Intelligent induction-based child medical limiting operation method and system

Abstract

The application discloses a child medical limiting operation method and system based on intelligent sensing, relates to the technical field of VR auxiliary limiting control, and obtains and recognizes a target child limb reference point, a first reference mark position and a second reference mark position in an initial frame image and a current frame image through a VR acquisition device, determines a limiting axis based on the two reference marks, calculates a ratio of a directed projection length of the target child limb reference point on the limiting axis to a distance between the two reference marks, obtains an initial relative limiting coefficient and a current relative limiting coefficient, obtains a limiting deviation coefficient according to a difference between the two, determines an adjustment distance and an adjustment direction, generates limiting adjustment prompt data and sends the limiting adjustment prompt data to a VR display device. Thus, the actual deviation of the target child limb relative to a limiting component can be quantified.

Description

Technical Field

[0001] This invention relates to the field of VR-assisted limit control technology, and more specifically, this application relates to a child medical limit operation method and system based on intelligent sensing. Background Technology

[0002] In pediatric medical procedures, restraint components are used to maintain limb posture. However, children's limbs are small and their movements are sudden. Furthermore, the flexible restraint structure and soft tissue will undergo compression, rebound, and synchronous movement. Even when the restraint component is still in place, the actual relative position between the limb and the restraint component may have already changed.

[0003] Existing technologies often rely on absolute position in a single image, the fixed state of the limiting component, or simple displacement thresholds to determine the effectiveness of the limiting mechanism. When the acquisition angle, imaging scale, or the overall position of the limiting component changes, absolute displacement makes it difficult to distinguish between limb movement with the component and limb slippage relative to the component. At the same time, static references may be affected by flexible settlement, and single-shot threshold judgments cannot take into account both short-term reciprocating movements and slow cumulative slippage. Therefore, the generated adjustment prompts are prone to problems such as unclear direction and mismatch between distance and actual recovery amount.

[0004] To address the aforementioned issues, there is an urgent need in this field for a technical solution capable of continuously quantifying the actual positional changes of a limb relative to the effective restraint segment in pediatric medical restraint scenarios, and translating these changes into executable adjustments of direction and distance. The core problem with existing technologies lies in the difficulty of accurately identifying and quantifying the true relative offset under the objective condition that the flexible restraint component and the child's limb may move, sink, or reciprocate synchronously. Summary of the Invention

[0005] To address the aforementioned technical problems, this technical solution provides a child medical restraint operation method and system based on intelligent sensing, which solves the problems mentioned in the background section.

[0006] In a first aspect, embodiments of this application provide a child medical restraint operation method based on intelligent sensing, comprising the following steps: acquiring and identifying a target child limb reference point, a first reference marker position representing the start of the effective restraint segment, and a second reference marker position representing the end of the effective restraint segment in an initial frame image and a current frame image, respectively, using a VR acquisition device; in the initial frame image and the current frame image, taking the direction from the first reference marker position to the second reference marker position as the restraint axis direction, projecting the target child limb reference point onto the corresponding restraint axis, calculating the ratio of the directed projection length from the first reference marker position to the projection point to the distance between the two reference marker positions, and obtaining the initial relative restraint coefficient and the current relative restraint coefficient. Calculate the difference between the current relative limit coefficient and the initial relative limit coefficient to obtain the limit offset coefficient; if the absolute value of the limit offset coefficient is less than the preset offset threshold, the adjustment distance of the limit component is recorded as zero; if the absolute value of the limit offset coefficient is not less than the preset offset threshold, the absolute value of the limit offset coefficient is multiplied by the distance between the two reference marks in the current frame image to obtain the adjustment distance of the limit component, and when the limit offset coefficient is positive, the adjustment direction is recorded as from the first reference mark position to the second reference mark position, and when the limit offset coefficient is negative, the adjustment direction is recorded as from the second reference mark position to the first reference mark position; generate limit adjustment prompt data according to the adjustment direction and adjustment distance and send it to the VR display device.

[0007] Secondly, embodiments of this application provide a child medical restraint operating system based on intelligent sensing, including: a VR acquisition module: used to acquire and identify the target child limb reference point, the position of the first reference marker representing the start end of the effective restraint segment, and the position of the second reference marker representing the end end of the effective restraint segment in the initial frame image and the current frame image respectively through a VR acquisition device; a restraint coefficient processing module: used to project the target child limb reference point onto the corresponding restraint axis in the initial frame image and the current frame image, respectively, with the direction from the first reference marker position to the second reference marker position as the restraint axis direction, and calculate the ratio of the directed projection length from the first reference marker position to the projection point to the distance between the two reference marker positions, to obtain the initial relative restraint coefficient and the current relative restraint coefficient; an offset coefficient processing module: used to calculate the current... The difference between the previous relative limit coefficient and the initial relative limit coefficient is used to obtain the limit offset coefficient; the first processing module of the limit component is used to record the adjustment distance of the limit component as zero if the absolute value of the limit offset coefficient is less than the preset offset threshold; the second processing module of the limit component is used to multiply the absolute value of the limit offset coefficient with the distance between the two reference marks in the current frame image to obtain the adjustment distance of the limit component if the absolute value of the limit offset coefficient is not less than the preset offset threshold, and record the adjustment direction as from the first reference mark position to the second reference mark position when the limit offset coefficient is positive, and record the adjustment direction as from the second reference mark position to the first reference mark position when the limit offset coefficient is negative; the sending and receiving module is used to generate limit adjustment prompt data according to the adjustment direction and adjustment distance and send it to the VR display device.

[0008] Thirdly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described intelligent sensing-based child medical restraint operation method.

[0009] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0010] 1. During the medical restraint procedure for children, the initial relative restraint coefficient and the current relative restraint coefficient are obtained by using the target child's limb reference point, the position of the first reference marker and the position of the second reference marker in the initial frame image and the current frame image, respectively. This allows the position determination to be based on the effective restraint segment of the restraint component, rather than relying solely on the absolute image position of the target child's limb, thereby more accurately representing the actual positional change of the target child's limb relative to the restraint component.

[0011] 2. In scenarios involving slight movement of a child's limbs, changes in the overall position of the limiting component, or changes in the VR acquisition distance, the positional changes in the image can be normalized by calculating the ratio of the directional projection length to the distance between the positions of the two reference markers. This reduces the impact of changes in imaging scale and overall movement of the limiting component on the limiting judgment results, thereby improving the stability of the limiting offset recognition results during continuous medical operations.

[0012] 3. When the target child's limb actually deviates from the limiting component, the limiting offset coefficient is obtained by calculating the difference between the current relative limiting coefficient and the initial relative limiting coefficient. The absolute value of the limiting offset coefficient is compared with the preset offset threshold. This can distinguish between position fluctuations within the allowable range and limiting offsets that need to be adjusted, reducing unnecessary adjustment prompts due to the child's short-distance activities.

[0013] 4. In scenarios where the limit components need to be rearranged, the adjustment distance is obtained by multiplying the absolute value of the limit offset coefficient by the distance between the two reference marker positions in the current frame image, and the adjustment direction is determined according to the sign of the limit offset coefficient. The relative position change can be transformed into an adjustment distance and adjustment direction with practical operational significance, so that the generated limit adjustment prompt data can provide medical staff with a clear and actionable reference for rearranging the limit components. Attached Figure Description

[0014] Figure 1 A schematic diagram illustrating the steps of a child medical restraint operation method based on intelligent sensing provided in an embodiment of this application;

[0015] Figure 2 A schematic diagram of the logic flow of the child medical restraint operation method based on intelligent sensing provided in the embodiments of this application;

[0016] Figure 3 This is a schematic diagram of the structure of a child medical limiting operating system based on intelligent sensing, provided in an embodiment of this application. Detailed Implementation

[0017] This application's embodiments address the technical problem in the prior art of accurately quantifying the actual offset of the target child's limb relative to the effective limiting segment of the limiting component, and generating adjustment directions and distances that can be used to rearrange the limiting component, in medical limiting scenarios where the child's limb and flexible limiting component may move together and the VR image acquisition scale changes.

[0018] During pediatric medical procedures, while restraint components can provide support and constraint for the child's limbs, the small size of children's limbs, the sudden nature of their movements, and the flexible contact between the restraint components and soft tissue mean that the child's limb may slip relative to the effective restraint segment while the restraint components are still in place. Current position determination methods typically rely on absolute coordinates in images or single displacement results. When the distance between the VR acquisition device and the target area changes, or when the restraint components and the child's limb move together as a whole, the absolute position change cannot accurately reflect the true relative offset between them, making it difficult to provide a clear basis for adjustment.

[0019] This solution first acquires the initial frame image when the limiting component is initially deployed and the current frame image during the medical operation. From these two types of images, it identifies the target child's limb reference point and the positions of the first and second reference markers, representing the start and end points of the effective limiting segment, respectively. Since the two reference markers are fixed to the same effective limiting segment, their connection reflects the current extension direction of the limiting component, and their spacing reflects the relative scale in the current image. Therefore, this solution determines the limiting axis direction in both the initial and current frame images by pointing from the first reference marker position to the second reference marker position. The target child's limb reference point is projected onto the corresponding limiting axis, and the ratio of the directed projection length from the first reference marker position to the projection point to the distance between the two reference marker positions is calculated to obtain the initial relative limiting coefficient and the current relative limiting coefficient.

[0020] This process converts the position of the target child's limb into a dimensionless positional relationship relative to the effective limiting segment. This ensures that when the limiting component moves as a whole or the image acquisition scale changes, the constant relative relationship between the target child's limb and the limiting component will not be misjudged as limiting offset. Subsequently, the difference between the current relative limiting coefficient and the initial relative limiting coefficient is calculated to obtain the limiting offset coefficient. The absolute value of the limiting offset coefficient quantifies the degree to which the target child's limb deviates from the initial limiting position, and the positive or negative sign of the limiting offset coefficient represents the direction of offset.

[0021] When the absolute value of the limit offset coefficient is less than the preset offset threshold, it indicates that the current position change is still within the allowable range, and the adjustment distance of the limit component is recorded as zero. When the absolute value of the limit offset coefficient is not less than the preset offset threshold, the absolute value of the limit offset coefficient is multiplied by the distance between the two reference marker positions in the current frame image, so that the dimensionless relative offset is restored to an adjustment distance with actual length significance, and the adjustment direction along the effective limit segment is determined according to the sign of the limit offset coefficient. Finally, limit adjustment prompt data is generated based on the adjustment direction and adjustment distance and sent to the VR display device, so that the actual offset of the target child's limb relative to the limit component can be transformed into a directly referable basis for the rearrangement of the limit component.

[0022] The core innovation of this solution lies in using the effective limiting segment of the limiting component itself as both a directional and scale reference. This transforms changes in limb position in the image into normalized position changes relative to the limiting component, and further restores these to adjustment directions and distances with practical operational significance. This solves the problem of difficulty in distinguishing between the overall movement of the limiting component, changes in imaging scale, and actual limb slippage in pediatric medical limiting scenarios.

[0023] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0024] like Figure 1 The diagram shown is a step-by-step illustration of the child medical limiting operation method based on intelligent sensing provided in this application embodiment. This embodiment can use the effective limiting segment of the limiting component as the relative position judgment benchmark, reducing the impact of the overall movement of the limiting component and changes in the image acquisition scale on the limiting offset judgment.

[0025] The target child's limb reference points, the first reference marker position representing the start of the effective restraint segment, and the second reference marker position representing the end of the effective restraint segment are acquired and identified in the initial frame image and the current frame image respectively through VR acquisition equipment.

[0026] The preset sampling period is set based on the image acquisition frequency of the VR acquisition device and the duration of the target child's limb position change that needs to be identified. When the VR acquisition device continuously outputs images, it can select the current frame image at fixed frame intervals, so that there is a definite sampling order between different current frame images.

[0027] The target child's limb reference point is the location of a key human body point corresponding to a preset limb part type. The preset limb part type is configured before the medical procedure begins and is used to select the target child's limb reference point from the human body key point recognition results. The first and second reference marker positions are represented by the center positions of their corresponding preset graphic codes.

[0028] In the initial frame image and the current frame image, the direction from the first reference mark position to the second reference mark position is taken as the direction of the limiting axis. The reference point of the target child's limb is projected onto the corresponding limiting axis. The ratio of the directed projection length from the first reference mark position to the projection point to the distance between the two reference mark positions is calculated to obtain the initial relative limiting coefficient and the current relative limiting coefficient.

[0029] Figure 2This is a schematic diagram of the logic flow of a child medical restraint operation method based on intelligent sensing provided in an embodiment of this application. Specifically, for an initial frame image, the target child limb reference point, the position of the first reference marker, and the position of the second reference marker in the initial frame image are read. The position of the first reference marker in the initial frame image is used as the starting reference for the restraint axis, and the direction from the first reference marker position to the second reference marker position is used as the positive direction of the restraint axis. The restraint axis corresponding to the initial frame image is determined based on the positions of the first and second reference markers.

[0030] The target child's limb reference point in the initial frame image is projected onto the corresponding limiting axis along a direction perpendicular to the limiting axis, thus obtaining the projection point corresponding to the initial frame image. The directed projection length from the position of the first reference mark to the projection point along the limiting axis is determined. When the projection point is located on the side of the first reference mark position facing the second reference mark position, the directed projection length is recorded as a positive value; when the projection point is located on the side of the first reference mark position away from the second reference mark position, the directed projection length is recorded as a negative value.

[0031] Calculate the distance between the first reference marker position and the second reference marker position in the initial frame image, and calculate the ratio of the directional projection length corresponding to the initial frame image to the distance between the two reference marker positions in the initial frame image. Record the calculation result as the initial relative limit coefficient.

[0032] For the current frame image, read the target child's limb reference point, the position of the first reference marker, and the position of the second reference marker in the current frame image. Use the position of the first reference marker in the current frame image as the starting reference of the limiting axis, and use the direction from the position of the first reference marker to the position of the second reference marker as the positive direction of the limiting axis. Determine the limiting axis corresponding to the current frame image based on the positions of the first and second reference markers.

[0033] Project the target child's limb reference point in the current frame image along a direction perpendicular to the limiting axis onto the corresponding limiting axis of the current frame image to obtain the projection point of the current frame image. Following the same positive and negative direction determination rules as the initial frame image, obtain the directed projection length from the position of the first reference mark in the current frame image to the projection point.

[0034] Calculate the distance between the first reference marker position and the second reference marker position in the current frame image, and calculate the ratio of the directional projection length corresponding to the current frame image to the distance between the two reference marker positions in the current frame image. Record the calculation result as the current relative limit coefficient.

[0035] The initial frame image and the current frame image independently complete the determination of the limiting axis, projection of the target child's limb reference point, calculation of the directed projection length and ratio in their respective image coordinate systems. The position data in the initial frame image does not directly participate in the calculation of the current relative limiting coefficient, nor does the position data in the current frame image directly participate in the calculation of the initial relative limiting coefficient.

[0036] This technique does not directly compare the absolute coordinates of the target child's limb reference points in the initial frame image and the current frame image. Instead, it establishes a relative position reference corresponding to the effective limiting segment based on the first reference marker position and the second reference marker position in each frame image.

[0037] The direction from the first reference marker to the second reference marker is used to determine the extension direction of the effective limiting segment and serves as the positive direction of the directed projection length. Projecting the target child's limb reference point onto the limiting axis allows for the extraction of the positional component of the target child's limb reference point along the effective limiting segment direction, reducing the impact of positional changes perpendicular to the effective limiting segment direction on the determination of the limiting relationship.

[0038] The directed projection length is used to represent the position and orientation of the target child's limb reference point relative to the first reference marker position along the limiting axis. When the projection point is located between the first and second reference marker positions, the corresponding relative position is within the effective limiting segment; when the projection point crosses the second reference marker position, the resulting ratio can be greater than one; when the projection point is located on the side of the first reference marker position away from the second reference marker position, the resulting ratio can be negative. This allows the retention of the directional and positional information of the target child's limb reference point relative to the effective limiting segment.

[0039] The distance between the two reference marker positions is used to characterize the current image scale of the effective restraint segment in the corresponding frame image. By calculating the ratio of the directed projection length to the distance between the two reference marker positions, the position of the target child's limb reference point along the restraint axis can be converted into a dimensionless positional relationship relative to the length of the effective restraint segment.

[0040] When the acquisition distance between the VR acquisition device and the target area changes, or when there is a difference in the imaging ratio between the initial frame image and the current frame image, the directional projection length and the distance between the two reference marker positions usually change at a similar ratio. The ratio of the two can reduce the impact of image scaling on the position judgment result. The initial relative limit coefficient is used to characterize the position of the target child's limb reference point relative to the effective limit segment when the limit component is initially deployed. The current relative limit coefficient is used to characterize the position of the target child's limb reference point relative to the effective limit segment at the current moment, thus providing input data with the same data meaning and the same scale basis for subsequent calculation of the limit offset coefficient.

[0041] Calculate the difference between the current relative limit coefficient and the initial relative limit coefficient to obtain the limit offset coefficient.

[0042] The absolute value of the limit offset coefficient is used to characterize the degree to which the target child's limb deviates from the initial limit position, and the sign of the limit offset coefficient is used to characterize the direction of the target child's limb position change along the limit axis.

[0043] If the absolute value of the limit offset coefficient is less than the preset offset threshold, the adjustment distance of the limit component will be set to zero.

[0044] The preset offset threshold is determined based on the ratio of the allowable offset distance corresponding to the effective limiting segment to the calibration distance between the first reference mark and the second reference mark. The allowable offset distance can be set according to the structural dimensions of the limiting component, the allowable range of motion of the target child's limbs, and the equipment calibration results.

[0045] If the absolute value of the limit offset coefficient is not less than the preset offset threshold, the absolute value of the limit offset coefficient is multiplied by the distance between the two reference marks in the current frame image to obtain the adjustment distance of the limit component. When the limit offset coefficient is positive, the adjustment direction is determined to be from the position of the first reference mark to the position of the second reference mark. When the limit offset coefficient is negative, the adjustment direction is determined to be from the position of the second reference mark to the position of the first reference mark.

[0046] The adjustment distance uses the same coordinate scale as the distance between the two reference marker positions in the current frame image to determine the display length or corresponding adjustment amount of the adjustment prompt in the VR display device. The adjustment distance can be reserved according to a predetermined number of bits, which is determined based on the image resolution and the minimum adjustment distance to be expressed.

[0047] Based on the adjustment direction and distance, limit adjustment prompt data is generated and sent to the VR display device.

[0048] Through the above processing, the positional changes of the target child's limbs in the image are converted into limit offset coefficients relative to the effective limit segment, and further converted into adjustment direction and adjustment distance, enabling the VR display device to output operation prompts that can be used to rearrange the limit components.

[0049] Furthermore, this implementation is used to distinguish between the gradual settling of the limiting components and the child's soft tissue after initial placement and the continuous sliding of the target child's limb in the same direction, so as to avoid incorrectly updating the initial relative limiting coefficient.

[0050] Within the preset benchmark confirmation window, the current frame image is continuously acquired according to the preset sampling period, and multiple current relative limit coefficients are obtained sequentially based on each current frame image.

[0051] The preset reference confirmation window begins after the initial frame image acquisition is completed, and its duration is determined based on the calibration time required for the limiting component to complete the initial bonding. The preset reference confirmation window includes at least several preset sampling periods capable of generating multiple current relative limiting coefficients.

[0052] The difference between two adjacent current relative limit coefficients is calculated according to the sampling order to obtain the baseline change sequence.

[0053] If the absolute values ​​of a preset number of consecutive differences in the benchmark change sequence decrease sequentially, and the absolute value of the last difference is less than a preset stable threshold, then it is determined that multiple current relative limit coefficients meet the first preset benchmark change condition. The initial relative limit coefficient is replaced with the corresponding last current relative limit coefficient, and the subsequent limit offset coefficient is calculated based on the replaced initial relative limit coefficient.

[0054] The preset stability threshold is determined based on the allowable range of change in adjacent relative restraint coefficients when the target child's limbs and the restraint components remain relatively stationary. A successive decrease in the absolute value of the difference indicates that the relative position change is gradually attenuating, and the absolute value of the last difference being less than the preset stability threshold indicates that the restraint relationship has entered a stable range.

[0055] If the signs of a preset number of consecutive differences in the benchmark change sequence are the same, and the sum of the absolute values ​​of the consecutive differences is not less than the preset sliding threshold, then it is determined that multiple current relative limit coefficients meet the second preset benchmark change condition, the initial relative limit coefficient remains unchanged, and the limit offset coefficient is updated according to the difference between the last current relative limit coefficient and the initial relative limit coefficient.

[0056] The preset slip threshold is determined by the minimum continuous slip distance to be identified and the calibrated distance of the effective limiting segment. Consecutive differences with the same sign indicate that the target child's limb is continuously changing in the same direction, and the sum of the absolute values ​​of the consecutive differences is used to characterize the cumulative degree of continuous change.

[0057] If multiple current relative limit coefficients do not meet the first and second preset reference change conditions, the current frame image continues to be acquired until the preset reference confirmation window ends. If none of the conditions are met after the preset reference confirmation window ends, the original initial relative limit coefficients remain unchanged.

[0058] Through the above processing, the initial relative limit coefficient can only be updated when the relative position change gradually decays and enters a stable range; when the relative position continues to change in the same direction, the original initial relative limit coefficient is retained, thereby avoiding the absorption of the actual slip into the new initial limit reference.

[0059] Furthermore, this implementation distinguishes between instantaneous reciprocating movements of the target child's limbs and slow, cumulative sliding movements, reducing erroneous adjustment prompts caused by instantaneous movements and identifying continuously cumulative positional changes.

[0060] After updating the initial relative limit coefficient according to the first preset benchmark change condition, when multiple limit offset coefficients obtained in the preset action judgment window meet the preset action change condition, the limit offset coefficient used to determine the adjustment distance or the current preset offset threshold is processed.

[0061] Obtain multiple limit offset coefficients formed in the sampling order within the preset action determination window, and calculate the difference between two adjacent limit offset coefficients to obtain the offset increment sequence.

[0062] The preset motion determination window is set according to the duration of the instantaneous reciprocating motion and the duration of the slow cumulative sliding motion that need to be identified.

[0063] If the offset increment sequence includes both positive and negative values, and the absolute value of the limit offset coefficient at the end of the preset action determination window is less than the preset regression threshold, then multiple limit offset coefficients are determined to meet the first preset action change condition, and the limit offset coefficient used to determine the adjustment distance is set to zero.

[0064] The offset increment sequence includes both positive and negative values, indicating that the target child's limb has reversed direction within the preset action judgment window; the absolute value of the limit offset coefficient at the end of the window is less than the preset regression threshold, indicating that the target child's limb has returned to the vicinity of the initial limit position.

[0065] If the signs of a preset number of consecutive offset increments in the offset increment sequence are the same, and the sum of the absolute values ​​of the consecutive offset increments is not less than the preset cumulative offset threshold, then it is determined that multiple limit offset coefficients meet the second preset action change condition. The preset cumulative judgment threshold that is less than the preset offset threshold is selected as the current preset offset threshold, and the adjustment distance and adjustment direction are determined according to the limit offset coefficient at the end of the preset action judgment window.

[0066] The same sign for consecutive offset increments indicates that the target child's limbs move gradually in the same direction. The sum of the absolute values ​​of the consecutive offset increments is used to characterize the cumulative degree of slow slippage. The preset cumulative judgment threshold is only used when the second preset action change condition is met, and does not replace the preset offset threshold in the normal processing state.

[0067] When multiple limit offset coefficients do not meet the first preset action change condition and the second preset action change condition, the general processing continues to be performed according to the preset offset threshold.

[0068] Through the above processing, the limit offset coefficient that reverses direction and eventually returns will not generate unnecessary adjustment distance, while the limit offset coefficient that accumulates continuously in the same direction can trigger an adjustment prompt when the accumulation judgment condition is met.

[0069] Furthermore, this implementation is used to correct subsequent adjustment distances based on the actual recovery after a single adjustment of the limiting component, thereby reducing under-adjustment or over-adjustment caused by the target child's limbs moving together with the limiting component.

[0070] When the absolute value of the limit offset coefficient at the end of the preset action judgment window is not less than the current preset offset threshold, and the limit component has been adjusted once according to the corresponding non-zero adjustment distance, the subsequent adjustment distance is corrected.

[0071] Receive adjustment completion confirmation data returned by the VR display device, and in response to the adjustment completion confirmation data, obtain the adjusted current frame image.

[0072] After the operator adjusts the limit component according to the limit adjustment prompt data output by the VR display device, they input an adjustment completion confirmation operation through the VR display device. The VR display device generates adjustment completion confirmation data based on the adjustment completion confirmation operation. Receiving the adjustment completion confirmation data before acquiring the current frame image after adjustment avoids using intermediate images during the limit component movement as the adjustment result.

[0073] The adjusted limit offset coefficient is obtained based on the adjusted current frame image.

[0074] The adjusted limit offset coefficient adopts the same processing rules as the general limit offset coefficient, and is obtained based on the difference between the adjusted current relative limit coefficient and the currently used initial relative limit coefficient.

[0075] Calculate the difference between the absolute value of the limit offset coefficient before adjustment and the absolute value of the limit offset coefficient after adjustment, and multiply the difference by the distance between the two reference marker positions in the current frame image before adjustment to obtain the actual recovery distance.

[0076] The actual recovery distance is used to characterize the actual reduction in offset distance of the target child's limb relative to the effective limiting segment after this adjustment of the limiting component.

[0077] The ratio between the actual recovery distance and the current adjustment distance is calculated to obtain the adjustment response coefficient.

[0078] The adjustment response coefficient characterizes the proportion of the actual adjustment distance converted into relative positional recovery. When the target child's limb moves together with the limiting component, the actual recovery distance is less than the adjustment distance, and the adjustment response coefficient decreases accordingly.

[0079] If the adjustment response coefficient is greater than zero and is outside the preset response range, the adjustment distance obtained based on the adjusted limit offset coefficient will be divided by the adjustment response coefficient to obtain the corrected adjustment distance, and the next limit adjustment prompt data will be updated based on the corrected adjustment distance.

[0080] The preset response range is determined based on the allowable proportional deviation between the actual recovery distance and the current adjustment distance. When the adjusted response coefficient is within the preset response range, the adjustment distance obtained based on the adjusted limit offset coefficient continues to be used, and no response correction is performed.

[0081] If the adjustment response coefficient is not greater than zero, a pause adjustment prompt is generated and sent to the VR display device.

[0082] An adjustment response coefficient not greater than zero indicates that the absolute value of the limit offset coefficient has not decreased after adjustment, or that the offset of the target child's limb relative to the effective limit segment has further increased. At this point, stop increasing the adjustment distance to avoid the accumulation of erroneous adjustments.

[0083] When the absolute value of the adjusted limit offset coefficient is less than the current preset offset threshold, the subsequent adjustment distance is set to zero, and the current adjustment response processing ends.

[0084] Through the above processing, the next adjustment distance is no longer generated solely based on the theoretical offset, but is corrected by combining the actual recovery ratio of the previous adjustment, making the limit adjustment prompt data more consistent with the actual contact response between the target child's limb and the flexible limit component.

[0085] Furthermore, this implementation method is used to ensure that the target child's limb reference point, the first reference marker position, and the second reference marker position have a clear and stable data source.

[0086] Identify different preset graphic codes set at the start and end of the effective limit segment in the initial frame image and the current frame image, respectively, and determine the center position of each preset graphic code as the first reference mark position and the second reference mark position, respectively.

[0087] The recognition of the preset graphic code can be accomplished using existing visual marker recognition technology. The first reference marker and the second reference marker use different codes to fix the relationship between the start and end of the effective limit segment.

[0088] Human key point recognition is performed on the initial frame image and the current frame image. Key point positions that match the preset limb part type are selected from the human key point recognition results and used as reference points for the target child's limbs.

[0089] Human body key point recognition can be accomplished using existing human posture recognition technology. Preset limb part types are used to select the target child's limb reference point corresponding to the current restricted part from the recognition results.

[0090] The recognition result corresponding to the initial frame image is used to calculate the initial relative limit coefficient, and the recognition result corresponding to the current frame image is used to calculate the current relative limit coefficient.

[0091] Furthermore, this embodiment is used to convert the position of the target child's limb reference point into a directional relative position along the direction of the effective limiting segment.

[0092] In the image coordinate systems corresponding to the initial frame image and the current frame image, the limiting axis passing through the positions of the first reference mark and the second reference mark is determined.

[0093] Using the direction from the first reference mark position to the second reference mark position as the positive direction of the limiting axis, the reference point of the target child's limb is projected onto the limiting axis along a direction perpendicular to the limiting axis to obtain the projection point.

[0094] The above projection process can be achieved using existing orthogonal projection processing, without changing the internal calculation rules of orthogonal projection.

[0095] The projected length from the first reference mark position to the projection point along the positive direction of the limiting axis is determined as a positive value, and the projected length along the opposite direction of the limiting axis is determined as a negative value, thus obtaining the directional projection length.

[0096] Calculate the ratio of the directional projection length to the distance between the two reference marker positions in the corresponding frame image, and retain the calculation result according to a predetermined number of bits to obtain the corresponding initial relative limit coefficient or current relative limit coefficient.

[0097] The number of preset bits is determined based on the image resolution of the VR acquisition device, the accuracy of the target child's limb reference point recognition, and the preset offset threshold, ensuring that the numerical retention accuracy can distinguish the minimum offset that needs to be recognized.

[0098] Furthermore, this implementation is used to convert the limit offset processing result into unified data that can be directly parsed by VR display devices.

[0099] The limit adjustment prompt data includes the adjustment status field, adjustment direction field, and adjustment distance field.

[0100] When the adjustment distance is zero, the adjustment status field is set to the no-adjustment status indicator, the adjustment direction field is set to the preset no-adjustment direction indicator, and the adjustment distance field is set to zero.

[0101] When the adjustment distance is not zero, the adjustment status field is set to the status identifier to be adjusted, the adjustment direction determined by the sign of the limit offset coefficient is written into the adjustment direction field, and the adjustment distance is written into the adjustment distance field.

[0102] The adjustment status field, adjustment direction field, and adjustment distance field are encapsulated according to the preset data format to obtain limit adjustment prompt data, which is then sent to the VR display device for output.

[0103] The preset data format can adopt a data packet structure with a fixed field order, enabling the VR display device to select the display state based on the adjustment state field, determine the display direction based on the adjustment direction field, and determine the prompt length or display the corresponding value based on the adjustment distance field.

[0104] Figure 3 This is a schematic diagram of the structure of a child medical restraint system based on intelligent sensing provided in an embodiment of this application. The child medical restraint system based on intelligent sensing includes: a VR acquisition module: used to acquire and identify the target child limb reference point, the position of the first reference marker representing the start end of the effective restraint segment, and the position of the second reference marker representing the end end of the effective restraint segment in the initial frame image and the current frame image, respectively, through a VR acquisition device; a restraint coefficient processing module: used to project the target child limb reference point onto the corresponding restraint axis in the initial frame image and the current frame image, respectively, with the direction from the first reference marker position to the second reference marker position as the restraint axis direction, and calculate the ratio of the directed projection length from the first reference marker position to the projection point to the distance between the two reference marker positions, to obtain the initial relative restraint coefficient and the current relative restraint coefficient; and an offset coefficient. Processing module: used to calculate the difference between the current relative limit coefficient and the initial relative limit coefficient to obtain the limit offset coefficient; Limit component first processing module: used to record the adjustment distance of the limit component as zero if the absolute value of the limit offset coefficient is less than the preset offset threshold; Limit component second processing module: used to multiply the absolute value of the limit offset coefficient with the distance between the two reference marks in the current frame image to obtain the adjustment distance of the limit component if the absolute value of the limit offset coefficient is not less than the preset offset threshold, and to record the adjustment direction as from the first reference mark position to the second reference mark position when the limit offset coefficient is positive, and from the second reference mark position to the first reference mark position when the limit offset coefficient is negative; Transmitting and receiving module: used to generate limit adjustment prompt data according to the adjustment direction and adjustment distance and send it to the VR display device.

[0105] This application also provides a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements a child medical limiting operation method based on intelligent sensing.

[0106] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0108] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0109] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0110] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0111] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A child medical restraint operation method based on intelligent sensing, characterized in that, Includes the following steps: The VR acquisition device was used to acquire and identify the target child's limb reference points, the first reference marker position representing the start of the effective restraint segment, and the second reference marker position representing the end of the effective restraint segment in the initial frame image and the current frame image, respectively. In the initial frame image and the current frame image, the direction from the first reference mark position to the second reference mark position is taken as the direction of the limiting axis. The reference point of the target child's limb is projected onto the corresponding limiting axis. The ratio of the directed projection length from the first reference mark position to the projection point to the distance between the two reference mark positions is calculated to obtain the initial relative limiting coefficient and the current relative limiting coefficient. Calculate the difference between the current relative limit coefficient and the initial relative limit coefficient to obtain the limit offset coefficient; If the absolute value of the limit offset coefficient is less than the preset offset threshold, the adjustment distance of the limit component is recorded as zero. If the absolute value of the limit offset coefficient is not less than the preset offset threshold, the absolute value of the limit offset coefficient is multiplied by the distance between the two reference marks in the current frame image to obtain the adjustment distance of the limit component. When the limit offset coefficient is positive, the adjustment direction is recorded as from the position of the first reference mark to the position of the second reference mark. When the limit offset coefficient is negative, the adjustment direction is recorded as from the position of the second reference mark to the position of the first reference mark. Based on the adjustment direction and distance, limit adjustment prompt data is generated and sent to the VR display device.

2. The child medical restraint operation method based on intelligent sensing according to claim 1, characterized in that, Within the preset benchmark confirmation window, the current frame image is continuously acquired according to the preset sampling period, and multiple current relative limit coefficients are obtained sequentially based on each current frame image; The difference between two adjacent current relative limit coefficients is calculated according to the sampling order to obtain the baseline change sequence; If the absolute values ​​of a preset number of consecutive differences in the benchmark change sequence decrease sequentially, and the absolute value of the last difference is less than a preset stable threshold, then it is determined that multiple current relative limit coefficients meet the first preset benchmark change condition, and the initial relative limit coefficient is replaced by the corresponding last current relative limit coefficient, and the subsequent limit offset coefficient is calculated based on the replaced initial relative limit coefficient. If the signs of a preset number of consecutive differences in the benchmark change sequence are the same, and the sum of the absolute values ​​of the consecutive differences is not less than the preset sliding threshold, then it is determined that multiple current relative limit coefficients meet the second preset benchmark change condition, the initial relative limit coefficient remains unchanged, and the limit offset coefficient is updated according to the difference between the last current relative limit coefficient and the initial relative limit coefficient.

3. The child medical restraint operation method based on intelligent sensing according to claim 2, characterized in that, After updating the initial relative limit coefficient according to the first preset benchmark change condition, when multiple limit offset coefficients obtained within the preset action determination window meet the preset action change condition, the limit offset coefficients used to determine the adjustment distance or the current preset offset threshold are processed, including: Obtain multiple limit offset coefficients formed in the sampling order within the preset action determination window, and calculate the difference between two adjacent limit offset coefficients to obtain the offset increment sequence; If the offset increment sequence includes both positive and negative values, and the absolute value of the limit offset coefficient at the end of the preset action determination window is less than the preset regression threshold, then multiple limit offset coefficients are determined to meet the first preset action change condition, and the limit offset coefficient used to determine the adjustment distance is recorded as zero. If the signs of a preset number of consecutive offset increments in the offset increment sequence are the same, and the sum of the absolute values ​​of the consecutive offset increments is not less than the preset cumulative offset threshold, then it is determined that multiple limit offset coefficients meet the second preset action change condition. The preset cumulative judgment threshold that is less than the preset offset threshold is selected as the current preset offset threshold, and the adjustment distance and adjustment direction are determined according to the limit offset coefficient at the end of the preset action judgment window.

4. The child medical restraint operation method based on intelligent sensing according to claim 3, characterized in that, When the absolute value of the limit offset coefficient at the end of the preset action judgment window is not less than the current preset offset threshold, and one adjustment of the limit component has been completed according to the corresponding non-zero adjustment distance, the subsequent adjustment distance is corrected, including: Receive adjustment completion confirmation data returned by the VR display device, and in response to the adjustment completion confirmation data, obtain the adjusted current frame image; The adjusted limit offset coefficient is obtained based on the adjusted current frame image; Calculate the difference between the absolute value of the limit offset coefficient before adjustment and the absolute value of the limit offset coefficient after adjustment, and multiply the difference by the distance between the two reference marker positions in the current frame image before adjustment to obtain the actual recovery distance; Calculate the ratio of the actual recovery distance to the adjusted distance to obtain the adjustment response coefficient; If the adjustment response coefficient is greater than zero and is outside the preset response range, the adjustment distance obtained based on the adjusted limit offset coefficient will be divided by the adjustment response coefficient to obtain the corrected adjustment distance, and the next limit adjustment prompt data will be updated based on the corrected adjustment distance. If the adjustment response coefficient is not greater than zero, a pause adjustment prompt is generated and sent to the VR display device.

5. The child medical restraint operation method based on intelligent sensing according to claim 1, characterized in that, The acquisition and identification of the target child's limb reference points, the position of the first reference marker, and the position of the second reference marker specifically include: Identify the different preset graphic codes set at the start and end of the effective limit segment in the initial frame image and the current frame image, respectively, and record the center position of each preset graphic code as the first reference mark position and the second reference mark position, respectively. Human key point recognition is performed on the initial frame image and the current frame image. Key point positions that match the preset limb part type are selected from the human key point recognition results and used as reference points for the target child's limbs. The recognition result corresponding to the initial frame image is used to calculate the initial relative limit coefficient, and the recognition result corresponding to the current frame image is used to calculate the current relative limit coefficient.

6. The child medical restraint operation method based on intelligent sensing according to claim 1, characterized in that, The process of obtaining the initial relative limit coefficient and the current relative limit coefficient includes: In the image coordinate systems corresponding to the initial frame image and the current frame image, determine the limiting axis through the positions of the first reference mark and the second reference mark; Using the direction from the first reference mark position to the second reference mark position as the positive direction of the limiting axis, the reference point of the target child's limb is projected onto the limiting axis along a direction perpendicular to the limiting axis to obtain the projection point; The projection length from the first reference mark position to the projection point along the positive direction of the limiting axis is recorded as a positive value, and the projection length along the opposite direction of the limiting axis is recorded as a negative value, thus obtaining the directional projection length; Calculate the ratio of the directional projection length to the distance between the two reference marker positions in the corresponding frame image, and retain the calculation result according to a predetermined number of bits to obtain the corresponding initial relative limit coefficient or current relative limit coefficient.

7. The child medical restraint operation method based on intelligent sensing according to claim 1, characterized in that, When the absolute value of the limit offset coefficient is not less than the preset offset threshold, the dimensionless limit offset coefficient is converted into an adjustment distance with the same length unit as the distance by using the distance between the two reference marker positions in the current frame image. After adjusting the distance and retaining it according to the predetermined number of digits, it is used to generate limit adjustment prompt data.

8. The child medical restraint operation method based on intelligent sensing according to claim 1, characterized in that, The limit adjustment prompt data includes the adjustment status field, adjustment direction field, and adjustment distance field; When the adjustment distance is zero, set the adjustment status field to the no-adjustment status indicator, set the adjustment direction field to the preset no-adjustment direction indicator, and set the adjustment distance field to zero; When the adjustment distance is not zero, the adjustment status field is set to the status to be adjusted, the adjustment direction determined by the sign of the limit offset coefficient is written into the adjustment direction field, and the adjustment distance is written into the adjustment distance field. The adjustment status field, adjustment direction field, and adjustment distance field are encapsulated according to the preset data format to obtain limit adjustment prompt data, which is then sent to the VR display device for output.

9. A child medical positioning and restraint system based on intelligent sensing, characterized in that, include: VR Acquisition Module: Used to acquire and identify the target child's limb reference points, the position of the first reference marker representing the start of the effective restraint segment, and the position of the second reference marker representing the end of the effective restraint segment in the initial frame image and the current frame image, respectively, through the VR acquisition device; Limiting coefficient processing module: In the initial frame image and the current frame image, the direction from the first reference mark position to the second reference mark position is used as the limiting axis direction. The target child limb reference point is projected onto the corresponding limiting axis, and the ratio of the directed projection length from the first reference mark position to the projection point to the distance between the two reference mark positions is calculated to obtain the initial relative limiting coefficient and the current relative limiting coefficient. Offset coefficient processing module: used to calculate the difference between the current relative limit coefficient and the initial relative limit coefficient to obtain the limit offset coefficient; The first processing module of the limit component is used to record the adjustment distance of the limit component as zero if the absolute value of the limit offset coefficient is less than the preset offset threshold. The second processing module of the limiting component is used to multiply the absolute value of the limiting offset coefficient with the distance between the two reference marks in the current frame image if the absolute value of the limiting offset coefficient is not less than the preset offset threshold, so as to obtain the adjustment distance of the limiting component. When the limiting offset coefficient is positive, the adjustment direction is recorded as from the position of the first reference mark to the position of the second reference mark, and when the limiting offset coefficient is negative, the adjustment direction is recorded as from the position of the second reference mark to the position of the first reference mark. Transmit / receive module: Used to generate limit adjustment prompt data based on the adjustment direction and distance and send it to the VR display device.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

Images