A home monitoring method, device, equipment and storage medium

Abstract

The present invention provides a home monitoring method, device, equipment and storage medium. The method is simple and efficient, reasonably and reliably designed, capable of accurately detecting a target sound source, and realizing timely home monitoring. By detecting pulse signals in the sound source signal, using energy and statistical analysis to extract measurement features, and finally using a threshold to screen to obtain the target signal, it is possible to accurately identify the target sound source in non-stationary signals, so as to ensure that in a noisy real environment, the sound we need to detect can be identified, achieving the purpose of medical sound source detection, and extracting the sound we need from the continuous signal stream.

Description

Technical Field

[0001] The present invention relates to home health monitoring, and specifically to a home monitoring method, system, device, and storage medium. Background Art

[0002] With the increasingly significant trend of population aging, more health facilities are needed to ensure the health of the elderly. Especially for the elderly living alone, when they are at home alone and an accident occurs but outsiders cannot know in the first time, it may delay the treatment time and cause adverse consequences. In traditional Chinese culture, sending the elderly to a nursing home is not a good choice. Therefore, how to detect the health accidents that occur to the elderly living alone at home in the first time is a pressing issue.

[0003] Currently, the common monitoring method on the market is a home camera. Through the camera, children can understand the every move of the elderly at home to achieve the purpose of monitoring. For example, Chinese Patent CN108335458A discloses a home intelligent monitoring system and its monitoring method for watching the home and people. By using the provided intelligent robot to redefine the nursing function in an intelligent way of recognizing the home, people, voice, posture, heart rate and body temperature, preventing theft and intrusion, promptly detecting and intervening in the emergency situations where children want to leave the safe area and the elderly suddenly encounter serious illness and cannot call for help, and notifying the guardian to prevent the situation from deteriorating; at the same time, monitoring the behavior of the nanny, promptly capturing and recognizing the bad events of beating and abusing, and notifying the guardian to prevent the situation from deteriorating; mainly based on image recognition and supplemented by other methods. However, if the home is full of cameras, for the elderly living alone or the people in need of care, their privacy will be leaked while being monitored, and they will be infringed. Therefore, a more concealed and accurate way is needed for monitoring, and the sound monitoring and alarm technology for specific sounds is a better choice.

[0004] However, there are various fields in signal detection, such as the detection of digital signals in noise, radar signal detection, and voice activity detection. There are many methods for defining measurement features, including the likelihood of statistical models, energy, etc. Currently, most of the existing detection systems focus on detecting human speech rather than impulse sounds. There is less research on impulse sound detection. Some research can only achieve good results in white noise. Currently, most of the existing detection systems focus on detecting human speech and remove other sounds as noise. However, it is difficult to separate the target sound signal in a noisy environment, resulting in the inability to accurately detect the detection object, thus making the detection accuracy and efficiency relatively low and unable to achieve good home monitoring. Summary of the Invention

[0005] In view of the problems existing in the prior art, the present invention provides a home monitoring method, device, equipment and storage medium, which is simple and efficient, reasonably and reliably designed, can accurately detect the target sound source, and realize timely home monitoring.

[0006] The present invention is realized through the following technical solutions:

[0007] A home monitoring method,

[0008] Collect the sound source signals in the home environment in real time;

[0009] Detect the pulse signals in the sound source signals;

[0010] Perform energy and statistical analysis on the pulse signals to obtain the measurement features of sound source detection;

[0011] After comparing the measurement features with the set threshold, identify the target signals for home monitoring and transmit them to the monitoring terminal.

[0012] Preferably, the discrete wavelet transform is used to detect the pulse signals in the sound source signals.

[0013] Furthermore, the wavelet basis in the discrete wavelet transform is generated by the translation and dilation of the mother wavelet.

[0014] Still further, the mother wavelet adopts the multi-Bessel wavelet.

[0015] Preferably, the energy and statistical analysis of the pulse signals specifically includes the following steps:

[0016] Process the pulse signals through the median filtering algorithm based on the energy condition to obtain the conditional median filtering energy;

[0017] Subtract the signal energy of the pulse signals from the conditional median filtering energy to obtain the measurement features of sound source detection.

[0018] Preferably, the real-time collection of the sound source signals x(t) in the home environment is expressed as follows:

[0019]

[0020] where t is time, u is time shift, k is a weighting constant, s is a scaling factor, and ψ u,s (t) is the general function of wavelet transform.

[0021] A home monitoring device, including:

[0022] A collection module for collecting the sound source signals in the home environment in real time;

[0023] A detection module for detecting the pulse signals in the sound source signals;

[0024] An analysis module for performing energy and statistical analysis on pulse signals to obtain measurement features for sound source detection;

[0025] A comparison module for identifying target signals for home monitoring after comparing the measurement features with a set threshold;

[0026] A transmission module for transmitting the target signals to a monitoring terminal.

[0027] Preferably, the analysis module is further configured to

[0028] Process the pulse signal through a median filtering algorithm based on energy conditions to obtain conditional median filtering energy;

[0029] Subtract the signal energy of the pulse signal from the conditional median filtering energy to obtain the measurement features for sound source detection.

[0030] A home monitoring device includes a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, it can implement the home monitoring method described in any one of the above.

[0031] A computer-readable storage medium has a computer program stored thereon, and is characterized in that when the computer program is executed by a processor, it implements the home monitoring method described in any one of the above.

[0032] Compared with the prior art, the present invention has the following beneficial technical effects:

[0033] The present invention detects pulse signals in sound source signals, uses energy and statistical analysis to extract measurement features, and finally uses threshold screening to obtain target signals, so as to accurately identify the target sound source in non-stationary signals, thereby ensuring that in a noisy real environment, the sound we need to detect can be recognized, achieving the purpose of medical sound source detection, and extracting the sound we need from a continuous signal stream. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flowchart of the home monitoring method in an embodiment of the present invention.

[0035] Figure 2 It is a schematic flowchart of signal processing in the home monitoring method in an embodiment of the present invention.

[0036] Figure 3 It is a structural block diagram of the home monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The present invention will be further described in detail below in conjunction with specific embodiments, which are explanations of the present invention rather than limitations.

[0038] An embodiment of the present invention provides a home monitoring method, device, equipment and storage medium. By detecting the pulse signal in the sound source signal, the preliminary signal is extracted, and then the pulse signal is subjected to energy and statistical analysis to detect the measurement characteristics. Finally, the target signal is obtained through threshold screening, ensuring that the sound we need to detect can be recognized in a noisy real environment.

[0039] The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or equipment comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or equipment.

[0040] For ease of understanding, the specific process of the embodiment of the present invention will be described below. Please refer to Figure 2 An embodiment of the home monitoring method in the embodiment of the present invention includes:

[0041] Step 101, collect the sound source signal in the home environment in real time;

[0042] It can be understood that the execution subject of this step can be an acquisition module, or a terminal, device or complete set of acquisition equipment, such as a single or systematically arranged microphone. Specifically, it is not limited here. In the embodiment of the present invention, the microphone is taken as an example of the acquisition module for illustration. The specific acquisition process can use any one of the existing technologies to collect the sound in the home environment.

[0043] Step 102, detect the pulse signal in the sound source signal;

[0044] It can be understood that the execution subject of this step can be a detection module, or a terminal or server. Specifically, it is not limited here. In the embodiment of the present invention, the server is taken as an example of the execution subject for illustration.

[0045] The server obtains the sound source signals in the home environment that have been collected. The sound source signals here are continuously collected and transmitted, so as to be able to detect and identify the pulse signals in real time. In the prior art, the pulse signals are usually removed as noise. Therefore, in order to retain more target signal features, the value and range of the pulse signals it detects are relatively small. In the present invention, in order to retain the pulse signals, the value and range are relatively large, which is equivalent to a preliminary screening.

[0046] It should be noted that after the server receives the sound source signals of a monitored object, it needs to detect and process the sound source signals. Only after the detection and processing can the signals be better analyzed, so that the server can finally identify more accurate information. This step of operation can also be directly completed on the terminal through the processor. As much as possible, ensure that the signals obtained in the subsequent processing are more accurate and efficient, provide high-quality parameters for signal measurement feature extraction, and improve the quality and efficiency of the subsequent processing.

[0047] In this preferred embodiment, the discrete wavelet transform is specifically used to detect the pulse signals in the sound source signals;

[0048] Due to its local characteristics in the time-frequency space and uneven time-frequency resolution, wavelet transform is often used in signal detection and audio processing. In the processing of sound source signals, it is necessary to discretize the continuous wavelet and its wavelet transform. In general computer implementations, binary discrete processing is used, and the wavelet and its corresponding wavelet transform after this discretization become the discrete wavelet transform (abbreviation: DWT). In fact, the discrete wavelet transform is obtained by discretizing the scale and displacement of the continuous wavelet transform according to the power of 2, so it is also called the binary wavelet transform.

[0049] Although the classical Fourier transform can reflect the overall connotation of the signal, its manifestation is often not intuitive enough, and noise will complicate the signal spectrum. In the field of signal processing, a family of band-pass filters has always been used to decompose the signal into different frequency components, that is, the signal f(x) is sent into the family of band-pass filters Hi(x).

[0050] Using the discrete wavelet transform (DWT) for the collected sound source signals, the significance of wavelet decomposition lies in the ability to decompose the signal at different scales, and the selection of different scales can be determined according to different targets.

[0051] For many signals, the low-frequency components are quite important as they often contain the characteristics of the signal, while the high-frequency components provide the details or differences of the signal. If the high-frequency components are removed from human speech, it may sound different from before, but the content being spoken can still be understood; if enough low-frequency components are removed, only meaningless sounds are heard. Approximation and detail are often used in wavelet analysis. Approximation represents the high scale of the signal, i.e., the low-frequency information; detail represents the high scale of the signal, i.e., the high-frequency information. Therefore, the original signal generates two signals through two mutually related filters.

[0052] Through continuous decomposition processes, the approximation signal can be continuously decomposed, and the signal can be decomposed into many low-resolution components. In theory, the decomposition can continue indefinitely, but in fact, the decomposition can proceed until the detail (high-frequency) contains only a single sample. Therefore, in practical applications, the appropriate number of decomposition levels is generally selected based on the characteristics of the signal or suitable criteria.

[0053] Specifically, all sound source signals x(t) can be decomposed into the sum of functions ψ u,s (t), and are weighted by k, u, s,

[0054]

[0055] where t is time, u is the time shift, k is a weighting constant, and s is the scale factor. The type of function ψ u,s (t) is a general function selected according to the short-time Fourier transform ("frequency" analysis) or wavelet transform ("time-scale" analysis) of the sound source signal.

[0056] The discrete wavelet transform (DWT) has uneven frequency and time resolutions. DWT is used for pulse signal detection because the time resolution in the high-frequency range is high, while the time resolution in the low-frequency range is poor. Therefore, DWT can well retain the time resolution in the high-frequency range, obtain the time points corresponding to the sound sources, and improve the accuracy and precision of detection. Among them, the wavelet basis is generated by the translation and dilation of the mother wavelet ψ. The Daubechies wavelets are used as the mother wavelet because they have good regularity for a large number of moments in the signal processing process.

[0057] Step 103, perform energy and statistical analysis on the pulse signal to obtain the measurement characteristics for sound source detection; after detecting the pulse signal, perform the following operations,

[0058] After the server obtains the pulse signal, based on the energy-based conditional median filtering algorithm, it performs energy and statistical analysis on the pulse signal after DWT; uses the energy and statistical analysis to extract the measurement characteristics of the sound source information from the pulse signal.

[0059] It is understandable that the sound source signal contains very rich characteristic parameters, and different feature vectors represent different physical and acoustic meanings. The measurement features extracted by the server are of great significance to the success or failure of sound source detection. If suitable measurement parameters are selected, it will help to improve the recognition rate. Extracting measurement features is to try to remove or reduce the influence of information irrelevant to recognition in the sound source signal, reduce the amount of data to be processed in the subsequent recognition stage, and generate measurement features that represent the information of the recognition object carried in the sound source signal.

[0060] The analysis of energy and statistics is a statistical processing method for studying and analyzing vibration and sound from the perspective of energy. The basic idea is to avoid solving complex mathematical and physical equations and instead use statistical methods to study the energy transfer and balance between various parts of the system to obtain a concise physical solution.

[0061] Generally speaking, in medical sound monitoring, the detection of the sound source is crucial. Because once the sound source is lost (not detected) in the first step, the subsequent monitoring steps cannot be carried out. On the other hand, if a lot of "false sound sources" are detected, the entire detection process will tend to be saturated, resulting in the "true sound source" not being accurately detected. Therefore, the detection of the sound source is of great importance.

[0062] The detection of the sound source is essentially the detection of a signal. Detection includes identifying the required signal in a noisy environment. The assumptions are:

[0063]

[0064] Among them, o(t) is the signal to be analyzed, b(t) is the noise, and a(t) is the target signal to be detected. The basic function of the detection algorithm in the present invention is to extract some measured features or quantities from the input signal and compare these values with a threshold.

[0065] The method described in the present invention is an energy-based conditional median filtering algorithm. The measurement feature is the difference between the signal energy and the conditional median filtering energy. After comparing the measurement feature with the set threshold, the detection result is obtained.

[0066] The pulse signal detected in the present invention is actually an energy signal, that is, a signal with limited energy. Because in daily home care, for the detection of the sound source, it is actually necessary to collect a specific sound source when an accident occurs. A major feature of this sound source is that the energy is concentrated and limited, and it occurs within a limited time interval and will not occur continuously.

[0067] The energy signal itself is a pulsed signal, which usually exists only within a limited time interval. Of course, there are also some energy signals that exist within an infinite time interval, but the main part of their energy is concentrated within a limited time interval. Such signals can well characterize the accidents in home monitoring. Therefore, through the discrete wavelet transform of the sound source signal, the energy signal therein can be extracted in the form of pulses. At the same time, the pulse signal of the energy signal is a rectangular pulse signal, and the magnitude of the energy is the product of the amplitude and pulse width of the rectangular pulse signal. Therefore, it is very convenient and fast to obtain the energy of the obtained pulse signal.

[0068] Among them, the Median filtering algorithm based on energy conditions actually uses the median filtering algorithm to filter and eliminate noise points well for the energy signal represented by the rectangular pulse signal, making the characteristics of the energy more concentrated; median filtering is a non-linear signal processing technology based on the theory of sorting statistics. Its basic principle is to replace the value of a point in a digital image or digital sequence with the median of the values of each point in a neighborhood of that point, making the surrounding Pixel values close to the true value, thereby eliminating isolated noise points. Therefore, through the median filtering algorithm, the signal energy represented by the pulse signal can be expressed as the median filtering energy. Each median filtering energy represents the statistical energy of adjacent pulse signals. The difference between this statistical energy and the pulse signal energy is used to obtain the measurement characteristics of the pulse signal, which actually represents the degree of prominence of the energy of this pulse signal relative to the energy of adjacent pulse signals, further strengthening the specificity of its signal; finally, the degree of prominence is limited by a threshold, and other non-detection pulse signals in the sound source signal are further filtered, such as the interference of continuous pulse signals, so as to detect and monitor the accidents during home care more accurately.

[0069] The pulse signal is processed by the median filtering algorithm based on energy conditions to obtain the conditional median filtering energy; that is, taking the energy of the pulse signal as the processing object, median filtering is performed to obtain the conditional median filtering energy as a reference, and then the difference between each pulse signal and the reference is limited by a threshold, so that the target sound source can be accurately identified in non-stationary signals (such as: pulse signals), thus ensuring that the sound we need to detect can be identified in a noisy real environment.

[0070] Step 104, after comparing the measurement characteristics with the set threshold, the target signal for home care is identified and transmitted to the care terminal.

[0071] The above process is actually reflected in the signal processing flow, that is, the continuous change and utilization of signal types, such as Figure 2As shown in the figure, the processing process of the sound source signal 201 in an embodiment of the present invention is as follows:

[0072] Perform discrete wavelet transform (DWT) on the sound source signal 201 to obtain a pulse signal 202;

[0073] Perform energy and statistical analysis on the pulse signal 202 to obtain measurement features 203;

[0074] Perform threshold judgment on the measurement features 203 to determine whether the corresponding sound source signal 201 is a target signal 204; finally, realize the screening of the target signal 204.

[0075] An embodiment of the present invention also provides a home monitoring device, as Figure 3 shown, which includes:

[0076] An acquisition module 301, configured to collect sound source signals in the home environment in real time;

[0077] A detection module 302, configured to detect pulse signals in the sound source signal;

[0078] An analysis module 303, configured to perform energy and statistical analysis on the pulse signal to obtain measurement features for sound source detection;

[0079] A comparison module 304, configured to identify the target signal for home monitoring after comparing the measurement features with a set threshold;

[0080] A transmission module 305, configured to transmit the target signal to the monitoring terminal.

[0081] Among them, the analysis module 303 is further configured to:

[0082] Process the pulse signal through a median filtering algorithm based on energy conditions to obtain conditional median filtering energy;

[0083] Subtract the signal energy of the pulse signal from the conditional median filtering energy to obtain the measurement features for sound source detection.

[0084] An embodiment of the present invention also provides a home monitoring device, including a processor and a memory. When the computer program stored in the memory is executed by the processor, it can implement the home monitoring method described in any of the above solutions.

[0085] An embodiment of the present invention further provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the home monitoring method described in any of the above solutions is implemented.

[0086] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0087] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0089] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the functions in the processFigure 1 one process or multiple processes and / or blocks Figure 1 steps of functions specified in one block or multiple blocks.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A home monitoring method, characterized in that, it collects sound source signals in the home environment in real time; uses discrete wavelet transform to detect pulse signals in the sound source signals; performs energy and statistical analysis on the pulse signals, processes the pulse signals through a median filtering algorithm based on energy conditions to obtain conditional median filtering energy; subtracts the signal energy of the pulse signals from the conditional median filtering energy to obtain the measurement characteristics of sound source detection; compares the measurement characteristics with a set threshold value, identifies the target signal for home monitoring, and transmits it to the monitoring terminal.

2. The home monitoring method according to claim 1, characterized in that, the wavelet basis in the discrete wavelet transform is generated by the translation and dilation of the mother wavelet.

3. The home monitoring method according to claim 2, characterized in that, the mother wavelet adopts the multi-Bessel wavelet.

4. The home monitoring method according to claim 1, characterized in that, Real-time collection of sound source signals in the home environment It is shown as follows, where t is time, u is time shift, k is a weighted constant, s is a scale factor, is the general function of wavelet transform.

5. A home monitoring device, characterized in that, it includes, a collection module for collecting sound source signals in the home environment in real time; a detection module for using discrete wavelet transform to detect pulse signals in the sound source signals; an analysis module for performing energy and statistical analysis on the pulse signals, processing the pulse signals through a median filtering algorithm based on energy conditions to obtain conditional median filtering energy; subtracting the signal energy of the pulse signals from the conditional median filtering energy to obtain the measurement characteristics of sound source detection; a comparison module for comparing the measurement characteristics with a set threshold value to identify the target signal for home monitoring; a transmission module for transmitting the target signal to the monitoring terminal.

6. A home monitoring device, characterized in that, it includes a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, it can implement the home monitoring method according to any one of claims 1-4.

7. A computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by the processor, it implements the home monitoring method according to any one of claims 1-4.

Images