Emotion sensing decompression method based on acceleration sensing recognition and toy thereof

Abstract

The application discloses an emotion sensing decompression method based on acceleration sensing recognition and a toy thereof, and belongs to the technical field of decompression toys. The method comprises the following steps: collecting acceleration signals received by a flexible main body; identifying a free-fall section, a collision section and a rebound section in sequence according to the acceleration signals; determining a motion mode based on the three-section time sequence combination; when the free-fall section exists, the collision section peak value is greater than a second threshold value, and a reverse acceleration peak value is detected, the motion mode is determined as a flailing motion; when the free-fall section does not exist, the motion mode is determined as a patting motion or a tapping motion according to the size relationship between the peak value and a third threshold value; determining a corresponding emotion level according to the motion mode, and outputting corresponding soothing voice; when the motion mode is identified, continuous frame de-bouncing confirmation and static counting are performed. The application can accurately distinguish flailing, patting and tapping motions, and output corresponding soothing voice, thereby improving the interactive feedback effect.

Description

Technical Field

[0001] This invention relates to the field of stress-relief toy technology, and in particular to an emotion-sensing stress-relief method based on acceleration sensing and the toy thereof. Background Technology

[0002] Emotional release and stress relief toys allow users to release negative emotions by patting or throwing them. Most existing stress relief toys rely solely on physical deformation to relieve stress, lacking intelligent emotion recognition and voice feedback; a few smart toys with built-in sensors simply judge the degree of force based on the peak magnitude of the impact acceleration.

[0003] Relying solely on peak impact values ​​has significant drawbacks: a user forcefully slapping a toy can generate a large peak acceleration; similarly, a toy falling and hitting the ground can also produce a high peak acceleration. These two completely different interactive actions cannot be effectively distinguished by peak values ​​alone, making it easy for the device to misjudge; consequently, the voice feedback may not match the user's actual actions and emotional state, resulting in poor intelligent feedback. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an emotion-sensing decompression method and its toy based on acceleration sensor recognition, which can identify the three-segment waveform timing combination of the throwing action through the acceleration sensor, accurately distinguish between throwing, slapping and light patting, and output corresponding soothing voice accordingly.

[0005] To address the aforementioned technical problems, this invention provides an emotion-based decompression method based on acceleration sensing recognition, comprising the following steps: S1. Collect the acceleration signal of the flexible body; S2. Identify the free fall segment, impact segment, and rebound segment that occur sequentially based on the acceleration signal; the free fall segment is the period when the acceleration value is less than a first threshold and the duration is greater than a first duration threshold; the impact segment is the period after the free fall segment where the peak acceleration value is greater than a second threshold; the rebound segment is the period after the impact segment where the acceleration sensor detects a reverse acceleration peak value; S3. Action mode determination based on three-segment timing combination; S4. Determine the emotion level based on the action pattern; S5. Output the corresponding soothing voice message based on the emotion level; S6. Perform continuous frame de-jitter confirmation and static counting.

[0006] Furthermore, the action pattern determined in S3 includes: When the free fall segment exists, and the peak acceleration of the impact segment is greater than the second threshold, and a reverse acceleration peak is detected, it is determined to be a slamming action; If the free fall segment is not detected, and the peak acceleration is greater than the third threshold and less than or equal to the second threshold, it is determined to be a slapping action; if the peak acceleration is less than or equal to the third threshold, it is determined to be a light patting action.

[0007] Furthermore, the anti-shake confirmation in S6 includes: when the acceleration sampling of a consecutive preset number of frames meets the determination condition of the same action mode, confirming that the action mode is established, so as to filter out interference signals caused by single vibration or accidental touch.

[0008] Furthermore, the static counting in S6 includes: after an action is confirmed, if no new impact segment is detected within a preset static time, the action is determined to be over and the action count is incremented by one.

[0009] Accordingly, the present invention also provides an emotion-sensing decompression toy based on acceleration sensing recognition, including a flexible body, an acceleration sensor, a controller, a voice output module, and a power supply; the acceleration sensor is disposed inside the flexible body; the controller receives the signal from the acceleration sensor, identifies the timing characteristics of the free fall segment, the impact segment, and the rebound segment to distinguish between three types of actions: falling, patting, and light patting; then, based on the action pattern, it obtains the emotion level and drives the voice output module to play soothing voice; the controller also has anti-shake confirmation logic and static counting logic to eliminate vibration misjudgment.

[0010] The beneficial effects of this invention are as follows: This invention no longer relies solely on the magnitude of the acceleration peak to determine the action, but introduces a three-segment time sequence waveform of free fall-impact-rebound as the basis for recognizing the falling action, which can effectively distinguish between two types of impact: falling off the hand and hitting with force. Furthermore, continuous frame anti-shake confirmation and static counting are used to reduce the probability of misrecognition caused by environmental vibrations. Soothing voice feedback of corresponding levels is output for different actions, which improves the human-computer interaction effect of the intelligent cathartic toy. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the overall modules of the present invention; Figure 2 This is a circuit connection block diagram of the present invention; Figure 3 This is a schematic diagram of the process by which the controller of the present invention recognizes the falling action; Figure 4 This is a schematic diagram of the timing of the three waveforms of the present invention. Detailed Implementation

[0012] The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0013] This embodiment discloses an emotion-sensing decompression toy based on acceleration sensing recognition. The flexible body is made of silicone and a three-axis accelerometer is installed inside. The controller is a microcontroller. The voice output module is a speaker. The power supply is a rechargeable lithium battery.

[0014] After the controller determines whether a free fall segment exists: If a freefall segment exists, the process proceeds to the slamming action judgment branch: further determine whether the peak acceleration is greater than the second threshold (2G). If it is, further determine whether a reverse acceleration peak is detected. If detected, it is judged as a slamming action. If there is no free fall segment, then proceed to the peak threshold judgment branch: determine whether the peak acceleration is greater than the third threshold (1G) and less than or equal to the second threshold (2G). If so, it is judged as a slapping action; if the peak acceleration is less than or equal to the third threshold (1G), it is judged as a light patting action.

[0015] Regarding the specific implementation of anti-shake and static counting: When the controller confirms an action is valid, it performs continuous frame de-jitter verification (e.g., five consecutive frames of sampling meet the same action mode's criteria) to filter out transient impacts or noise. After an action confirmation, the controller enters a static waiting state. If no new impact segment is detected within a preset waiting time (e.g., 1 second), the interaction is considered complete, the action count is incremented, and the controller prepares to respond to the next action. If a new valid impact segment appears during the waiting period, the timer is reset or the judgment is recalculated.

[0016] The controller has a pre-stored emotion level mapping table: slamming action corresponds to a high emotion level; patting action corresponds to a medium emotion level; light patting action corresponds to a low emotion level; different levels call up the corresponding voice files; the voice content can include animal sounds plus soothing statements.

[0017] Furthermore, the controller also implements progressive feedback logic: after recognizing the throwing action multiple times in a row, the order of the soothing voice changes; when the cumulative number of throwing actions reaches a preset number, it switches to a deep soothing voice to adapt to the user's continuous venting scenario.

[0018] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent modifications or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An emotion-based decompression method based on acceleration sensing recognition, characterized in that, Includes the following steps: S1. Collect the acceleration signal experienced by the flexible body; S2. Identify the free fall segment, impact segment, and rebound segment that occur sequentially based on the acceleration signal; the free fall segment is the period when the acceleration value is less than a first threshold and the duration is greater than a first duration threshold; the impact segment is the period after the free fall segment where the peak acceleration value is greater than a second threshold; the rebound segment is the period after the impact segment where the acceleration sensor detects a reverse acceleration peak value; S3. Determine the action mode based on the temporal combination of the free fall segment, the impact segment, and the rebound segment; when the free fall segment exists, and the acceleration peak of the impact segment is greater than the second threshold, and a reverse acceleration peak is detected, it is determined to be a slamming action; when the free fall segment is not identified, if the acceleration peak is greater than the third threshold and less than or equal to the second threshold, it is determined to be a slapping action; if the acceleration peak is less than or equal to the third threshold, it is determined to be a light patting action. S4. Determine the corresponding emotion level based on the described action pattern; S5. Output the corresponding soothing voice according to the emotional level; S6. When identifying the action mode, the anti-shake confirmation is performed based on the sampling results of multiple consecutive frames, and the static counting is entered after one action confirmation. If no new impact segment is detected within the preset static time, the action is determined to be over and the action count is incremented by one.

2. The method according to claim 1, characterized in that, The method described in S4 for determining the corresponding emotion level based on the action pattern includes: using a pre-stored emotion level mapping table, assigning a slapping action to a high emotion level, a patting action to a medium emotion level, and a light patting action to a low emotion level.

3. The method according to claim 1, characterized in that, The soothing voice in S5 includes preset animal sounds and soothing statements output sequentially, and the content of the animal sounds and / or the soothing statements is different for different emotion levels.

4. The method according to claim 1, characterized in that, It also includes a progressive feedback step: when the number of consecutive slamming actions is detected to increase, different soothing voices are output in a preset order; when the number of consecutive slamming actions is detected to reach a preset number, a soothing voice different from that of the first slamming action is output.

5. The method according to claim 1, characterized in that, The anti-shake confirmation mentioned in S6 includes: when the acceleration sampling of a consecutive preset number of frames meets the determination condition of the same action mode, the action mode is confirmed to be established, so as to filter out interference signals caused by single vibration or accidental touch.

6. The method according to claim 1, characterized in that, The first threshold and the first duration threshold are used to distinguish between free fall and everyday slight shaking; the second threshold is used to distinguish between a fall and a normal pat; and the third threshold is used to distinguish between a pat and a light tap.

7. An emotion-sensing stress-relieving toy based on acceleration sensing recognition, characterized in that, include: A flexible body; an acceleration sensor, disposed within the flexible body, for detecting the acceleration experienced by the flexible body and outputting an acceleration signal; A controller, connected to the accelerometer, is used to identify the freefall segment, impact segment, and rebound segment based on the acceleration signal, and to determine the action pattern based on the combination of the three time segments. When the freefall segment exists, and the peak acceleration of the impact segment is greater than a second threshold, and a reverse acceleration peak is detected, it is determined to be a slapping action. When the freefall segment is not identified, it is determined to be a patting action or a light patting action based on the relationship between the peak value and a third threshold. The controller is also used to determine the corresponding emotion level based on the action pattern. A voice output module, connected to the controller, is used to output a corresponding soothing voice based on the emotion level. A power supply is used to power the accelerometer, the controller, and the voice output module. The controller is further configured to: when recognizing an action mode, perform anti-shake confirmation based on the sampling results of multiple consecutive frames, and enter a static counting after an action confirmation; if no new impact segment is detected within a preset static time, the action is determined to be over and the action count is incremented by one.

8. The toy according to claim 7, characterized in that, The controller has a pre-stored emotion level mapping table, which assigns a high emotion level to a slapping action, a medium emotion level to a patting action, and a low emotion level to a light patting action.

9. The toy according to claim 7, characterized in that, The controller is also configured to: when the number of consecutively detected falling actions increases, control the voice output module to progressively output different soothing voices in a preset order; when the number of consecutively detected falling actions reaches a preset number, output a soothing voice different from the first falling action.

10. The toy according to claim 7, characterized in that, The accelerometer is a triaxial accelerometer; the flexible body is made of flexible fabric or flexible silicone; the power source is a button battery or a rechargeable lithium battery.

Images