CN109757928B - Prevent falling intelligent crib - Google Patents
Prevent falling intelligent crib Download PDFInfo
- Publication number
- CN109757928B CN109757928B CN201811653925.0A CN201811653925A CN109757928B CN 109757928 B CN109757928 B CN 109757928B CN 201811653925 A CN201811653925 A CN 201811653925A CN 109757928 B CN109757928 B CN 109757928B
- Authority
- CN
- China
- Prior art keywords
- infant
- crossing
- baby
- crib
- risk level
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Alarm Systems (AREA)
Abstract
The utility model provides an prevent falling intelligent crib, includes the crib body at least and with the crib bedstead of crib body rigid coupling, prevent falling intelligent crib and still include that the baby climbs over detection module, a plurality of hand action detection module and a plurality of foot action detection module. When at least one specific event occurs, pressure change information of the anti-falling intelligent crib is obtained and recorded through at least one of the foot behavior detection modules and at least one of the hand behavior detection modules, infant crossing estimation analysis is carried out based on the received pressure change information, and an infant crossing estimation result and/or an infant crossing risk grade matched with the infant crossing estimation result and/or the infant crossing estimation grade are/is output to the central processing module, and the central processing module sends out an early warning prompt to an intelligent terminal operated by a guardian based on the received infant crossing estimation result and/or the infant crossing risk grade and/or sends out first acousto-optic information through the acousto-optic prompt module.
Description
Technical Field
The invention relates to the technical field of intelligent infant monitoring, in particular to an anti-falling intelligent infant bed.
Background
After 10 months of birth, infants and young children generally have a strong desire to try to walk, and parents often place an object of interest to the infant at a distance or higher to guide the infant to begin learning to walk, gradually entering another growth stage. According to the widely applied baby development reference index 'Danver development screening test table' at present, it is shown that the infants begin to learn to walk in the period of 10 months to 14 months and 15 months all belong to the normal age range, the walking age of the academic conference is different due to genetic genes, breeding modes, character characteristics or birth months, but the walking age of the infant academic conference is not greatly related to any ability in future life.
Generally, an infant can walk by holding objects when being eight months old, starts to learn how to bend the knees to squat or stand tired to sit when being nine to ten months old, is not as simple as an adult from standing to squat or from standing to sitting, and is a great difficult and dangerous project for the infant regardless of the actions of holding objects to walk or stand to squat, so that a guardian needs to invest a great deal of time and energy to take care of the infant. Moreover, the baby just born is easy to lack the sense of safety and hope to be in the arms of parents all the time, but after all, the guardian has limited energy and cannot continuously take care of the baby, and particularly for young parents with high working and living pressure, the rest time is more valuable.
Therefore, the transition from crawling to trying to stand or squat is often not observed by the guardian, and the baby still cannot walk and stays on the crib for most of the time, so that the guardian is easy to ignore the problem of whether the baby is placed on the crib alone or not. The crib among the prior art usually fixes the infant on the bed body of crib through the mode of fixed band or sleeping bag, can guarantee the safety of infant when guardians attend to the scope outside. However, the crib brings great movement restriction to the infants, is easy to cause strong discomfort and fear of the infants, and causes unnecessary crying and screaming, thereby bringing more worry to guardians.
A Chinese patent (publication number is CN108741926A) is an anti-falling safe crib and a metering anti-falling method thereof, comprising a bed enclosure body, wherein the bottom of the bed enclosure body is fixed with a supporting bed board, and a weighing supporting plate and a mattress are sequentially arranged on the supporting bed board; the weighing support plate comprises a main support plate, a first side support plate and a second side support plate, a first weighing sensor, a second sensor and a third weighing sensor are arranged on the support bed plate, the sensing end of the first weighing sensor is connected with the main support plate in a matching mode, the sensing end of the second weighing sensor is connected with the first side support plate in a matching mode, and the sensing end of the third weighing sensor is connected with the third side support plate in a matching mode; two sides of the bed surrounding body are respectively provided with an anti-falling blocking mechanism. This patent sets up special forced induction structure, can transmit the structure that different pressure concentration points adjusted the crib according to the baby, and the baffle of falling is prevented to corresponding the rising can stop the baby and turn over out the crib.
However, in the anti-falling safe crib and the metering anti-falling method thereof provided by the patent, a plurality of sensors are arranged on the mattress, when the detection value of one of the sensors at two sides exceeds a preset threshold value, whether the baby turns over, stands and is playing or tries to climb the crib cannot be accurately judged, the baby is totally judged to have the hidden trouble of falling and the supporting plate at the corresponding side of the crib body is lifted, so that the baby is prevented from moving to fall from the crib. Although the threshold method is relatively simple to implement, the accuracy of infant drop prediction is not high due to the fact that the subjectively selected single fixed threshold is not universal to various infant posture conditions, and a certain misjudgment rate or alarm missing phenomenon exists.
Chinese patent (the publication number is CN206044117U) discloses an intelligent control safe and reliable's crib, including bed board, main body frame, control box, music switch, alarm and battery case, main body frame is installed to the bed board top, main body frame internally mounted has safety barrier, and installs the crossbeam above the safety barrier, the headstock is installed on the main body frame right side, and installs the control box above the headstock, control box surface mounting has the music switch, the light is installed on music switch right side, switch is installed in music switch left side, the interface that charges is installed in the control box left side, the USB interface is installed on the control box right side, control box internally mounted has the battery case, battery case internally mounted has the battery.
Although this patent is through installing infrared inductor, just can sense and remind the user through the alarm is reported to the police when the baby climbs the bed, the effectual accident that prevents the baby crib that drops that appears. However, when the baby may just play to grasp the bed frame and does not want to turn over the baby bed, or when the height and strength of the baby in the current growing period are far insufficient enough to turn over the baby bed, the baby bed provided by the patent can frequently trigger an alarm to give an alarm to the guardian, which also brings more concerns and care difficulties to the guardian, and cannot accurately and effectively provide the guardian with the monitoring auxiliary effect. Therefore, in view of the above deficiencies, there is a need for a crash-proof intelligent baby crib which can accurately determine the over-turning behavior of the baby and give an early warning in time before the over-turning, while overcoming the deficiencies of the prior art.
Disclosure of Invention
In view of the defects of the prior art, the invention provides an anti-falling intelligent crib, which at least comprises a crib body and a crib frame fixedly connected with the crib body, and further comprises an infant crossing detection module, a plurality of hand behavior detection modules and a plurality of foot behavior detection modules, wherein the infant crossing detection module is arranged on the crib frame and is connected with a central processing module, the hand behavior detection modules are used for acquiring first pressure change information, the foot behavior detection modules are used for acquiring second pressure change information and acquiring related specific events, and the infant crossing detection module is configured to execute the following steps:
when at least one specific event occurs, the pressure change information of the anti-falling intelligent crib is obtained and recorded through at least one of the foot behavior detection modules and at least one of the hand behavior detection modules when the specific event occurs, infant crossing estimation analysis is carried out on the basis of the received pressure change information, and an infant crossing estimation result and/or an infant crossing risk level matched with the infant crossing estimation result and/or the infant crossing estimation result are/is output to the central processing module, and the central processing module sends out an early warning prompt to an intelligent terminal operated by a guardian on the basis of the received infant crossing estimation result and/or the infant crossing risk level and/or sends out first acousto-optic information through the acousto-optic prompt module.
According to a preferred embodiment, said infant crossing estimation analysis comprises at least the following steps: when the current time reaches a risk level judgment period, sending effective pressure change information to the infant crossing detection module according to the risk level of the infant crossing detection module in the last risk level judgment period within a first time length;
determining a duration of at least one effective pressure change message of the infant crossover detection module within the first duration;
and determining the risk level of the infant crossing detection module in the current risk level judgment period according to the obtained duration of the at least one piece of effective pressure change information of the infant crossing detection module in the first duration.
According to a preferred embodiment, the infant crossing prediction analysis further comprises the steps of: if the obtained duration in the current risk level judgment period is greater than the first risk level threshold, determining that the infant crossing detection module is at a first risk level in the current risk level period, and after the first duration, indicating the central processing module to send an early warning prompt to an intelligent terminal operated by a guardian by the infant crossing detection module according to the first risk level;
if the obtained duration in the current risk level judgment period is greater than the second risk level threshold, determining that the infant crossing detection module is at the second risk level in the current risk level period, indicating the central processing module to send an early warning prompt to an intelligent terminal operated by a guardian according to the second risk level by the infant crossing detection module after the first duration, and continuously sending first acousto-optic information through the acousto-optic prompt module located at the first position from the first time to the second time.
According to a preferred embodiment, if the obtained duration in the current risk level determination period is greater than the second risk level threshold, it is determined that the infant crossover detection module is at the second risk level in the current risk level period, high risk analysis is performed according to a first occurrence time of the first risk level and a second occurrence time of the second risk level to output a high risk time period matched with the first risk level, and after the first duration, the infant crossover detection module instructs the central processing module to send an early warning prompt to an intelligent terminal operated by a guardian according to the second risk level and the high risk time period.
According to a preferred embodiment, the infant crossing detection module at least comprises a data receiving unit, a data storage unit and a data comparison unit, wherein the data receiving unit is configured to obtain the pressure change information including the first pressure change information and the second pressure change information when at least one specific event occurs and output the pressure change information to the data comparison unit, the data storage unit is configured to obtain an infant crossing estimation model according to an infant configuration profile connected to the central processing module and output the infant crossing estimation model to the data comparison unit, and the data comparison unit is configured to perform infant crossing estimation analysis on the received pressure change information and the infant crossing estimation model to obtain the infant crossing estimation result and output the infant crossing estimation result to the central processing module.
According to a preferred embodiment, the anti-falling intelligent baby crib further comprises a plurality of bed leaving detection modules arranged on the crib frame, and the bed leaving detection modules are configured to execute the following steps: when the at least one specific event occurs, tracking video frame information in the anti-falling intelligent baby crib through at least one of the plurality of bed exit detection modules to obtain an actual baby crossing result and current time information, and instructing the central processing module to output the actual baby crossing result to the baby crossing detection module and/or sending out second acousto-optic information through the acousto-optic prompt module based on the actual baby crossing result.
According to a preferred embodiment, if the infant crossing detection module with the second risk level in the current risk level period receives the actual infant crossing result containing the information that the infant has crossed the bed, the infant crossing detection module is upgraded to the third risk level in the current risk level period, and after the second time, the infant crossing detection module instructs the central processing module to integrate the acquired current time and the second position into alarm information according to the third risk level and sends the alarm information to at least one medical care service platform.
According to a preferred embodiment, the infant-crossing detection module further comprises a data-updating module configured to perform the steps of: and comparing and analyzing the actual infant crossing result input by the out-of-bed detection module with the infant crossing estimation result input by the data comparison unit and corresponding to the current time information, and finishing the training and updating process of the infant crossing estimation model stored in the infant configuration profile based on the actual infant crossing result and the pressure change information under the condition that the comparison of the actual infant crossing result and the infant crossing estimation result fails.
According to a preferred embodiment, the infant crossing prediction model is configured to: acquiring the actual infant crossing result including the video frame information, the current time information and the pressure change information, which are input by the bed-leaving detection module;
preprocessing the video frame information according to a matching method and obtaining the matching rate of the images in the anti-falling intelligent crib;
extracting the matching rate of a period of time sequence based on the video frame information and the current time information as a classification feature and taking the obtained matching rate sequence as a feature vector X;
and obtaining the infant crossing estimation result based on the infant crossing estimation model established by the feature vector X and the classification discriminant function, and storing the infant crossing estimation result and the pressure change information to the infant configuration profile in a time information correlation mode.
According to a preferred embodiment, the matching method is configured to: the target image block is slid on the input source image, the actual source image block is matched with the target image, and the matching rate of the source image and the target image can be obtained every time the target image block is slid.
The invention has the beneficial technical effects that:
(1) according to the anti-falling intelligent crib, the problems that the false alarm rate is high and the assistance is poor when the crib climbing behavior is detected, and the problem that a guardian is fatigued to take an alarm frequently to reduce the confidence level of the provided early warning information due to the fact that the guardian is tired of taking care of the alarm frequently in the prior art are solved by simultaneously collecting the information of the limbs and the hands of the infants, monitoring whether the infants cross the crib in time and then accurately estimating and analyzing the infant crossing by combining the actual growth conditions of the infants in different periods. Meanwhile, the guardian can not immediately stop the behavior that the baby tries to overturn because the guardian can have a certain distance from the baby, and the sound and light information capable of attracting the attention of the baby is sent out by the bed body, so that double guarantee is provided for preventing the baby from overturning over the crib.
(2) Because the crib body has certain degree of difficulty to the baby that the height is not enough and strength is less, it is a lagged and comparatively slow process that the baby wants to overturn out of the crib body, and its result often needs to show through a period of time. Therefore, the invention divides the corresponding risk grade after the estimation and analysis of the baby crossing through the proposed baby crossing risk grade, and judges the risk grade of the baby under the corresponding condition so as to determine whether external intervention or alarm processing is needed, thereby ensuring the safety and controllability of the baby in the intelligent crib.
(3) According to the invention, the risk level of the baby crossing the bed body of the crib corresponding to different daily moments is evaluated and predicted in advance, so that the occurrence probability of the accident that the baby crosses the bed body of the crib can be further reduced, and the emergency remedial treatment after the original crossing accident is changed into early warning and precaution, so that the safety of the baby in the intelligent crib is well guaranteed.
Drawings
FIG. 1 is a schematic view of the connection relationship of the modules of the crash-proof intelligent crib of the present invention; and
fig. 2 is a schematic diagram of the module connection relationship of the preferred baby crossing detection module of the present invention.
List of reference numerals
1: the central processing module 2: infant crossing detection module 3: foot behavior detection module
4: hand behavior detection module 5: the acousto-optic prompt module 6: communication module
7: and (4) the intelligent terminal 8: off-bed detection module 9: data updating module
201: data receiving unit 202: the data comparing unit 203: data storage unit
Detailed Description
The following detailed description is made with reference to the accompanying drawings.
The invention provides an anti-falling intelligent crib, which at least comprises a crib body and a crib frame fixedly connected with the crib body, and further comprises a baby crossing detection module 2, a plurality of hand behavior detection modules 4 and a plurality of foot behavior detection modules 3, wherein the baby crossing detection module 2 is arranged on the crib frame and is connected with a central processing module 1, the hand behavior detection modules 4 are used for acquiring first pressure change information, the foot behavior detection modules 3 are used for acquiring second pressure change information and acquiring related specific events, and the baby crossing detection module 2 is configured to execute the following steps: when at least one specific event occurs, the pressure change information of the anti-falling intelligent crib is obtained and recorded through at least one of the foot behavior detection modules 3 and at least one of the hand behavior detection modules 4 when the specific event occurs, infant crossing estimation analysis is carried out on the basis of the received pressure change information, and an infant crossing estimation result and/or an infant crossing risk level matched with the infant crossing estimation result and/or the infant crossing estimation result are/is output to the central processing module 1, and the central processing module 1 sends out an early warning prompt to an intelligent terminal 7 operated by a guardian on the basis of the received infant crossing estimation result and/or the infant crossing risk level and/or sends out first acousto-optic information through an acousto-optic prompt module 5.
The anti-falling intelligent crib can timely monitor whether the infant is crossing the crib and accurately estimate and judge whether the infant can cross the crib or not by simultaneously collecting the hand and foot limb information of the infant and utilizing the hand holding action and the foot stepping off the ground action which are necessary for the infant to cross the guardrail on the crib, thereby solving the problems of high false alarm rate and poor assistance rate of alarming the guardian as long as the crib climbing action is detected in the prior art, because the estimation and judgment process is to combine the actual growth conditions of each infant in different periods to carry out infant crossing estimation and analysis, the false alarm rate is further reduced compared with the mode of singly setting a fixed threshold value to carry out early warning, therefore, scientific and reliable early warning information is provided for the guardian, and the condition that the guardian is tired of taking care of the early warning information due to frequent alarms is avoided, so that the trust degree of the provided early warning information is reduced. Meanwhile, the guardian can be away from the infant by a certain distance and cannot immediately stop the behavior that the infant tries to overturn, and the sound and light information capable of attracting the attention of the infant is sent out by the bed body, so that double guarantee is provided for preventing the infant from overturning over the crib.
When a guardian temporarily leaves the crib, the baby may have a behavior of trying to climb over the crib, when the baby holds the crib frame of the anti-falling intelligent crib, the hand behavior detection module 4 automatically collects the holding pressure and monitors the holding pressure to obtain first pressure change information, when the baby is at the edge of the crib body, the foot behavior detection module 3 automatically collects the pressure on the crib body and monitors the pressure to obtain second pressure change information, and it is determined that the baby possibly climbs over the crib edge position when the baby is at the edge of the crib body. And then, carrying out infant crossing estimation analysis and judgment on the variation trend of the collected first pressure variation information and the second pressure variation information, wherein the judgment at least comprises the following three judgment conditions: the collected first pressure change information exceeds a preset first pressure threshold value; the change trend of the collected second pressure change information is reduced after increasing; and none of the foot behavior detection modules 3 simultaneously acquires information on the pressure change applied to the bed body by the baby. The acquired first pressure change information exceeds a preset first pressure threshold, namely the acquired first pressure change information is that the infant not only holds the baby crib frame of the anti-falling intelligent baby crib, but also gradually increases the pressure applied to the baby crib frame until the first pressure change information exceeds the preset first pressure threshold, and one of the judgment conditions is met; the change trend of the collected second pressure change information is gradually increased and then reduced, namely the change trend of the second pressure change information is gradually increased when the infant moves to the edge of the crib body, and then the gravity center of the infant gradually leans against the hand when trying to climb over the crib, so that the change trend of the second pressure change information is gradually reduced again to meet one of the judgment conditions; meanwhile, none of the foot behavior detection modules 3 collects pressure change information applied to the bed body by the baby, namely, the baby leaves the bed body by the two feet and is completely attached to the bed frame of the baby bed, so that the second pressure change information is zero, and one of the judgment conditions is met.
Only under the condition that the acquired information simultaneously or respectively meets the judgment conditions, an infant crossing estimation result which is matched with the acquired information and is judged that the infant possibly crosses the crib is output, and an early warning prompt is sent to the intelligent terminal 7 by the central processing module 1 through the communication module 6 based on the infant crossing estimation result, the early warning information is timely sent to a guardian who temporarily leaves or does not notice the infant, and meanwhile, the first acousto-optic information is sent by the acousto-optic prompt module 5 arranged in the middle of the anti-falling intelligent crib to attract the attention of the infant. Preferably, the first acousto-optic information comprises one or more of a pre-recorded sound called by the infant's father-mother, cartoon joyful music and a continuously changing color soft light, so as to divert the dangerous behavior of the infant trying to cross the crib by attracting the attention of the infant before the guardian arrives. Preferably, the first acousto-optic information can be input or modified through the intelligent terminal 7 operated by the guardian, and the guardian can store the acousto-optic information capable of attracting infants in the daily interaction process with the infants as the first acousto-optic information, so that visual or auditory fatigue of the infants caused by long-term use of the same acousto-optic information is avoided, and the attention of the infants is lost. Since the baby can clearly see objects within 15-30 cm before the eyes for more than 1 month, the baby can watch the objects. The phenomenon of infant vision concentration became more and more obvious by 2 months, and the infant was interested in seeing moving objects and familiar faces of adults. The visual objects can be fixed at 3 months, objects about 75cm away can be seen clearly, and the vision is about 0.1. The fixation time is significantly extended and the line of sight can also follow moving objects, e.g. a baby sleeping on a small bed, while the mother walks around, his eyes can follow the mother's body. The color vision of the infant has been greatly developed for more than 2 months, and the color and the non-color can be distinguished for more than 3 months. Infants have active visual abilities, they can see the surroundings and have a preference for vibrant colors. The newborn baby has the directional power of the existing sound from one birth, can distinguish different timbres at 4 months old, can distinguish male voice and female voice, is sensitive to the emotion expressed in the language and can generate different responses. When the infant is full of 4.5-6 months, the infant is curious about various novel sounds, can locate the sound source, and can respond Eyahatically when hearing the sound, and respond to the change of the sound volume.
According to a preferred embodiment, said infant crossing estimation analysis comprises at least the following steps: when it is determined that the current time reaches the risk level judgment period, effective pressure change information is sent to the infant crossing detection module 2 within a first time length according to the risk level of the infant crossing detection module 2 in the previous risk level judgment period. After the acquired information is subjected to infant crossing estimation analysis, the infant crossing detection module 2 outputs an infant crossing estimation result which is matched with the acquired information and is judged that the infant is likely to cross the crib, the pressure change information corresponding to the infant crossing estimation result which is determined that the infant is likely to cross the crib is effective pressure change information, and the risk level judgment is performed on the effective pressure change information when the effective pressure change information is received.
Preferably, the duration of at least one valid pressure change message of the baby crossing detection module 2 within the first duration is determined. The duration of one effective pressure change message is the duration of the effective pressure change message within a first duration, and the first duration is the time length of one risk level determination period.
Preferably, the risk level of the infant crossing detection module 2 in the current risk level determination period is determined according to the obtained duration of the at least one effective pressure change information of the infant crossing detection module 2 in the first time period. Wherein the risk levels include at least a first risk level having a first risk level threshold, a second risk level having a second risk level threshold, and a third risk level. The first risk level threshold is less than the second risk level threshold, and both the first risk level threshold and the second risk level threshold are less than the first length of time.
Because the crib body has certain degree of difficulty to the baby that the height is not enough and strength is less, it is a lagged and comparatively slow process that the baby wants to overturn out of the crib body, and its result often needs to show through a period of time. Therefore, the invention divides the corresponding risk grade after the estimation and analysis of the baby crossing through the proposed baby crossing risk grade, and judges the risk grade of the baby under the corresponding condition so as to determine whether external intervention or alarm processing is needed, thereby ensuring the safety and controllability of the baby in the intelligent crib.
According to a preferred embodiment, the infant crossing prediction analysis further comprises the steps of: if the obtained duration in the current risk level judgment period is greater than the first risk level threshold, determining that the infant crossing detection module 2 is at a first risk level in the current risk level period, and after the first duration, instructing the central processing module 1 to send an early warning prompt to an intelligent terminal 7 operated by a guardian according to the first risk level by the infant crossing detection module 2; if the obtained duration in the current risk level judgment period is greater than the second risk level threshold, determining that the infant crossing detection module 2 is at the second risk level in the current risk level period, and after the first duration, instructing the central processing module 1 to send an early warning prompt to an intelligent terminal 7 operated by a guardian according to the second risk level by the infant crossing detection module 2, and continuously sending first acousto-optic information through an acousto-optic prompt module 5 located at the first position from the first time to the second time.
The continuous at least two risk grade judging cycles are determined as the second risk grade, the current moment is the first moment, the bed body edge is often held by the baby to exercise walking or turn over to the bed body edge, and the first acousto-optic information is played when the second risk grade is determined by setting the continuous at least two risk grade judging cycles, so that the frequency of playing the first acousto-optic information is reduced, more importantly, the frequency of listening to the first acousto-optic information by the baby is reduced, and the continuous attraction of the first acousto-optic information to the baby is further ensured. The second moment in time when the current risk level cycle degrades to the first risk level and any time thereafter, i.e. the playing of the first acousto-optic message is continued at least until the infant is guided into the safe area out of the danger situation. Meanwhile, at least one first position is determined according to the current position of the baby on the intelligent bed collected at the first moment, and the first position can be a position which is opposite to the current position and is arranged on the intelligent bed and provided with an acousto-optic prompt module.
According to a preferred embodiment, if the obtained duration in the current risk level determination period is greater than the second risk level threshold, it is determined that the infant crossover detection module 2 is at the second risk level in the current risk level period, high risk analysis is performed according to a first occurrence time of the first risk level and a second occurrence time of the second risk level to output a high risk time period matched with the first risk level, and after the first duration, the infant crossover detection module 2 instructs the central processing module 1 to send an early warning prompt to the intelligent terminal 7 operated by the guardian according to the second risk level and the high risk time period.
According to the invention, the risk level of the baby crossing the bed body of the crib corresponding to different daily moments is evaluated and predicted in advance, so that the occurrence probability of the accident that the baby crosses the bed body of the crib can be further reduced, and the emergency remedial treatment after the original crossing accident is changed into early warning and precaution, so that the safety of the baby in the intelligent crib is well guaranteed. Preferably, the first occurrence time of the first risk level may be a time of occurrence of the first risk level of the infant crossover detection module 2 in a current risk level cycle determined within 24H of a day, and the second occurrence time of the second risk level may be at least a time of occurrence of the second risk level of the infant crossover detection module 2 in the current risk level cycle determined within 24H of a day, and the high risk analysis process is to overlap a plurality of the first occurrence times and a plurality of the second occurrence times, generate at least one high risk time slot with the highest coincidence rate, and generate the coincidence rate corresponding to each time slot in 24H of a day. By providing the high-risk time period for the guardian in advance, the guardian can master the time period with the highest infant crossing risk in one day, and the guardian is reminded in the time period to enhance the monitoring degree of the infant activity, so that the aim of preventing the infant crossing time in advance can be fulfilled, and the better care auxiliary effect is ensured while the risk of the infant crossing is effectively reduced.
According to a preferred embodiment, the infant crossing detection module 2 at least comprises a data receiving unit 201, a data storage unit 203 and a data comparison unit 202, wherein the data receiving unit 201 is used for acquiring pressure change information including first pressure change information and second pressure change information when at least one specific event occurs and outputting the pressure change information to the data comparison unit 202, the data storage unit 203 is used for acquiring an infant crossing estimation model from an infant configuration profile connected with the central processing module 1 and outputting the infant crossing estimation model to the data comparison unit 202, and the data comparison unit 202 is used for performing infant crossing estimation analysis on the received pressure change information and the infant crossing estimation model to obtain an infant crossing estimation result and outputting the infant crossing estimation result to the central processing module 1.
Preferably, the specific event includes a case where the pressure variation information satisfies the determination condition while obtaining the pressure variation information through infant crossing prediction analysis of the collected pressure variation information. The infant crossing estimation analysis at least comprises the following three judgment conditions: the acquired first pressure change information exceeds a preset first pressure threshold, wherein the preset first pressure threshold is set according to basic infant information stored in an infant configuration profile, the infant configuration profile at least comprises basic information such as periodically updated infant height, weight, age and medical history and high length and width of the crib, for example, the height or the first pressure change information of an infant is far smaller than the height of a bedstead or the weight of the infant, namely, the infant is judged not to overturn out of the crib even if the infant tries to overturn the crib; the change trend of the collected second pressure change information is reduced after increasing, pressure change information is generated on a pressure sensor and the like arranged at the edge of the bed because the infant tries to climb out of the crib and is necessarily close to the edge of the bed, the change trend of the pressure change information is gradually increased when the infant is close to the edge of the bed, and the change trend is gradually reduced under the state that the body of the bedstead is held by hands and is semi-suspended; meanwhile, none of the foot behavior detection modules 3 collects the pressure change information applied to the bed body by the baby, namely the baby needs to leave the double bodies through the feet in the process of trying to overturn out of the baby bed, and no pressure change information is generated on any one of the foot behavior detection modules 3 at this time.
According to a preferred embodiment, the anti-falling intelligent baby crib further comprises a plurality of bed leaving detection modules 8 arranged on the crib frame, wherein the bed leaving detection modules 8 are configured to execute the following steps: when at least one specific event occurs, tracking video frame information in the anti-falling intelligent crib through at least one of the plurality of bed exit detection modules 8 to obtain an actual baby crossing result and current time information, and instructing the central processing module 1 to output the actual baby crossing result to the baby crossing detection module 2 and/or sending out second sound and light information through the sound and light prompt module 5 based on the actual baby crossing result. Preferably, the second acoustic light information includes at least an alarm sound that can alert the guardian.
According to a preferred embodiment, if the infant crossing detection module 2 with the second risk level in the current risk level period receives the actual infant crossing result containing the information that the infant has crossed the bed, the infant crossing detection module 2 is upgraded to the third risk level in the current risk level period, and after the second time period, the infant crossing detection module 2 instructs the central processing module 1 to integrate the acquired current time and the second position into alarm information according to the third risk level and sends the alarm information to at least one medical care service platform. And the second time length is an interval from the moment when the current risk level cycle is determined to be the third risk level to the moment when the sensor arranged on the bed body does not detect the infant image. Preferably, the second duration may also be an interval from when it is determined that the current risk level period is a third risk level to when the bed leaving detection module 8 generates the baby bed leaving information. The risk of the baby crossing after being upgraded to the third risk level reaches the highest, and the baby can not acquire the baby image when the baby crosses the bed body and falls off, so that the alarm is given to at least one medical service platform. The present moment is locating promptly when the bed body does not detect the baby's image, the service platform of doctorsing and nurses can be for nearest community street police service or other hospitals, the second position is the geographical position information at intelligence crib place, to sum up integrates into the alarm information that has time place personage incident, can take effectual remedial measure at the very first time that the baby fell.
According to a preferred embodiment, the infant-crossing detection module 2 further comprises a data-updating module 9, the data-updating module 9 being configured to perform the following steps: the actual infant crossing result input by the out-of-bed detection module 8 is compared and analyzed with the infant crossing estimation result input by the data comparison unit 202 and corresponding to the current time information, and under the condition that the comparison between the actual infant crossing result and the infant crossing estimation result fails, the training and updating process of the infant crossing estimation model stored in the infant configuration profile is completed based on the actual infant crossing result and the pressure change information. Preferably, the infant configuration profile stores basic information about the height, weight, age, activity and the like of the infant and corresponding infant crossing estimation information, the infant crossing estimation information can be stored in a manner corresponding to the pressure change information, and after an actual infant crossing result is acquired and the comparison between the actual infant crossing result and the infant crossing estimation result fails, the infant crossing estimation result, which is stored in history and is under the pressure change information under the condition is updated, so that the infant crossing estimation model is further corrected and trained.
According to a preferred embodiment, the infant crossing prediction model is configured to: the actual baby crossing result including the video frame information, the current time information, and the pressure change information input by the out-of-bed detection module 8 are acquired. And preprocessing the video frame information according to a matching method and obtaining the matching rate of the images in the anti-falling intelligent crib. Preferably, the matching method is template matching, and is a method for finding a specific target image block in a source image, the target image block is slid on an input source image, an actual source image block and a target image are matched, the matching rate of the source image and the target image can be obtained every time the target image block and the actual source image block are slid, and the matching rate is higher when the source image block is more similar to the target image.
Extracting matches for a time sequence based on video frame information and current time informationMatching rate is used as classification characteristic and the obtained matching rate sequence is used as a characteristic vector X. Preferably, the period of time is the time corresponding to 16-21 frames of images; the number of the image frames is preferably 18, and the matching rate sequence of 18 gate images is taken as a feature vector for classification and is used for identifying the behavior that the baby tries to cross; the feature vector is 1X 18 dimensional, i.e. (X) ═ X1,X2,...,Xi) And i is 18. And obtaining an infant crossing estimation result based on an infant crossing estimation model established by the feature vector X and the classification discriminant function, and storing the infant crossing estimation result and the pressure change information to an infant configuration profile in a manner of being related to time information.
According to a preferred embodiment, the matching method is configured to: the target image block is slid on the input source image, the actual source image block is matched with the target image, and the matching rate of the source image and the target image can be obtained every time the target image block is slid. Preferably, the matching method is a method for finding a specific target image block in a source image, such as template matching. According to a preferred embodiment, the functional formula of the classification discriminant function is:
wherein X ═ X (X)1,X2,...,XN) Is an N-dimensional feature vector; w is aiIs the ith class; p (w)i) Is the prior probability of class i; p (X | w)i) Is a conditional probability density function; n is the number of classification categories, and N is 4;
is an N-dimensional mean vector of the i-th class;
is an N-dimensional covariance matrix; si -1Is SiThe inverse matrix of (d); i SiL is SiDeterminant (c). Preferably, the obtained real-time feature vector X is substituted into the classification discriminant function if max { h }i(X),i＝1,2,...,4}＝h1(X), at this time, an infant crossing estimation result for judging that the infant may cross the crib is output. If the baby crossing estimation result that the baby possibly crosses the crib is output, alarming is carried out on the guardian through the communication module, alarming information is sent out through the acousto-optic prompting module, and the time of the event and the video frame of the baby crossing the crib are stored; if the baby crossing estimation result which judges that the baby can not cross the crib is output, the step that the baby crossing estimation model is configured and executed is repeated, and the detection and the monitoring are continued.
According to a preferred embodiment, the classification categories comprise at least: max { h }i(X),i＝1,2,...,4}＝h1(X), the matching rate in the feature vector is increased from low to high, the matching rate in the feature vector is decreased from high to low, and other matching rate feature vectors. Since the classification of the samples is very important in the bayesian decision method, if the classification is not well done, the error will be very large. The principle of classification is to classify feature vectors into classes separately, close to the feature vectors of the event that the baby tries to flip over, to reduce misclassification.
According to a preferred embodiment, the hand behavior detection module 4 comprises a pressure sensor or an infrared temperature sensor, the foot behavior detection module 3 comprises one or more of a pressure sensor, an infrared temperature sensor and a displacement sensor, and the bed leaving detection module 8 comprises a camera or an image collector. The data receiving unit 201 comprises one or more of a microprocessor, a single chip microcomputer, an application specific integrated circuit, a special chip and/or a computing server, the data storage unit 203 comprises one or more of a semiconductor memory device, a magnetic disk and a magneto-optical disk, and the data comparison unit 202 comprises one or more of a comparator, a numerical comparison chip or a comparison integrated chip.
It should be noted that the above-mentioned embodiments are exemplary, and that those skilled in the art, having benefit of the present disclosure, may devise various arrangements that are within the scope of the present disclosure and that fall within the scope of the invention. It should be understood by those skilled in the art that the present specification and figures are illustrative only and are not limiting upon the claims. The scope of the invention is defined by the claims and their equivalents.
Claims (9)
1. An anti-falling intelligent baby crib at least comprising a baby crib body and a baby crib frame fixedly connected with the baby crib body, characterized in that the anti-falling intelligent baby crib further comprises a baby crossing detection module (2) arranged on the baby crib frame and connected with a central processing module (1), a plurality of hand behavior detection modules (4) used for collecting first pressure change information and a plurality of foot behavior detection modules (3) used for collecting second pressure change information and acquiring related specific events, wherein the baby crossing detection module (2) is configured to execute the following steps:
when at least one specific event occurs, the pressure change information of the anti-falling intelligent baby crib is obtained and recorded through at least one of the foot behavior detection modules (3) and at least one of the hand behavior detection modules (4) when the specific event occurs, baby crossing estimation analysis is carried out on the basis of the received pressure change information, and a matched baby crossing estimation result and/or baby crossing risk level are/is output to the central processing module (1), the central processing module (1) sends out early warning prompt to an intelligent terminal (7) operated by a guardian on the basis of the received baby crossing estimation result and/or baby crossing risk level and/or sends out first acousto-optic information through an acousto-optic prompt module (5),
the infant crossing estimation analysis at least comprises the following steps:
when the current time reaches a risk grade judgment period, receiving the pressure change information according to the risk grade of the infant crossing detection module (2) in the last risk grade judgment period in a first time length;
determining a duration of at least one effective pressure change message of the infant crossover detection module (2) within the first duration;
determining the infant crossing risk level of the infant crossing detection module (2) in the current risk level determination period according to the obtained duration of the at least one effective pressure change information of the infant crossing detection module (2) in the first duration.
2. The anti-fall intelligent crib of claim 1, wherein the infant crossing prediction analysis further comprises the steps of:
if the obtained duration in the current risk level judgment period is greater than a first risk level threshold value, determining that the infant crossing detection module (2) is at a first risk level in the current risk level period, and after the first duration, indicating the central processing module (1) to send an early warning prompt to an intelligent terminal (7) operated by a guardian by the infant crossing detection module (2) according to the first risk level;
if the obtained duration in the current risk level judgment period is greater than a second risk level threshold, determining that the infant crossing detection module (2) is at a second risk level in the current risk level period, indicating the central processing module (1) to send an early warning prompt to an intelligent terminal (7) operated by a guardian according to the second risk level by the infant crossing detection module (2) after the first duration, and continuously sending first acousto-optic information through an acousto-optic prompt module (5) located at a first position from a first moment to a second moment.
3. The crash-proof intelligent crib according to claim 2, wherein if the obtained duration in the current risk level determination period is greater than the second risk level threshold, it is determined that the baby rollover detection module (2) is at the second risk level in the current risk level period, and a high risk analysis is performed according to a first occurrence time of the first risk level and a second occurrence time of the second risk level to output a high risk time period matching therewith, after the first duration, the baby rollover detection module (2) instructs the central processing module (1) to send an early warning prompt to the intelligent terminal (7) operated by the guardian according to the second risk level and the high risk time period.
4. The crash-proof smart crib according to claim 3, wherein said baby crossing detection module (2) comprises at least a data receiving unit (201), a data storage unit (203) and a data comparison unit (202), wherein,
the data receiving unit (201) is configured to obtain the pressure change information including the first pressure change information and the second pressure change information when at least one specific event occurs and output the pressure change information to the data comparing unit (202), the data storage unit (203) is configured to obtain an infant crossing prediction model from an infant configuration profile and output the infant crossing prediction model to the data comparing unit (202), the data comparing unit (202) is configured to perform infant crossing prediction analysis on the received pressure change information and the infant crossing prediction model to obtain an infant crossing prediction result and output the infant crossing prediction result to the central processing module (1),
the first pressure change information is obtained by automatically collecting and monitoring the holding pressure of the baby through the hand behavior detection module (4), and the second pressure change information is obtained by automatically collecting and monitoring the pressure on the baby crib body through the foot behavior detection module (3).
5. The fall-resistant intelligent crib of claim 4, further comprising a number of bed exit detection modules (8) provided on the crib frame, the bed exit detection modules (8) being configured to perform the steps of:
tracking video frame information in the anti-falling intelligent baby crib by at least one of a plurality of bed-leaving detection modules (8) when at least one specific event occurs to obtain an actual baby crossing result and current time information, and instructing the central processing module (1) to output the actual baby crossing result to the baby crossing detection module (2) and/or sending out second acousto-optic information by the acousto-optic prompt module (5) based on the actual baby crossing result.
6. The crash-proof intelligent infant bed according to claim 5, wherein if the infant crossing detection module (2) with the second risk level in the current risk level cycle receives the actual infant crossing result containing the information that the infant has crossed the bed, the infant crossing detection module (2) is upgraded to the third risk level in the current risk level cycle, and after the second time period, the infant crossing detection module (2) instructs the central processing module (1) to integrate the collected current time and the second position into alarm information according to the third risk level and sends the alarm information to at least one medical service platform,
wherein the second position is used for representing the geographical position of the intelligent baby crib.
7. The crash-proof smart crib of claim 6, wherein said baby crossing detection module (2) further comprises a data update module (9), said data update module (9) being configured to perform the steps of:
comparing and analyzing the actual infant crossing result input by the out-of-bed detection module (8) with the infant crossing estimation result input by the data comparison unit (202) and corresponding to the current time information, and finishing a training and updating process of the infant crossing estimation model stored in the infant configuration profile based on the actual infant crossing result and the pressure change information under the condition that the comparison of the actual infant crossing result and the infant crossing estimation result fails.
8. The anti-fall intelligent crib of claim 5, wherein the infant rollover prediction model is configured to:
acquiring the actual infant crossing result including the video frame information, the current time information and the pressure change information input by the out-of-bed detection module (8);
preprocessing the video frame information according to a matching method and obtaining the matching rate of the images in the anti-falling intelligent crib;
extracting the matching rate of a period of time sequence based on the video frame information and the current time information as a classification feature and taking the obtained matching rate sequence as a feature vector X;
and obtaining the infant crossing estimation result based on the infant crossing estimation model established by the feature vector X and the classification discriminant function, and storing the infant crossing estimation result and the pressure change information to the infant configuration profile in a time information correlation mode.
9. The fall-resistant smart crib of claim 8, wherein the matching method is configured to: the target image block is slid on the input source image, the actual source image block is matched with the target image, and the matching rate of the source image and the target image can be obtained every time the target image block is slid.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811653925.0A CN109757928B (en) | 2018-12-29 | 2018-12-29 | Prevent falling intelligent crib |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811653925.0A CN109757928B (en) | 2018-12-29 | 2018-12-29 | Prevent falling intelligent crib |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109757928A CN109757928A (en) | 2019-05-17 |
| CN109757928B true CN109757928B (en) | 2022-01-25 |
Family
ID=66453232
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811653925.0A Active CN109757928B (en) | 2018-12-29 | 2018-12-29 | Prevent falling intelligent crib |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109757928B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110833397A (en) * | 2019-11-25 | 2020-02-25 | 深圳市码影科技有限公司 | Intelligent bed foot detection method, system and device based on Internet of things |
| CN111462922A (en) * | 2020-03-27 | 2020-07-28 | 南通大学附属医院 | Evaluation method and evaluation system for reasonable body constraint of ICU adult catheterization patient |
| JP7406737B2 (en) * | 2020-03-31 | 2023-12-28 | インフィック株式会社 | Bed leaving prediction notification device and program |
| CN116612603A (en) * | 2023-07-17 | 2023-08-18 | 微网优联科技(成都)有限公司 | Mobile object monitoring system and method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202489609U (en) * | 2012-01-19 | 2012-10-17 | 中国矿业大学 | Automatic alarm children bed |
| CN204071330U (en) * | 2014-09-24 | 2015-01-07 | 中国人民解放军第三军医大学 | Anti-fall bed medical care mattress |
| CN106343826A (en) * | 2016-11-29 | 2017-01-25 | 天津创潮至上科技发展有限公司 | Manufacturing process for intelligent anti-falling structure of baby crib |
| CN206924156U (en) * | 2017-01-10 | 2018-01-26 | 新疆维吾尔自治区人民医院 | A kind of ward nursing bed of intelligent alarm |
| CN107248261A (en) * | 2017-06-27 | 2017-10-13 | 苏州见真物联科技有限公司 | A kind of intelligent safety protection device for preventing that baby from falling |
-
2018
- 2018-12-29 CN CN201811653925.0A patent/CN109757928B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109757928A (en) | 2019-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109757928B (en) | Prevent falling intelligent crib | |
| JP6978111B2 (en) | Identification of fall risk using machine learning algorithms | |
| CN104799826B (en) | A kind of intelligence Ankang service system and the reliable detection method of alarm | |
| US6968294B2 (en) | Automatic system for monitoring person requiring care and his/her caretaker | |
| JP6823160B2 (en) | Devices, systems, and methods for patient monitoring to predict and prevent bed falls | |
| CN107920754A (en) | Monitor infant's emotional state and determine the physiological measure associated with infant | |
| Yuan et al. | Power-efficient interrupt-driven algorithms for fall detection and classification of activities of daily living | |
| US20160310067A1 (en) | A baby monitoring device | |
| CN109480868B (en) | Intelligent infant monitoring system | |
| JP2004523849A (en) | An automated system that monitors people living alone who need assistance on an irregular basis | |
| CN107851356A (en) | Determine the posture of infant and wearable the infant's monitoring device and system of motion | |
| CN108209871A (en) | Sleep monitor method, apparatus, system and electronic equipment | |
| Naranjo-Hernandez et al. | Personalization and adaptation to the medium and context in a fall detection system | |
| CN107411753A (en) | A kind of wearable device for identifying gait | |
| Dovgan et al. | Intelligent elderly-care prototype for fall and disease detection | |
| CN108012560A (en) | Intelligent infant monitors system and infant monitoring center and infant learns acceptance detecting system | |
| CN109643585A (en) | For carrying out patient-monitoring to predict and prevent the equipment, system and method for falling from bed | |
| CN108024712A (en) | Prediction infant sleeping sleep mode simultaneously derives infant's model based on the observation result associated with infant | |
| CN114668388A (en) | Intelligent elderly health monitoring method, device, terminal and storage medium | |
| CN109758281B (en) | Safety system based on body position adjustment | |
| CN107924643A (en) | Remotely assemble and analyze the measurement data that system is monitored from multiple infants | |
| CN112262416B (en) | Nursing auxiliary system for nursing person | |
| TW202133798A (en) | Care system and method of automated care | |
| JP2020053070A5 (en) | ||
| TWI731673B (en) | Image sleep analysis method and system thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |