Lamp strip module atmosphere creating method based on scene induction control

Abstract

The invention discloses a lamp strip module atmosphere creating method based on scene induction control, and relates to the technical field of intelligent illumination control, and the method comprises the following steps: S1, collecting an original audio signal of a target space, extracting amplitude fluctuation, a spectrum structure and energy mutation features, recording the starting time and duration of a sound event, and constructing a sound source time sequence; according to the invention, through sound source time sequence modeling, abnormal behavior identification and environment trend analysis, in combination with a multi-dimensional consistency judgment mechanism, accurate response to an abnormal sound source is realized, safe transition is carried out through neutral color temperature lighting effect, and then light effect smooth regression is realized through dynamic recovery evaluation and a slow changing algorithm, so that the accuracy of the abnormal sound source is improved. An intelligent light control system with real-time perception, context understanding and closed-loop regulation and control capabilities is constructed, immersion experience and safety are improved, and the intelligent light control system has remarkable practical value and technical innovation.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and in particular to a method for creating an atmosphere of a light strip module based on scene sensing control. Background Art

[0002] "Creating an atmosphere with light strip modules based on scene-sensing control" means automatically identifying the type of light environment currently required by sensing the specific scene information of the current environment or user behavior (such as light intensity, sound frequency, human movements, time period or spatial position, etc.), and intelligently adjusting parameters such as the brightness, color temperature, and color change rhythm of the light strip module to achieve an atmosphere lighting effect that matches the scene. For example, when the system detects that the user enters the bedroom and sits on the edge of the bed, the light strip will automatically switch to soft warm tones to create a relaxing atmosphere; when it recognizes that the speaker is playing music with a strong sense of rhythm, the light strip will flash and change color synchronously to enhance the entertainment atmosphere. This method combines multi-dimensional sensing technology with programmable light strips to achieve a high degree of linkage between light effects and user situations, providing a personalized and dynamic lighting atmosphere experience for homes, commercial spaces, or in-vehicle environments.

[0003] The existing technology has the following deficiencies: In the process of creating an atmosphere with light strips based on scene-sensing control, the current scene is usually determined by analyzing the rhythmic patterns, volume intensity changes, and frequency distribution characteristics of the ambient sound, and the lighting effects are driven accordingly. However, when abnormal noise events such as screaming, violent collisions, or heavy objects falling to the ground suddenly occur in the actual environment, if the system fails to perceive the sudden changes in the properties of the sound source and its discontinuous evolution trend in the temporal environment in real time, and still relies on static audio parameters to execute a single rule response, it may misjudge the sudden acoustic events with warning properties as normal entertainment sound sources (such as party music, game sound effects, or crowd cheers), thereby erroneously triggering "active atmosphere" lighting modes such as highlight, rhythm, and color. The root of this problem is that the existing light strip control method lacks an adaptive judgment mechanism for abnormal sound source behavior, and cannot integrate the current sound source status and contextual sound field environment characteristics to dynamically adjust the lighting effect control strategy, resulting in a serious deviation between the ambient lighting effect and the actual semantic scene. On the one hand, this type of response mismatch weakens the scene immersion and user experience of lighting rendering; on the other hand, it can easily mislead occupants in potentially dangerous scenarios, obscure the perceptual clues of real risk events, and increase the safety hazard of environmental interaction systems failing to respond at critical moments.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for creating an atmosphere for a light strip module based on scene sensing control. Through sound source time series modeling, abnormal behavior recognition and environmental trend analysis, combined with a multi-dimensional consistency judgment mechanism, it can achieve accurate response to abnormal sound sources, and safely transition through neutral color temperature light effects. Dynamic recovery evaluation and slow-changing algorithms are then used to achieve smooth regression of lighting effects, thereby constructing an intelligent lighting control system with real-time perception, context understanding and closed-loop control capabilities, improving immersive experience and safety, and having significant practical value and technological innovation to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a method for creating an atmosphere of a light strip module based on scene sensing control, comprising the following steps: S1, collects the original audio signal of the target space, extracts the amplitude fluctuation, spectrum structure and energy mutation characteristics, records the start time and duration of the sound event, and constructs the sound source time series; S2, analyze the continuity, change slope and amplitude difference of the sound source time series, identify non-periodic mutation behavior, and determine whether it is an abnormal sound source event; S3, after identifying the abnormal sound source event, extract the environmental background characteristics within a fixed time window before and after the abnormal event, including the duration of silence, sound pressure stability and spatial sound source distribution, and establish an environmental background evolution trend model; S4: Calculate the abnormal behavior consistency index based on the evolution trend model. If the preset conditions are met, it is determined to be a sudden non-entertainment event, suspend the current light strip control logic, and execute the neutral color temperature lighting effect solution; S5, during the execution of the neutral color temperature lighting effect scheme, continuously collects sound source evolution data, analyzes amplitude trends, spectrum changes and sound background stability in multiple time windows, and updates the sound source state recovery level; S6: When the recovery level reaches the set threshold, a smooth transition strategy is executed to gradually switch the lighting effect from the neutral color temperature state to the normal atmosphere mode, completing the closed-loop control.

[0007] Preferably, step S1 includes: Collect the original audio signal in the target space and perform frame segmentation and windowing processing; Extract the amplitude fluctuation, spectral structure and energy mutation characteristics of each frame of audio signal to construct composite sound source parameters; Extract Mel-frequency cepstral coefficients, spectrum centroid, spectrum bandwidth and spectrum roll-off point based on spectrum structure features; The start time and duration of the sound event are marked, and combined with the composite sound source parameters to form a sound source time series for subsequent evolution trend analysis.

[0008] Preferably, step S2 includes: The sound source time series is segmented according to equally spaced time windows to extract the amplitude average, inter-frame energy change rate, spectral centroid and spectral roll-off point fluctuations; Analyze the instantaneous increase and decrease trend of the characteristic curve based on the local change slope index to determine whether there is non-periodic mutation behavior; Perform cluster analysis on the multi-dimensional sound source feature vectors within the mutation segment to identify abnormal signal clusters with consistent features and continuous time. The abnormal signal cluster is compared with the background baseline model constructed in the adjacent time period for feature differences to confirm whether it constitutes an abnormal sound source event.

[0009] Preferably, step S3 includes: The time range of 1 to 3 seconds before and after the abnormal sound source event was selected as the analysis window to extract the silence duration, sound pressure change stability and spatial sound source distribution information; The extracted time series features are normalized and time-aligned with the center point of the abnormal event as the origin to construct the sound field evolution dataset; Regression fitting and time period cluster analysis are performed based on the sound field evolution dataset to identify acoustic context segments; Construct an environmental background evolution trend model and use it to determine the semantic attributes of abnormal events and subsequent lighting effect control strategies.

[0010] Preferably, step S4 includes: The feature vector of the abnormal sound source event and the environmental background trend vector are subjected to multi-dimensional Euclidean distance weighted fusion calculation to obtain the abnormal consistency index; Compare the consistency index with a preset consistency threshold based on historical sound source data statistics. If the threshold is exceeded, it is determined to be a sudden non-entertainment event. Pause the current light strip control logic, execute priority interrupt operation, and freeze the lighting effect output status; Implement a neutral color temperature lighting effect scheme with a color temperature range of 4000K to 5000K, keeping the brightness stable to achieve a safe response.

[0011] Preferably, step S5 includes: Continuously collect ambient audio signals and use a sliding window mechanism to extract amplitude, spectrum, and sound pressure features; Analyze the amplitude change trend and energy fluctuation in each time window to determine whether they are converging; Calculate the changing trend of the spectrum morphology characteristics to determine whether it is transitioning from a disordered to a concentrated state; A multi-dimensional restoration scoring model is constructed by comprehensively considering the changes in various features, and the sound source state restoration level is dynamically updated.

[0012] Preferably, step S6 includes: Freeze the current neutral color temperature lighting effect state and set it as the transition starting reference state; Construct a nonlinear gradient curve for each lighting effect parameter based on the normal atmosphere mode settings; Set independent transition step and time interval for each parameter, and use slow-changing algorithm to achieve continuous adjustment; During the transition process, the sound source state changes are continuously monitored. If an abnormality occurs, the transition is terminated and the neutral color temperature state is restored.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: This invention achieves accurate identification and classification of abnormal sound sources by constructing a sound source time series, identifying non-periodic mutation behavior, extracting environmental background features, and establishing an evolutionary trend model. Combined with a multi-dimensional consistency indicator judgment mechanism, this method introduces a neutral color temperature lighting effect solution as a safety interruption strategy. After the environment stabilizes, it updates the recovery level through dynamic analysis of multiple time windows, and uses a nonlinear slow-varying algorithm to achieve smooth transition regression of the lighting effect, thus constructing an intelligent lighting control system with real-time judgment, contextual awareness, and closed-loop self-adjustment capabilities. The overall solution combines robustness, adaptability, and semantic linkage, enhancing the user's immersive experience in scenarios such as home, business, or in-vehicle use while also improving the environmental interaction system's ability to identify and respond to potential hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 This is a flow chart of a method for creating an atmosphere using a light strip module based on scene sensing control according to the present invention. DETAILED DESCRIPTION

[0016] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0017] The present invention provides Figure 1 The method for creating an atmosphere of a light strip module based on scene sensing control shown includes the following steps: S1, collects the original audio signal in the target space, extracts the composite sound source parameters including amplitude fluctuation, spectral structure and energy mutation characteristics, and records the start time and duration of the sound event to construct the sound source time series; The specific steps for collecting and processing sound source features are as follows: An audio acquisition unit for picking up acoustic information is deployed in the target space to continuously collect the original audio signal in the space. The sampling frequency is preferably set to above 16kHz to ensure that the frequency range of common environmental sound sources is covered. During the audio signal acquisition process, a bandpass filter is introduced to effectively retain the frequency band from 20Hz to 20kHz, and to remove DC components and high-frequency noise interference to ensure the cleanliness of the collected data and the accuracy of subsequent feature extraction. Based on the sampling of the original audio signal, it is short-term processed through frame segmentation and windowing operations. The duration of each frame is set to 20 milliseconds, and the frame shift is set to 10 milliseconds, providing a temporal resolution basis for feature extraction.

[0018] After completing short-term signal preprocessing, the multiple dimensional sound source parameters contained in the audio signal are extracted in sequence. The amplitude fluctuation feature is achieved by calculating the instantaneous energy, short-term average amplitude, and inter-frame change rate of each frame, corresponding to the characterization of sound pressure intensity and transient fluctuation state. The spectral structure feature extracts its frequency domain distribution based on the fast Fourier transform, and further extracts the Mel-frequency cepstral coefficients (MFCC), spectral centroid, spectral bandwidth, and spectral roll-off point to reflect the timbre and structural complexity of the sound source. The energy mutation feature detects energy surges or sharp decreases in a short period of time by introducing a high-order derivative model, thereby identifying possible non-stationary sound source signals, such as abnormal sounds such as explosions and knocks. These characteristic parameters are uniformly organized into a composite sound source feature vector, representing the acoustic state collected at each moment.

[0019] Mel-frequency cepstral coefficients (MFCCs), spectral centroids, spectral bandwidths, and spectral roll-off points are key acoustic parameters used to analyze the spectral characteristics of audio signals. They effectively characterize the timbre, structural complexity, and frequency distribution of a sound source. In the actual extraction process, the original audio signal is first frame-segmented and windowed (e.g., using a Hamming window). A Fast Fourier Transform (FFT) is then performed on each frame to obtain its spectrum. Next, the spectral power map is mapped to a filter bank on the Mel-frequency scale, and the energy corresponding to each filter is calculated. The energy is then logarithmically transformed and then subjected to a discrete cosine transform (DCT) to obtain the MFCC coefficients, which effectively simulate the timbre characteristics perceived by the human ear. The spectral centroid represents the "center of gravity" of the spectral distribution. It is obtained by calculating the frequency-weighted average of the energy and is used to reflect the brightness and clarity of the sound. The spectral bandwidth represents the range of spectral energy distribution around the center and can be used to describe the frequency spread of an audio signal. The spectral roll-off point is the frequency position where the cumulative spectral energy reaches a certain percentage (e.g., 85%) and is often used to determine the proportion of high-frequency components in a sound source. These characteristic parameters have clear physical meanings and distinguishing capabilities in sound source analysis, and can collaboratively describe the spectral structure and complexity of sound events. They are an important basis for determining the sound source category and identifying sudden abnormal sounds.

[0020] After completing the extraction of composite sound source features, in order to realize the evolution analysis of sound source events, each sound signal event needs to be time-labeled, including the start time, end time, and duration. Specifically, by setting the energy threshold and change threshold, it is determined whether a certain signal constitutes an independent sound event, and combining the sliding window mechanism to continuously track its boundary changes to achieve accurate segmentation of events. All identified sound events are numbered in chronological order and associated with their corresponding composite sound source parameters to form a complete event-level audio feature identification dataset.

[0021] All annotated sound events are arranged in chronological order to construct a time series structure for subsequent sound source behavior modeling. This sound source time series not only includes the start time, duration, amplitude characteristics, spectral shape, and energy mutation information of each event, but also embeds dynamic change indicators between the previous and next frames, such as the short-time spectrum crossover rate and the inter-frame energy ratio. This time series structure can comprehensively reflect the evolutionary trends and changing characteristics of the sound source along the timeline, providing a key data foundation for subsequent non-periodic mutation behavior recognition, environmental semantic modeling, and lighting effect control strategies.

[0022] Collecting the raw audio signal within the target space and extracting composite sound source parameters, including amplitude fluctuations, spectral structure, and energy mutation characteristics, while recording the start time and duration of the sound event and constructing a sound source time series, is a fundamental step in the entire adaptive light strip atmosphere control method. Its core role is to achieve the key data conversion from low-level physical sound waves to high-level behavioral semantic modeling. Through this process, continuous environmental sounds can be converted into acoustic data entities with temporal structure and multidimensional feature descriptions, making it possible to subsequently understand the evolution trend of the sound source, mutation behavior, and environmental context. Specifically, amplitude fluctuation characteristics are used to capture the intensity variation patterns of sound, spectral structure characteristics reveal the timbre and complexity of the sound, and energy mutation characteristics can be used to identify non-periodic events such as impacts and explosions. The start and end time and duration information of the event can help construct a complete time series, supporting the modeling of causal and change chains between events. The composite features generated in this process not only have high resolution but also retain the dynamic properties of the time dimension, providing solid data support and behavioral basis for subsequent abnormal sound source identification, environmental background trend modeling, and lighting effect control strategy decision-making, ensuring that the entire control process has response accuracy and intelligent judgment capabilities.

[0023] S2: Analyze the continuity, change slope, and amplitude difference of the sound source time series in the time dimension to determine whether there is non-periodic mutation behavior, and identify abnormal sound source events based on the judgment results; To identify abnormal sound source events, it is necessary to conduct an in-depth analysis of the constructed sound source time series, which includes the following steps: The constructed sound source time series is segmented into equally spaced time windows, each containing a fixed number of frames to maintain consistent temporal granularity. Within each time window, the amplitude average, inter-frame energy change rate, and fluctuations in the spectral centroid and spectral roll-off points of consecutive sound events are calculated to construct a time evolution curve of the sound source characteristics within that time period. To enhance sensitivity to abnormal events, standard deviation normalization is used to unify the scale of various acoustic features, ensuring comparability and stability in subsequent analysis.

[0024] Based on the above-mentioned feature evolution curve, the continuity in the time series is evaluated. Specifically, the local change slope indicator is introduced to characterize the instantaneous increase or decrease trend of the feature on the time axis, and the slope extreme points and directional mutation frequency between three or more consecutive time frames are calculated in combination with the sliding window. If the slope change shows a high-amplitude reversal continuously in a short period of time, accompanied by an abnormal increase in the energy peak, it is preliminarily judged that non-periodic mutation behavior exists in this time period. In addition, by detecting the synergy between features (such as the synchronization of spectral centroid and energy changes), the multi-dimensional consistency of the mutation behavior is further verified to enhance the recognition accuracy.

[0025] The "local change slope index" is a numerical measure used to measure the speed and direction of change of a feature value between consecutive time frames in an audio feature time series. Essentially, it approximates the first-order derivative of the feature curve in the time dimension, capturing the instantaneous fluctuation trends of the sound signal. In sound source analysis, this metric primarily identifies whether audio features exhibit sudden, non-stationary changes, such as sharp increases or decreases in energy or rapid shifts in the spectral center of gravity. These phenomena are often highly correlated with unusual sound source events (such as impacts, screams, and explosions). The method for obtaining this metric involves first constructing a time series of selected acoustic features (such as energy or spectral center of gravity) by time frame. Then, using a sliding window approach, the difference between the numerical changes between two or more adjacent frames is calculated. For example, if each frame is set to 10 milliseconds, the difference between each two frames is divided by the time interval to estimate the local change slope of the feature segment. Furthermore, the slopes can be continuously calculated within multiple windows and their changing trends (such as continuous reversals of slope sign or high-frequency fluctuations) analyzed to determine whether there is unusual fluctuation behavior. The local change slope index can effectively improve the recognition sensitivity of sound source mutation events by strengthening the dynamic analysis capability in the time dimension, and is one of the key parameters for realizing multi-dimensional acoustic intelligent recognition.

[0026] After determining the time period of non-periodic mutation behavior, a local cluster analysis is performed on this mutation segment to determine whether it constitutes an independent abnormal sound event. Cluster analysis constructs a Euclidean distance matrix based on multidimensional feature vectors and combines density thresholds with time length thresholds to identify abnormal behavior clusters. When the identified abnormal behavior clearly deviates from the duration of a regular sound source event in terms of time, and the degree of change in its amplitude and spectrum exceeds the set threshold, it is marked as a suspected abnormal event. This process avoids the misjudgment of occasional non-significant fluctuations and improves the reliability of identifying abnormal sound source events.

[0027] The main purpose of performing local cluster analysis is to further verify whether non-periodic mutation behavior possesses structural independence and characteristic aggregation after detection, thereby identifying whether it constitutes a true abnormal sound event with behavioral boundaries. Non-periodic mutations may simply be brief fluctuations in certain high-frequency background sounds. Directly treating them as abnormal events can easily lead to misjudgment. However, local cluster analysis can aggregate and evaluate the composite sound source features of adjacent frames in a time series to determine whether these frames form a densely distributed area in the feature space. In other words, whether there is a cluster of abnormal signals that are highly similar in dimensions such as amplitude, spectrum, and energy, and continuous in time. The specific steps include: first, selecting the multidimensional sound source feature vectors of all frames within the time period in which the mutation behavior was detected in the previous step; second, calculating the Euclidean distance or cosine similarity between these feature vectors to construct a distance matrix between features; third, performing a clustering operation based on a density threshold (such as the neighborhood radius and minimum number of samples in the DBSCAN algorithm) to identify a group of frames that are highly similar in feature dimensions; finally, determining whether the clustering result has temporal continuity and feature consistency. If it meets the clustering scale and boundary stability requirements, it is confirmed as an independent abnormal sound source event. The core value of this method lies in avoiding the misidentification of occasional unstructured mutations as valid events through multidimensional similarity judgment and density detection, improving the accuracy and robustness of abnormal sound source identification, and is an important step in building an intelligent acoustic understanding mechanism.

[0028] After identifying a suspected abnormal event, a comparison mechanism based on background baseline modeling is required to further confirm its abnormal properties. The baseline model is constructed by selecting the sound source characteristics of the adjacent time periods before and after the event. The baseline model is then compared with the characteristic curve of the current suspected abnormal event. If the amplitude deviation and spectral distribution alienation significantly exceed the background fluctuation range, the event is ultimately confirmed as an abnormal sound source event.

[0029] Analyzing the temporal continuity, slope, and amplitude differences of sound source time series, and identifying anomalous sound source events based on these findings, plays a crucial role in creating an atmosphere using scene-sensing controlled light strips. This step aims to identify non-periodic and abrupt acoustic behaviors within complex ambient sound signals, thereby distinguishing normal background sounds (such as voice communication and musical rhythm) from events with alarming or anomalous properties (such as screaming, falling objects, and shattering glass). By analyzing the temporal continuity of various acoustic features (such as energy, spectrum, and sound pressure) in the sound source time series, we can determine whether the feature values ​​exhibit a stable evolution trend or experience drastic fluctuations between frames. Furthermore, the slope of change metric is introduced to capture the speed and direction of feature changes over short timescales, assisting in identifying abrupt signal changes. Amplitude difference analysis assesses the degree of deviation between local extreme values ​​and the background average, revealing the intensity and significance of anomalous fluctuations. The coordinated analysis of these three elements enables early warning and significance assessment of unexpected events, preventing misidentification of high-intensity, yet semantically sound, entertainment-related sounds as abnormal events, or conversely, missing dangerous signals. This step not only strengthens the semantic understanding of sound source perception but also provides a basis for the subsequent proper switching of lighting control logic. It is a key technical step in ensuring that the output of ambient lighting effects is highly consistent with the semantics of the real scene.

[0030] S3, after identifying the abnormal sound source event, extract the environmental background characteristics within a fixed time window before and after the abnormal sound source event, including the duration of silence, the stability of sound pressure changes, and the spatial sound source distribution information, and establish an environmental background evolution trend model; After successfully identifying an abnormal sound source event, in order to improve the accuracy of the judgment of the event nature and the ability to understand the contextual semantics, it is necessary to further extract the environmental background features before and after the abnormal event and establish an environmental background evolution trend model to assist in the adaptive decision-making of the lighting effect response. This process includes the following steps: A fixed time range before and after the occurrence of the abnormal sound source event is selected as the analysis window, and a time span of 1 to 3 seconds is preferably set on each side to ensure that the contextual acoustic change state of the event is covered. Within the set time window, the continuous audio data is time-series scanned and feature extracted. The extracted content includes three core indicators: silence duration, sound pressure change stability and spatial sound source distribution information. The silence duration is calculated by accumulating the time segments where the energy frame value is lower than the preset threshold to reflect whether the current environment is in a relatively low sound pressure state. The stability of sound pressure change calculates the standard deviation and change rate of each frame energy through a sliding window to determine the degree of fluctuation of the background sound, which is used to distinguish between quiet environments and active scenes. The spatial sound source distribution information is based on multi-point audio acquisition or single-point directional algorithm to estimate the propagation position and distribution density of the sound in space, which is used to identify whether the sound source is quickly cut in from a specific direction, and to assist in determining whether it is an external interference signal.

[0031] After extracting environmental background features, the extracted time series feature values ​​are normalized and time-aligned, unifying them to the center of the abnormal event as the reference origin to construct a symmetrical time axis, forming a bidirectional comparative sound field evolution dataset. This dataset records the changes in silence state before and after the event, energy fluctuation trends, and the spatial displacement path of the sound source, allowing subsequent models to utilize structured data for trend fitting and semantic induction.

[0032] Based on this structured dataset, an environmental background evolution trend model was constructed. This model combines regression fitting with cluster analysis, first fitting a time series curve to each feature type (using methods such as local weighted regression and sliding average). The fitting results are then clustered and identified by time period, dividing the sound field state before and after the event into several acoustic context segments. For example, if the environment is silent for a long time before the event, with extremely low sound pressure and no obvious directional distribution of the sound source, and then the sound pressure suddenly jumps and shows directional concentration after the event, the trend model can identify it as a "silent burst anomaly." This modeling process not only retains the temporal evolution logic but also integrates the synergistic relationship between multidimensional features, making it a key analytical tool for distinguishing different types of sound source events.

[0033] After the model is built, the identified abnormal sound source events can be attributed and contextually analyzed based on the evolutionary pattern characteristics within the environmental background evolution trend model. This helps determine whether the subsequent light strip control strategy will enter a safe response logic or maintain a normal atmosphere. For example, if the model determines that the abnormal event is a high-confidence, sudden, non-entertainment event, it indicates that the event is inconsistent with the trend of changes in the ambient sound field, is intrusive, and abnormal, and the current lighting effect needs to be interrupted and the preset neutral light mode needs to be entered.

[0034] After identifying an abnormal sound source event, extracting environmental background features within a fixed time window before and after the event and building a model of the environmental background evolution trend based on this information is a key step in implementing an intelligent response mechanism for light strip atmosphere control. This provides contextual information to determine the true semantic attributes of the abnormal event. Ambient sound is a continuous and contextual signal. A single sudden change in a sound source often makes it difficult to accurately determine its semantic category, potentially leading to misidentification or omission due to a lack of contextual support. This step allows analysis of whether the abnormal sound source event aligns with the surrounding sound field conditions, effectively distinguishing between sudden, non-entertainment anomalies (such as accidents and danger signals) and loud, acceptable sound fluctuations in a normal context (such as musical climaxes, applause, and children's screams). Specifically, the duration of silence reflects whether the event was preceded by a low sound pressure background. A sudden burst of loud sound after a long period of silence often indicates a sudden event. The stability of sound pressure fluctuations assesses the level of dynamic fluctuations in the ambient sound, determining whether the event is part of a naturally evolving background. Spatial sound source distribution information provides clues as to whether the sound source suddenly enters from a specific direction, helping to identify environmental disturbances or specific behavioral changes in the room. The environmental background evolution trend model constructed based on this data not only enhances the basis for determining the attributes of abnormal events, but also improves the adaptability and semantic consistency of lighting effect response strategies, effectively avoiding the false triggering of inappropriate lighting effects such as high brightness and color, thereby improving user experience and environmental safety. This mechanism provides a non-obvious technological advancement in semantic-level sound source understanding and is a key support for achieving intelligent ambient light linkage.

[0035] S4: Calculate the consistency index of abnormal sound source behavior based on the environmental background evolution trend model. When the consistency index meets the preset conditions, it is determined to be a sudden non-entertainment event, suspend the current light strip control logic, and execute the corresponding neutral color temperature lighting effect solution; To improve the semantic judgment capability of abnormal sound source events and ensure the high consistency between the light effect response strategy and the actual environmental status, after extracting and constructing the environmental background evolution trend model, it is necessary to further calculate the consistency index of abnormal sound source behavior and use it to determine whether to enter the safety response process. The specific implementation process includes the following steps: Based on the established environmental background evolution trend model, the evolutionary trajectories of various acoustic features before and after the abnormal sound source event are extracted, mainly including the changing trend of silence duration, the stability trend of the sound pressure fluctuation range, and the distribution expansion trajectory of the sound source in space. The above characteristic curves are time-aligned, and the characteristic data corresponding to the abnormal sound source event is used as the current state vector. A difference comparison method is introduced between the background trend curve to measure the degree of deviation of the abnormal event from the environmental baseline. The deviation calculation method can adopt the multi-dimensional Euclidean distance weighted fusion method. After normalizing the indicators such as the degree of silence mutation, the energy change slope, and the degree of spatial distribution alienation, a comprehensive evaluation is conducted to form a unified abnormal consistency indicator system.

[0036] The multidimensional Euclidean distance weighted fusion method is a mathematical approach used to comprehensively assess the degree of deviation between multiple feature dimensions. It primarily quantifies differences by calculating the "distance" of each feature relative to a reference state in multidimensional space, and assigns weights based on the importance of each feature to achieve more accurate fusion judgments. In this invention, the method is used to measure the degree of deviation between the current abnormal sound source event and its surrounding environmental background across multiple acoustic dimensions, thereby forming a unified anomaly consistency indicator. Specifically, multiple acoustic features, such as the degree of silence mutation, energy change slope, and degree of spatial distribution alienation, are first normalized to unify their value ranges to the same scale range (e.g., 0 to 1) to eliminate bias caused by differences in dimension and numerical range. Subsequently, these normalized feature vectors are treated as coordinate points in multidimensional space and subjected to Euclidean distance calculation with a reference vector representing the environmental background trend. This involves taking the square root of the sum of the squared differences in each feature dimension to obtain the overall degree of deviation. Furthermore, based on the relative importance of each feature's contribution to anomaly detection in actual experience or training data, different weighting coefficients are assigned to the distance values ​​in each dimension. These are then weighted and summed to obtain the weighted fused Euclidean distance, which serves as the final anomaly consistency indicator. The larger this index, the more significant the overall deviation from the environmental context and the greater the suddenness of the event. This approach fully considers the combined influence of multiple acoustic variation characteristics, improving the sensitivity and robustness of anomaly detection and avoiding misjudgments caused by fluctuations in a single feature. This provides a more scientific and accurate basis for intelligent lighting response logic.

[0037] A preset consistency threshold is set and the calculated consistency index is compared with this threshold. When the consistency index exceeds the threshold, it indicates that the current sound source behavior is seriously inconsistent with the background trend in multiple dimensions, indicating abnormal behavior that is sudden, discontinuous, and significantly intrusive. This type of behavior usually corresponds to accidental, warning, or destructive events in daily life, rather than entertainment activities such as rhythmic music and crowd interaction. Therefore, it should be classified as "sudden non-entertainment events" in the judgment logic. This judgment mechanism avoids the risk of falsely triggering the warning mode for loud but semantically normal events (such as cheers and beats) and is clearly non-obvious.

[0038] The consistency threshold is set primarily based on statistical analysis of historical acoustic data and the required tolerance for false positives in real-world scenarios. By collecting a large number of sound source feature samples under normal conditions and calculating the multidimensional Euclidean distance distribution between them and the corresponding environmental background trends, a standard reference range is obtained. Based on this, a statistically significant percentile value (such as the 95th or 99th percentile) is selected as the threshold. Under normal conditions, the consistency index for the vast majority of sound source behaviors falls below this value. Only in the event of sudden or unusual behavior does the index significantly exceed this threshold, triggering the judgment logic for non-entertainment events. This threshold can also be adjusted based on the application scenario. For example, in a home environment, it can be set to a lower value for increased safety, while in entertainment scenarios, it can be appropriately increased to reduce false triggers.

[0039] After completing the above judgment, to prevent the lighting effect response from mismatching the semantics of the environment, all dynamic control logic of the current light strip is immediately paused. Specifically, the pause operation includes interrupting the currently executing lighting effect animation logic, freezing the light strip's output state, and stopping all lighting change sequences triggered by rhythm or emotion perception, ensuring that the light strip does not display misleading lighting effects during the judgment period. This operation takes priority over the regular lighting effect update process and is a real-time interruption, ensuring the immediacy and safety of the response.

[0040] While the control logic is paused, a neutral color temperature lighting scheme, matching the safe state, is immediately executed. This pre-set static lighting scheme uses white or warm white light with a color temperature between 4000K and 5000K, maintaining a moderately stable brightness level to create a neutral, safe, and non-intrusive optical atmosphere. This scheme minimizes sensory stimulation and is suitable for visual noise reduction and heightened alertness during the event assessment phase, while also avoiding distractions caused by dynamic lighting effects during unusual events.

[0041] The consistency index of abnormal sound source behavior is calculated according to the environmental background evolution trend model, and it is judged whether it is a sudden non-entertainment event based on this, and then the current light strip control logic is suspended and the neutral color temperature lighting effect scheme is executed. This step plays a key role in intelligent judgment and response conversion in the present invention. The core purpose of this step is to correlate the sound source identification results with the acoustic context in which it is located, and to improve the accuracy of understanding the semantics of the sound source by quantifying the degree of match between the current abnormal behavior and the environmental background, so as to avoid the light strip from mistakenly responding to inappropriate scenes. Abnormal sound sources do not always mean danger or interference. They may be part of scenes such as music climaxes and game sound effects. Only when their behavior patterns deviate significantly from the evolution trend of the surrounding sound field in terms of intensity, rate of change, and continuity, do they have suddenness and abnormality. By calculating the consistency index, the current sound source feature vector can be quantitatively compared with the background trend model, and it can be evaluated from multiple dimensions whether it constitutes an "environmental discontinuity event." When the consistency index exceeds a set threshold, indicating that the light strip is intrusive and non-entertainment in the background environment, the original dynamic response logic of the light strip is suspended to avoid the false triggering of "active atmosphere" lighting effects such as high brightness and strobe. The output is replaced with a neutral color temperature lighting effect, providing a stable, soft, and non-intrusive visual state. This not only enhances the safety and accuracy of the lighting effect response, but also effectively reduces the risk of misleading caused by lighting effect mismatch in abnormal scenarios, ensuring that the light strip is highly matched with the actual environment at the semantic level. It is an important component of the closed-loop semantic perception control mechanism implemented by the present invention.

[0042] S5: During the implementation of the neutral color temperature lighting effect scheme, continuously collect sound source evolution data, analyze the amplitude change trend, spectrum reconstruction characteristics and ambient sound background stability in multiple time windows, and update the sound source state recovery level; When implementing a neutral color temperature lighting effect solution to respond to sudden non-entertainment events, in order to determine whether the sound source state has returned to normal environmental levels, it is necessary to continuously collect sound source evolution data and dynamically update the sound source state recovery level based on multi-dimensional acoustic characteristics. This process not only ensures the stability and adaptability of the lighting effect response, but also constitutes the core link of intelligent closed-loop control. Specifically, it includes the following steps: While implementing the neutral color temperature lighting scheme, the ambient audio signal in the target space is continuously collected at a preset frame rate, maintaining a time series record of all key acoustic features. During the acquisition process, a fixed-length time window (e.g., 1 or 2 seconds) is set, and a sliding window mechanism is used to segment the continuous audio stream. Within each window period, characteristic parameters such as the sound source amplitude, energy envelope, and spectral distribution are extracted to ensure high-temporal-resolution data on the dynamic evolution of the sound source. Noise reduction algorithms and energy threshold screening strategies are applied during the sampling process to eliminate redundant background interference and ensure that the collected data truly reflects the changing trends of the sound source.

[0043] After acquiring sound source feature data from multiple windows, a fitting analysis is performed on the amplitude variation trends within each time window. Specifically, this involves calculating the inter-frame amplitude variation rate to detect whether there is continuous upward or downward movement; statistically analyzing the energy standard deviation and peak-to-valley difference to determine whether the fluctuation amplitude is converging; and introducing sliding average and smoothing filtering algorithms to enhance the stability and fault tolerance of trend analysis. When multiple consecutive windows exhibit characteristics such as slowing energy variation and stabilizing amplitude, it can be preliminarily concluded that the sound source state is evolving toward stable recovery.

[0044] The semantic changes of the sound source are further judged based on the spectrum reconstruction characteristics within each time window. The specific method is: perform fast Fourier transform on each frame of audio signal to obtain a spectrum diagram, then calculate the spectrum morphological characteristics such as spectrum centroid, spectrum bandwidth and spectrum roll-off point, and analyze its change trend between multiple time windows. When the spectrum reconstruction shows a gradual recovery from disordered, discrete, concentrated high-frequency bands to low-frequency dominance, spectrum concentration, and symmetrical energy distribution, it means that the sound source characteristics have gradually recovered from a sudden or fragmented state to a stable environment state. In addition, it is necessary to analyze the stability of the background sound pressure, that is, whether the fluctuation range of the sound pressure value in multiple time windows converges to a certain steady-state interval, in order to comprehensively judge the static stability level of the overall environment.

[0045] A multidimensional recovery scoring model is constructed by integrating the amplitude change trends, spectrum reconstruction status, and background sound pressure stability in each time window, and the sound source state recovery level is dynamically updated accordingly. The scoring model can use a weighted accumulation mechanism to calculate the degree of proximity of the current sound source state to the set "normal environment template" based on the degree of deviation, change trend, and duration of each feature. When the cumulative recovery score reaches the preset threshold, the current sound source state is determined to have returned to normal levels, and in subsequent steps, the light strip lighting effect transition strategy from the neutral color temperature state to the normal atmosphere mode is triggered.

[0046] During the implementation of the neutral color temperature lighting scheme, continuously collecting sound source evolution data, analyzing amplitude trends, spectral reconstruction characteristics, and ambient sound background stability over multiple time windows, and updating the sound source recovery level accordingly are key steps in achieving a closed-loop dynamic adaptive control logic. Its core function is to determine, through continuous and structured acoustic monitoring, whether the current abnormal sound source has subsided and whether the environment has gradually returned to stability, thereby providing reliable evidence for the smooth restoration of the subsequent lighting mode. In real-world scenarios, sudden non-entertainment events often exhibit short-term, high-energy, irregular, and spectrally complex sound source characteristics. Therefore, a single recognition result alone cannot accurately determine whether the event has ended or whether potential interference still exists. However, by continuously collecting new sound source data during the neutral color temperature response phase and performing segmented processing and trend analysis using a sliding window approach, the evolution of acoustic characteristics can be continuously observed. If amplitude changes gradually stabilize, spectral morphology regains regularity, and background sound pressure fluctuations decrease, it indicates that the sound source is gradually returning to normal. Quantifying this dynamic process into recovery levels and establishing a multi-level scoring standard not only prevents false triggering of recovery operations due to brief periods of silence, but also enables a smooth transition from alert mode to normal ambient lighting through refined control strategies. This step effectively enhances the contextual adaptability and semantic judgment capabilities of lighting control, improving the system's stability and intelligence in complex environments.

[0047] S6: When the sound source state recovery level reaches the set threshold, a smooth transition strategy is executed to gradually switch the current lighting effect from the neutral color temperature state to the normal atmosphere mode, completing the adaptive closed-loop control of the light strip atmosphere effect; To achieve a smooth transition from the neutral color temperature lighting effect state to the normal atmosphere mode and ensure that the lighting effect output always maintains semantic consistency with the current environmental state, a smooth transition strategy with phased and dynamic adjustment capabilities should be implemented after detecting that the sound source state recovery level reaches the preset threshold. This process ensures that the lighting effect returns to a natural and coherent process after the abnormal state is resolved, avoiding visual abrupt changes or uncontrolled responses. The specific implementation includes the following steps: Based on the sound source state recovery level value dynamically updated in the previous step, a real-time comparison is performed with the set recovery threshold. When the recovery level in multiple consecutive time windows is higher than the threshold, and the acoustic feature fluctuation range is stable within the predefined low-variance range, it can be determined that the current environmental state has returned from the abnormal interference state to the normal semantic range. At this time, the start signal of the smooth transition strategy is triggered, and the current neutral color temperature lighting effect state is frozen as the transition starting reference state, ensuring a stable baseline for the entire dimming process.

[0048] A multi-dimensional gradient curve is constructed based on the transition target—the difference between the lighting parameters required for normal ambient mode (including color temperature, brightness, color saturation, and frequency of change) and the current neutral color temperature state. Each parameter dimension is assigned a separate transition step size and time interval, and a slow-changing algorithm (such as sinusoidal interpolation, exponential decay, or piecewise linear transformation) is employed to achieve nonlinear, gradual adjustment. In particular, color temperature transitions should avoid sudden changes from cold to warm, and instead use color temperature gradient interpolation to create a gentle color temperature shift. For brightness adjustment, initial brightness fluctuations should be kept within 10% to prevent visual abruptness. Color and rhythm transitions should be delayed and synchronized with the dynamic rhythm of the lighting effects.

[0049] "A separate transition step size and time interval are set for each parameter dimension, and a slow-changing algorithm (such as sinusoidal interpolation, exponential decay, or piecewise linear transformation) is used to achieve nonlinear, gradual adjustment." This means that in the process of smoothly transitioning the lighting effect from a neutral color temperature state to a regular atmosphere mode, the change speed (step size) and update frequency (time interval) of each lighting control parameter (such as color temperature, brightness, saturation, color change rhythm, etc.) are set separately, and mathematical functions are used for slow and continuous adjustment, so that the overall lighting effect presents a soft, non-abrupt transition effect visually. The implementation method involves first determining the initial value (current neutral color temperature state) and target value (normal atmosphere mode setting) for each parameter. Then, based on the target change characteristics, an appropriate slow-change algorithm is selected—for example, sinusoidal interpolation for a natural, gradual transition with fluctuations, exponential decay for a rapid initial change followed by a steady state, and piecewise linear transformation for multi-stage linear changes. Next, a change step size (e.g., 1% brightness change per time) and time interval (e.g., updates every 50 milliseconds) are set for each parameter. During execution, each parameter gradually approaches the target value along its own change curve, ultimately completing the nonlinear gradual change of the overall lighting effect within a few cycles. This method achieves dynamic and flexible lighting effect transitions under scene perception by finely regulating the response rhythm of each parameter dimension, enhancing the natural fit between the ambient atmosphere and human perception.

[0050] During the transition process, the system continuously monitors changes in the state of ambient sound sources to prevent recurrence of abnormal events or failure to promptly identify new abnormal signals. Specifically, an "interruption path" is maintained during the transition cycle. If a sudden change in sound pressure, abnormal spectral reconstruction, or a shift in the concentration of spatial sound sources is detected within any time window, the transition operation is immediately terminated and the system is restored to a neutral color temperature state. The recovery level assessment process is also restarted. This dynamic monitoring mechanism ensures the safety and robustness of the transition operation, preventing inconsistent lighting effects caused by misjudgment of the scene.

[0051] When all lighting parameters successfully complete the transition curve and reach the normal ambient mode settings, and the ambient state remains stable throughout the transition, the transition is marked complete and the current light strip status is updated to "Normal Mode." This completes the entire adaptive closed-loop control process: from abnormal event recognition, neutral response, recovery judgment, smooth transition, and return to normal mode, forming a complete, continuous, self-aware and controllable intelligent light strip response mechanism.

[0052] When the sound source recovery level reaches a set threshold, a smooth transition strategy is implemented, gradually switching the current lighting effect from a neutral color temperature state to a normal ambiance mode. This step aims to achieve an intelligent and gradual return from an abnormal response state to a normal visual ambiance state, ensuring a dynamic balance between functionality and user experience. Previously, the system entered a neutral color temperature lighting effect state upon detecting abnormal sound source behavior. While this state provides safety and interference suppression, maintaining it for a long time can result in a dull and lacking sense of interaction. Therefore, the system needs to restore the normal ambiance lighting effect after the sound source environment stabilizes. However, a sudden jump from a neutral state to a bright, vibrant, or rhythmic ambiance mode can easily cause visual abruptness or misleading meaning. Therefore, this step employs a transition strategy to gradually transition various lighting control parameters (such as color temperature, brightness, color, and rhythm) over time, creating a smooth and fluid visual transition. For example, by using a slow-changing algorithm such as sinusoidal interpolation or exponential decay, the lighting transition from calm to active is gradually adjusted, giving the user the perception that the environment is gradually "returning to normal," rather than abruptly. At the same time, the transition strategy still retains the ability to monitor the sound source environment. If an abnormality occurs again during the transition, the conversion process will be immediately interrupted to ensure the stability and safety of the response.

[0053] This scene-sensing controlled light strip module atmosphere creation method effectively addresses technical challenges faced by existing light strip control methods, such as response mismatch, semantic judgment deficiencies, and false triggering of lighting effects when faced with unusual sound sources. This method significantly improves the consistency and intelligence between lighting output and the actual environmental context. By constructing sound source time series, identifying non-periodic mutation behavior, extracting environmental background features, and establishing an evolutionary trend model, combined with a multi-dimensional consistency indicator judgment mechanism, this method achieves accurate identification and classification of unusual sound sources. Furthermore, it introduces a neutral color temperature lighting effect scheme as a safety interruption strategy. Once the environment stabilizes, it dynamically updates the recovery level through multi-time window analysis, and achieves smooth transition regression of the lighting effect using a nonlinear, slowly varying algorithm. This results in an intelligent lighting control system with real-time judgment, contextual awareness, and closed-loop self-adjustment capabilities. This overall solution combines robustness, adaptability, and semantic linkage, enhancing the user experience in scenarios such as homes, businesses, and vehicles while also improving the environmental interaction system's ability to identify and respond to potential hazards.

[0054] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0055] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0056] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0057] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0058] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0059] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0060] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0061] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0062] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. A method for creating an atmosphere of a light strip module based on scene sensing control, characterized in that: The following steps are involved: S1, collects the original audio signal of the target space, extracts the amplitude fluctuation, spectrum structure and energy mutation characteristics, records the start time and duration of the sound event, and constructs the sound source time series; S2, analyze the continuity, change slope and amplitude difference of the sound source time series, identify non-periodic mutation behavior, and determine whether it is an abnormal sound source event; S3, after identifying the abnormal sound source event, extract the environmental background characteristics within a fixed time window before and after the abnormal event, including the duration of silence, sound pressure stability and spatial sound source distribution, and establish an environmental background evolution trend model; S4: Calculate the abnormal behavior consistency index based on the evolution trend model. If the preset conditions are met, it is determined to be a sudden non-entertainment event, suspend the current light strip control logic, and execute the neutral color temperature lighting effect solution; S5, during the execution of the neutral color temperature lighting effect scheme, continuously collects sound source evolution data, analyzes amplitude trends, spectrum changes and sound background stability in multiple time windows, and updates the sound source state recovery level; S6: When the recovery level reaches the set threshold, a smooth transition strategy is executed to gradually switch the lighting effect from the neutral color temperature state to the normal atmosphere mode, completing the closed-loop control.

2. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S1 includes: Collect the original audio signal in the target space and perform frame segmentation and windowing processing; Extract the amplitude fluctuation, spectral structure and energy mutation characteristics of each frame of audio signal to construct composite sound source parameters; Extract Mel-frequency cepstral coefficients, spectrum centroid, spectrum bandwidth and spectrum roll-off point based on spectrum structure features; The start time and duration of the sound event are marked, and combined with the composite sound source parameters to form a sound source time series for subsequent evolution trend analysis.

3. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S2 includes: The sound source time series is segmented according to equally spaced time windows to extract the amplitude average, inter-frame energy change rate, spectral centroid and spectral roll-off point fluctuations; Analyze the instantaneous increase and decrease trend of the characteristic curve based on the local change slope index to determine whether there is non-periodic mutation behavior; Perform cluster analysis on the multi-dimensional sound source feature vectors within the mutation segment to identify abnormal signal clusters with consistent features and continuous time. The abnormal signal cluster is compared with the background baseline model constructed in the adjacent time period for feature differences to confirm whether it constitutes an abnormal sound source event.

4. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S3 includes: The time range of 1 to 3 seconds before and after the abnormal sound source event was selected as the analysis window to extract the silence duration, sound pressure change stability and spatial sound source distribution information; The extracted time series features are normalized and time-aligned with the center point of the abnormal event as the origin to construct the sound field evolution dataset; Regression fitting and time period cluster analysis are performed based on the sound field evolution dataset to identify acoustic context segments; Construct an environmental background evolution trend model and use it to determine the semantic attributes of abnormal events and subsequent lighting effect control strategies.

5. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S4 includes: The feature vector of the abnormal sound source event and the environmental background trend vector are subjected to multi-dimensional Euclidean distance weighted fusion calculation to obtain the abnormal consistency index; Compare the consistency index with a preset consistency threshold based on historical sound source data statistics. If the threshold is exceeded, it is determined to be a sudden non-entertainment event. Pause the current light strip control logic, execute priority interrupt operation, and freeze the lighting effect output status; Implement a neutral color temperature lighting effect scheme with a color temperature range of 4000K to 5000K, keeping the brightness stable to achieve a safe response.

6. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S5 includes: Continuously collect ambient audio signals and use a sliding window mechanism to extract amplitude, spectrum, and sound pressure features; Analyze the amplitude change trend and energy fluctuation in each time window to determine whether they are converging; Calculate the changing trend of the spectrum morphology characteristics to determine whether it is transitioning from a disordered to a concentrated state; A multi-dimensional restoration scoring model is constructed by comprehensively considering the changes in various features, and the sound source state restoration level is dynamically updated.

7. The method for creating an atmosphere of a light strip module based on scene sensing control according to claim 1, characterized in that: Step S6 includes: Freeze the current neutral color temperature lighting effect state and set it as the transition starting reference state; Construct a nonlinear gradient curve for each lighting effect parameter based on the normal atmosphere mode settings; Set independent transition step and time interval for each parameter, and use slow-changing algorithm to achieve continuous adjustment; During the transition process, the sound source state changes are continuously monitored. If an abnormality occurs, the transition is terminated and the neutral color temperature state is restored.

Images