurse station.

10. A computing device comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the infrared-based bed exit monitoring alarm method according to any one of claims 1-8.

Images