Artificial intelligence visual identification and interaction control method

Abstract

The invention relates to the technical field of electric digital data processing, in particular to an artificial intelligence visual identification and interaction control method. Comprising the following steps: S1, acquiring a data acquisition period of a camera and an interval acquisition period of a sensor in a target scene; based on the data acquisition period, acquiring an intermittent work period of visual identification; s2, dividing data characteristics into steady state data and transient state data in the data acquisition period; obtaining change data based on the steady state data and the transient state data; and S3, in the interval acquisition period, adjusting the interval acquisition period of the change data. According to the visual identification method based on data feature classification, data types are dynamically divided, the period is optimized in combination with indexes and logarithmic functions, and acquisition and processing strategies are adjusted in a self-adaptive mode; and resources are dynamically allocated by using a difference quantification formula, energy consumption and precision are optimized, and the reliability and adaptability of complex terrain monitoring under low power consumption are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to an artificial intelligence visual recognition and interactive control method. Background Art

[0002] Visual recognition and interactive control in outdoor environments face severe challenges. Existing visual recognition technologies face problems such as low data processing efficiency and the lack of dynamic classification mechanisms in complex scenarios. Traditional methods rely on fixed acquisition cycles, making it difficult to optimize energy consumption through data feature analysis. In addition, they lack adaptive algorithms based on mathematical models, resulting in redundant data occupying computing resources and insufficient target recognition accuracy and real-time performance. For example, the continuous monitoring mode cannot distinguish between steady-state and transient data characteristics, resulting in a large amount of invalid data processing, exacerbating equipment energy consumption pressure. An adaptive method is urgently needed to optimize energy consumption while ensuring recognition accuracy and achieve precise control. Summary of the Invention

[0003] In order to overcome the shortcomings of low efficiency and high energy consumption in data processing, the present invention provides an artificial intelligence visual recognition and interactive control method.

[0004] The technical implementation scheme of the present invention is: an artificial intelligence visual recognition and interactive control method, comprising the following steps:

[0005] S1: Obtaining a data acquisition cycle of a camera and an intermittent acquisition cycle of a sensor in a target scene; and obtaining an intermittent working cycle of visual recognition based on the data acquisition cycle;

[0006] S2: within the data acquisition period, dividing data features into steady-state data and transient data; and obtaining change data based on the steady-state data and the transient data;

[0007] S3: adjusting the intermittent collection period of the change data within the intermittent collection period; and adjusting the data collection period of the camera and the intermittent working period of the visual recognition based on the adjusted intermittent collection period;

[0008] S4: constructing a first interactive control sequence and a second interactive control sequence based on the adjusted data collection cycle and intermittent working cycle, and determining an interactive control method according to the degree of difference between the first interactive control sequence and the second interactive control sequence.

[0009] Preferably, the step of acquiring the data collection period of the camera and the intermittent collection period of the sensor in the target scene includes:

[0010] Obtaining the data acquisition cycle of the camera and the intermittent acquisition cycle of the sensor to form N data acquisition cycles and intermittent acquisition cycles;

[0011] In the N data collection cycles, interactive control nodes are collected and defined as first interactive control nodes;

[0012] Collecting interactive control nodes during the N intermittent collection cycles and defining them as second interactive control nodes;

[0013] The data collection cycle refers to the cyclic time unit in which the camera starts and stops regularly due to energy constraints, usually manifested as alternating activation phases and dormancy phases of fixed duration;

[0014] The intermittent collection cycle refers to the discrete data acquisition rhythm designed by the environmental sensor to reduce energy consumption, which is usually manifested as a sampling trigger mechanism with discontinuous and equal time intervals.

[0015] Preferably, the step of obtaining the intermittent working cycle of visual recognition based on the data collection cycle includes:

[0016] During the data collection period, N-1 intermittent working periods of visual recognition are formed;

[0017] The intermittent working cycle refers to an intermittent operation mode set to adapt to the power supply cycle of the camera, which is usually manifested as selecting only N-1 cycles out of N device working cycles to perform image analysis tasks.

[0018] Preferably, the data characteristics are divided into steady-state data and transient data during the data acquisition period, including:

[0019] In the data collection cycle, data characteristics are divided into steady-state data and transient data based on each intermittent working cycle;

[0020] If the position and morphological characteristics of the data feature do not exceed the preset threshold value during the intermittent working cycle, the data is classified as steady-state data;

[0021] If the position and morphological features of the data feature exceed a preset threshold value during the intermittent working cycle, the data is classified as transient data.

[0022] Preferably, obtaining the change data based on the steady-state data and the transient data includes:

[0023] In the intermittent working cycle, if the steady-state data or the transient data in the next intermittent working cycle is transformed into the transient data or the steady-state data, the steady-state data or the transient data is classified as changing data.

[0024] Preferably, adjusting the intermittent collection period of the change data within the intermittent collection period includes:

[0025] If the time interval between the first interactive control node and the second interactive control node is greater than a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node exceeds a preset amount threshold, shortening the interval collection period of the change data;

[0026] If the time interval between the first interactive control node and the second interactive control node is less than or equal to a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node does not exceed a preset amount threshold, then the intermittent collection period of the change data is extended.

[0027] Preferably, adjusting the data acquisition cycle of the camera and the intermittent working cycle of visual recognition based on the adjusted intermittent acquisition cycle includes:

[0028] If the interval collection period is shortened, the weather conditions in the monitoring area are obtained;

[0029] If the interval collection period is extended, emergencies in the monitoring area are acquired;

[0030] Based on the weather conditions, adjusting the data collection period of the camera according to a first adjustment formula;

[0031] Based on the emergency event, the intermittent working cycle of the visual recognition is adjusted according to a second adjustment formula.

[0032] Preferably, the data acquisition period of the camera is adjusted according to a first adjustment formula based on the weather conditions, including: the first adjustment formula is as follows,

[0033]

[0034] in, is the adjusted data collection period, is the data collection period before adjustment, W severity is the severity of the weather, and α is the weather impact coefficient.

[0035] Preferably, the intermittent working cycle of visual recognition is adjusted according to a second adjustment formula based on the emergency event, including: the second adjustment formula is as follows,

[0036]

[0037] in, is the adjusted intermittent duty cycle, is the intermittent working cycle before adjustment, E level is the emergency level, and β is the event response coefficient.

[0038] Preferably, constructing a first interactive control sequence and a second interactive control sequence based on the adjusted data collection cycle and intermittent working cycle, and determining an interactive control method according to the degree of difference between the first interactive control sequence and the second interactive control sequence includes:

[0039] Combining the data collection period, the adjusted data collection period and the time interval into a first interactive control sequence;

[0040] Combining the intermittent working cycle, the adjusted intermittent working cycle and the time interval into a second interactive control sequence;

[0041] Based on the difference between the first interactive control sequence and the second interactive control sequence, the interactive control method of the data feature is determined; the quantitative formula of the difference is as follows:

[0042]

[0043] Where D is the degree of difference, w1 is the weight coefficient of the data acquisition cycle, w2 is the weight coefficient of the intermittent working cycle, and w1+w2=1.

[0044] Beneficial Effects: This paper proposes a visual recognition method based on data feature classification and mathematical modeling. By dynamically dividing steady-state, transient, and variable data, combined with a periodic optimization method driven by exponential and logarithmic functions, this method achieves adaptive adjustment of data acquisition and processing strategies, significantly reducing the amount of redundant data processing and improving the real-time and accuracy of target recognition. Based on a quantitative formula for the degree of difference, computing resources are dynamically allocated, optimizing system energy consumption and recognition accuracy, while balancing system stability and real-time response. This method improves the reliability and adaptability of complex terrain monitoring while maintaining low power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of the artificial intelligence visual recognition and interactive control method of the present invention;

[0046] Figure 2 Schematic diagram of the adjustment process of the intermittent collection period of variable data in the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] An artificial intelligence visual recognition and interactive control method, such as Figure 1 As shown, the following steps are included:

[0049] S1: Obtaining a data acquisition cycle of a camera and an intermittent acquisition cycle of a sensor in a target scene; and obtaining an intermittent working cycle of visual recognition based on the data acquisition cycle;

[0050] S2: within the data collection period, dividing data features into steady-state data and transient data; and obtaining change data based on the steady-state data and the transient data;

[0051] S3: adjusting the intermittent collection period of the change data within the intermittent collection period; and adjusting the data collection period of the camera and the intermittent working period of the visual recognition based on the adjusted intermittent collection period;

[0052] It should be noted that, based on the adjusted intermittent collection cycle, the camera's data collection cycle and the intermittent working cycle of visual recognition are adjusted by the first adjustment formula and the second adjustment formula respectively according to weather parameters and event levels.

[0053] S4: constructing a first interactive control sequence and a second interactive control sequence based on the adjusted data collection cycle and intermittent working cycle, and determining an interactive control method according to the degree of difference between the first interactive control sequence and the second interactive control sequence.

[0054] Obtain the data collection cycle of the camera and the intermittent collection cycle of the sensor in the target scene, including:

[0055] Obtaining the data acquisition cycle of the camera and the intermittent acquisition cycle of the sensor to form N data acquisition cycles and intermittent acquisition cycles;

[0056] In the N data collection cycles, interactive control nodes are collected and defined as first interactive control nodes;

[0057] Collecting interactive control nodes during the N intermittent collection cycles and defining them as second interactive control nodes;

[0058] The data collection cycle refers to the cyclic time unit in which the camera starts and stops regularly due to energy constraints, usually manifested as alternating activation phases and dormancy phases of fixed duration;

[0059] The intermittent collection cycle refers to the discrete data acquisition rhythm designed by the environmental sensor to reduce energy consumption, which is usually manifested as a sampling trigger mechanism with discontinuous and equal time intervals.

[0060] It’s important to note that in the wild, visual recognition and interactive control require efficient data collection under energy-constrained conditions. Traditional methods consume too much energy due to continuous monitoring and are difficult to adapt to complex terrain changes.

[0061] Data collection method,Data collection cycle of the camera: Due to energy constraints, the camera is alternately activated (working) and dormant (energy saving) according to a fixed cycle to ensure data capture during critical periods.

[0062] Intermittent sensor acquisition cycle: Environmental sensors trigger sampling at a discrete rhythm, reducing energy consumption by acquiring data at discontinuous and equal intervals.

[0063] By dividing the camera into periodic working and dormant phases and combining them with a discrete sensor triggering mechanism, we balance data integrity and energy efficiency. The camera's active phase covers key monitoring windows, while the sensor intermittently samples additional environmental parameters, forming a coordinated collection strategy.

[0064] Interactive Control Node Definition, First Interactive Control Node: The data collection cycle refers to the cyclical time unit during which a camera regularly starts and stops due to energy constraints. It typically consists of alternating active and dormant phases of fixed duration. The first interactive control node collects data during the camera's active phase and is used for visual data synchronization and target tracking.

[0065] Second Interactive Control Node: The intermittent data acquisition cycle refers to the discrete data acquisition rhythm designed by environmental sensors to reduce energy consumption. It is usually manifested as a sampling trigger mechanism with non-continuous, evenly spaced intervals. The second interactive control node collects data when the sensor triggers sampling, which is used for environmental parameter fusion and state calibration.

[0066] For example, in a field monitoring scenario, the camera's data collection cycle is 10 minutes, consisting of a 2-minute active phase and an 8-minute dormant phase. During the active phase, images are collected, forming the first interactive control node. The sensor triggers temperature and humidity sampling every 30 seconds (intermittent collection cycle), generating the second interactive control node upon triggering. These two types of nodes collaborate to drive system decision-making. If an anomaly (such as a rockfall) is detected during the active phase, the sensor is immediately awakened for high-frequency sampling, improving real-time response.

[0067] Based on the data collection cycle, an intermittent working cycle of visual recognition is obtained, including:

[0068] During the data collection period, N-1 intermittent working periods of visual recognition are formed;

[0069] The intermittent working cycle refers to an intermittent operation mode set to adapt to the power supply cycle of the camera, which is usually manifested as selecting only N-1 cycles out of N device working cycles to perform image analysis tasks.

[0070] It's important to note that intermittent operation optimizes energy consumption by adapting to the camera's power supply rhythm, proactively skipping some cycles while ensuring critical data analysis. This approach essentially selects only N-1 cycles out of N camera operating cycles to perform image analysis tasks. The skipped cycles are used for device cooling or low-power standby, avoiding energy overload caused by continuous computing.

[0071] For example, assume a camera's data collection cycle is 10 minutes (consisting of five cycles, each with 2 minutes of active and 8 minutes of dormant time). The intermittent working cycle of visual recognition selects four of these five cycles for analysis (N-1 = 4). For example, analysis is active in cycles 1-4, while cycle 5 only captures images but does not process them. Linked control: During the skipped fifth cycle, the system relies on environmental data from the sensor's intermittent collection cycles (such as sudden changes in temperature and humidity) to determine whether to initiate emergency analysis.

[0072] During the data acquisition cycle, data features are divided into steady-state data and transient data, including:

[0073] In the data collection cycle, data characteristics are divided into steady-state data and transient data based on each intermittent working cycle;

[0074] If the position and morphological characteristics of the data feature do not exceed the preset threshold value during the intermittent working cycle, the data is classified as steady-state data;

[0075] If the position and morphological features of the data feature exceed a preset threshold value during the intermittent working cycle, the data is classified as transient data.

[0076] It should be noted that within the data collection cycle, the intermittent working period is used as the time basis, and preset thresholds are used to quantify the position and morphological changes of data features to achieve the classification of steady-state and transient data. Steady-state data represents stable and unchanging targets in the scene (such as fixed field ridges), while transient data reflects dynamically changing targets (such as moving animals).

[0077] The classification basis is: Steady-state data: during the intermittent working cycle, the target position offset ≤ the preset threshold and the morphological change rate ≤ the preset threshold; transient data: the target position offset > threshold or the morphological change rate > preset threshold.

[0078] The steady-state data refers to data units whose position and morphological characteristics remain stable during intermittent working cycles. They usually appear as background targets or long-term stationary objects and require low-frequency maintenance to reduce computing load.

[0079] The transient data refers to data units whose position or morphological characteristics change significantly during an intermittent working cycle. They usually correspond to dynamic targets or sudden disturbances and require high-frequency tracking to ensure real-time control.

[0080] Preset threshold setting method and historical data analysis: Based on the distribution patterns of target position offset and morphological changes from long-term monitoring data, the upper limit of the 95% confidence interval is used as the threshold. Environmental adaptation adjustment: The threshold is dynamically adjusted based on terrain complexity (such as slope and vegetation density), and the threshold is relaxed by 20%-30% in steep slope areas. Online learning mechanism: The threshold is updated in real time through a sliding window algorithm to adapt to seasonal or sudden environmental changes.

[0081] For example, assuming a camera's data collection cycle is 10 minutes (consisting of five 2-minute activation phases), four of the intermittent working cycles are selected for analysis: Steady-state data: In the first cycle, a 0.1-meter offset (preset threshold 0.5 meters) and a 2% morphological change rate (preset threshold 5%) of the ridge position are detected. This data is marked as steady-state data, triggering the generation of a monthly maintenance report. Transient data: In the third cycle, a 1.2-meter offset (exceeding the preset threshold) and an 8% morphological change rate (exceeding the preset threshold) of the target are detected. This data is marked as transient data, immediately activating drone tracking and initiating acoustic and visual repelling. Dynamic threshold adjustment: During the rainy season, frequent landslides are detected, and the position offset threshold is automatically raised from 0.5 meters to 0.7 meters to reduce false positives.

[0082] Obtaining change data based on the steady-state data and the transient data, including:

[0083] In the intermittent working cycle, if the steady-state data or the transient data in the next intermittent working cycle is transformed into the transient data or the steady-state data, the steady-state data or the transient data is classified as changing data.

[0084] It should be noted that variable data is defined as data features that undergo state transitions between adjacent intermittent operating cycles. If data marked as steady-state or transient in one cycle switches to the opposite state (steady-state to transient or transient to steady-state) in the next cycle, it is considered variable data. This type of data reflects abnormal and sudden changes in the environment or target and requires priority processing to trigger a control response.

[0085] The variable data refers to the data units that cross the boundary between steady state and transient classification during the intermittent working cycle, which usually manifests as the target characteristics of sudden state migration and requires dynamic adjustment of the control strategy to cope with abnormal changes.

[0086] For example, from steady state to transient state: In the first cycle, the ridge position shifted by 0.1 meter (steady-state data). In the second cycle, due to heavy rain, the displacement reached 0.8 meter (exceeding the threshold of 0.5 meters). This is marked as changing data, triggering a landslide warning and activating the drainage system.

[0087] Transient → Steady State: In the third cycle, a sudden increase in cucumber pests was detected (transient data). In the fourth cycle, the target disappeared and the characteristics were stable (classified as steady-state data). It was marked as changing data, triggering the automatic spraying system and generating a pest control report.

[0088] Control response: Changing data triggers the camera's data collection cycle to be shortened to 5 minutes (originally 10 minutes), and additional computing power is allocated to continuously track high-risk targets.

[0089] This mechanism converts state transition events into control instructions, improving adaptability to dynamic environments.

[0090] Adjusting the intermittent collection period of the change data within the intermittent collection period includes:

[0091] If the time interval between the first interactive control node and the second interactive control node is greater than a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node exceeds a preset amount threshold, shortening the interval collection period of the change data;

[0092] If the time interval between the first interactive control node and the second interactive control node is less than or equal to a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node does not exceed a preset amount threshold, then the intermittent collection period of the change data is extended.

[0093] It should be noted that if Figure 2 As shown, the logic of shortening the intermittent collection cycle is that when the time interval between the first interactive control node (the control point within the camera's data collection cycle) and the second interactive control node (the control point within the sensor's intermittent collection cycle) is greater than the preset interval threshold, and the amount of variable data during this period exceeds the preset threshold, it indicates that: environmental disturbances are increasing: a large amount of variable data appears in a long time interval, such as sudden disasters (such as landslides) or high-frequency dynamic events (such as animal activities). Response urgency is increasing: it is necessary to shorten the collection cycle of variable data and increase the frequency of data updates to ensure that the control system captures anomalies in a timely manner and triggers actions (such as starting drainage or alarms). Conclusion: Shortening the intermittent collection cycle can enhance real-time monitoring capabilities and avoid missing key events.

[0094] The logic for extending the intermittent data collection period is that when the interval is less than or equal to the preset interval threshold and the amount of data changes does not exceed the preset threshold, it indicates that the environment is stable: there is little data change within a short interval, indicating a low-risk state (such as clear weather and no animal activity). Energy optimization requirements: Extending the intermittent data collection period reduces the number of sensor and camera wake-up times, lowering energy consumption and extending device battery life. Conclusion: Extending the intermittent data collection period prioritizes energy efficiency and reduces redundant data collection.

[0095] Threshold setting method, preset interval threshold: Basis: Combine the device endurance target (e.g., 30 days of continuous operation) and the event response delay tolerance (e.g., disaster response must be within 5 minutes). Calculation: Determine the average safe interval through historical data analysis (e.g., the longest interval when there is no risk is 20 minutes), and use 80% as the threshold (16 minutes).

[0096] Preset threshold: Based on the statistical fluctuation range of the change data under normal conditions (e.g., the average daily change data on sunny days is 10 times), the threshold is taken as 3 times the standard deviation (e.g., 15 times). Dynamic adjustment: The threshold is automatically increased by 20% (18 times) during the rainy season to avoid frequent adjustments due to weather interference.

[0097] For example, the camera's data collection cycle is 10 minutes (including five 2-minute activation phases), and the sensor samples every 5 minutes. Trigger conditions for shortening the intermittent collection cycle: Time interval: The interval between the first interactive control node (end of the camera's data collection cycle) and the second interactive control node (end of the sensor's intermittent collection cycle) is 18 minutes (preset threshold 15 minutes). Changed data: 20 changes in data were detected during the period (preset threshold 15 times), including 12 mountain displacements and 8 sudden increases in cucumber pests. Action: Shorten the intermittent collection cycle of changed data from 5 minutes to 3 minutes, increase the monitoring density, and immediately start the drone spraying operation.

[0098] Trigger condition for extending the intermittent data collection period: Interval: 12 minutes between nodes (≤ 15-minute threshold). Data changes: Only 8 data changes were detected (< 15 threshold), all of which were minor leaf drop disturbances. Action: Extend the intermittent data collection period from 5 minutes to 8 minutes and disable secondary sensors to save energy.

[0099] Adjusting the data acquisition cycle of the camera and the intermittent working cycle of the visual recognition based on the adjusted intermittent acquisition cycle includes:

[0100] If the interval collection period is shortened, the weather conditions in the monitoring area are obtained;

[0101] If the interval collection period is extended, the emergency events in the monitoring area are obtained;

[0102] Based on the weather conditions, adjusting the data collection period of the camera according to a first adjustment formula;

[0103] Based on the emergency event, the intermittent working cycle of the visual recognition is adjusted according to a second adjustment formula.

[0104] It should be noted that by shortening the interval collection period of the sensor, the system obtains weather data (such as rainstorm, sandstorm) in real time and dynamically optimizes the data collection period of the camera based on the first mathematical model (exponential function). severity By formula Computing drives the compression of camera data acquisition cycles to increase image data density and ensure the integrity of key scenes. When unexpected events are detected during extended intermittent acquisition cycles, processing resources are adjusted based on event priority. Extended intermittent acquisition cycles are often associated with low-risk environments. In these situations, computing power is released to track unexpected events (such as the sudden appearance of wild animals), avoiding idle resources.

[0105] This design makes up for the deficiency of the traditional method in separating the environment and event response. Through a two-way coupling mechanism (weather-driven collection frequency, event-driven processing intensity), it achieves a dynamic balance between energy consumption and accuracy, rather than a one-way adjustment under fixed rules.

[0106] For example, the basic camera's data collection cycle is 10 minutes (including five activation phases), and the sensor's intermittent collection cycle is 5 minutes. To shorten the intermittent collection cycle (to obtain weather data), the sensor detects heavy rain (the interval decreases from 5 minutes to 3 minutes), triggering the first adjustment formula and reducing the data collection cycle from 10 minutes to 6 minutes (with 3 minutes per activation phase). To extend the intermittent collection cycle (to obtain events), the sensor interval is extended to 8 minutes on a sunny day. During this time, the infrared sensor detects a sudden change in banana fruit ripeness (an unexpected event), triggering the second adjustment formula and reducing the intermittent duty cycle from 1 to 0 (full cycle analysis).

[0107] Based on the weather conditions, the data acquisition period of the camera is adjusted according to a first adjustment formula, including: the first adjustment formula is as follows,

[0108]

[0109] in, is the adjusted data collection period, is the data collection period before adjustment, W severity is the severity of the weather, and α is the weather impact coefficient.

[0110] It should be noted that the parameter description: Basic data collection period (e.g. 10 minutes by default on sunny days). severity: Weather severity (0 = sunny, 1 = extremely bad), quantifying the intensity of environmental interference (e.g., heavy rain is 0.8, dust is 0.6). α: Weather influence coefficient (0.1-0.3), controlling the adjustment range of the cycle. The larger the value, the more sensitive the response.

[0111] The exponential function dynamically compresses the period. The worse the weather (W severity ↑), The smaller the value, This design proactively increases the frequency of data collection in extreme weather conditions, ensuring that key environmental changes (such as mountain displacement and water level surges) are captured in a timely manner, avoiding the risk of missed detections caused by sudden environmental changes in traditional fixed cycles.

[0112] Example, basic data collection cycle Minutes, α=0.2. Heavy rain weather (W severity =0.8): Minutes, Action: Data collection cycle is shortened from 10 minutes to 8.5 minutes, the activation phase ratio is increased, and two additional mountain images are collected per cycle. severity =0.5): minute.

[0113] Based on the emergency, the intermittent working cycle of the visual recognition is adjusted according to the second adjustment formula, including: the second adjustment formula is as follows,

[0114]

[0115] in, is the adjusted intermittent duty cycle, is the intermittent working cycle before adjustment, E level is the emergency level, and β is the event response coefficient.

[0116] It should be noted that the parameter description: The basic intermittent working cycle indicates the default frequency of visual recognition execution (such as 4 times every 5 device cycles). level : Emergency level (0 = no event, 1 = low risk, 2 = medium risk, 3 = high risk), quantifies the urgency of the event. β: Event response coefficient (0.1∽0.5), controls the adjustment range of the recognition frequency. A larger value indicates a more aggressive response.

[0117] Logarithmic function: ln(E level +1) Converts linear event levels to nonlinear gains, with faster gains at low levels and slower gains at high levels to avoid over-response.

[0118] Formula meaning: The higher the emergency level (E level ↑), visual recognition frequency increased ( ↑), reduce the number of cycle jumps to strengthen monitoring.

[0119] For example, the camera performs 4 visual recognitions per 5 cycles (N=5) by default. β=0.3.

[0120] Banana plant disease outbreak (E level =2):

[0121] Action: The camera's data collection cycle is 10 minutes (including five 2-minute activation phases), and the visual recognition defaults to performing 4 analyses (1 skip) every 5 cycles. level =3), the analysis is performed 5 times in the full cycle (skipping cycle is canceled).

[0122] Heavy rain landslide (E level =3):

[0123] Action: Continuously perform 5 identifications, simultaneously shorten the camera's data collection cycle to 6 minutes (originally 10 minutes), and double the density of mountain displacement monitoring.

[0124] Based on the adjusted data collection cycle and intermittent working cycle, a first interactive control sequence and a second interactive control sequence are constructed, and an interactive control method is determined according to the degree of difference between the first interactive control sequence and the second interactive control sequence, including:

[0125] Combining the data collection period, the adjusted data collection period and the time interval into a first interactive control sequence;

[0126] Combining the intermittent working cycle, the adjusted intermittent working cycle and the time interval into a second interactive control sequence;

[0127] Based on the difference between the first interactive control sequence and the second interactive control sequence, the interactive control method of the data feature is determined; the quantitative formula of the difference is as follows:

[0128]

[0129] Where D is the degree of difference, w1 is the weight coefficient of the data acquisition cycle, w2 is the weight coefficient of the intermittent working cycle, and w1+w2=1.

[0130] It should be noted that the construction and function of the interactive control sequence, the first interactive control sequence: component elements: data collection cycle The basic working cycle of the camera (such as 10 minutes). The adjusted data collection cycle The cycle adjusted according to the weather formula (e.g., shortened to 8.5 minutes during heavy rain). Time interval (Δt): The time span between adjacent control nodes within a sequence (e.g., the 2-minute activation phase within each camera cycle). Purpose: Comprehensively reflects the dynamic adjustment process of data acquisition and records the impact of environmental changes on the camera's operating rhythm.

[0131] Second interactive control sequence: Component: intermittent working cycle The default number of skip cycles for visual recognition (e.g. skip once every 5 cycles). Adjusted intermittent working cycle

[0132] The period adjusted according to the event formula (e.g., skipping the period in the event of an emergency). Time interval (Δt): The time span synchronized with the first sequence. Purpose: Indicates changes in visual recognition processing intensity and records the driving effect of emergencies on computing power allocation.

[0133] Parameter analysis of the formula for quantifying the degree of difference: w1, w2: weight coefficients (w1+w2=1), reflecting the optimization priority of data acquisition and visual processing. If the focus is on energy consumption optimization: w1=0.7, w2=0.3 (more attention is paid to camera cycle adjustment). If the focus is on response speed: w1=0.3, w2=0.7 (more attention is paid to visual processing adjustment). D value: degree of difference, measuring the comprehensive amplitude of parameter adjustment. Function: High D value: indicates that the environment or event has caused a significant adjustment of the parameters, and a high-intensity control strategy (such as a disaster protocol) needs to be activated. Low D value: indicates stable operation, maintaining the baseline strategy or energy-saving mode.

[0134] The determination logic of the interactive control method dynamically selects the control mode based on the difference degree D: D>D threshold (High difference): Action: Synchronize the camera cycle and the visual recognition jump cycle to start multi-device collaboration (such as drone + sensor). Example: Heavy rain causes D = 5.2 (threshold D threshold =4.0), triggering all-weather monitoring and shutting down non-core sensors. D≤D threshold (Low variance): Action: Maintain operation according to adjusted parameters, responding only to changes in data. Example: Sunny, D = 2.3, maintain camera cycle for 10 minutes, and visual recognition jump once every 5 cycles.

[0135] Example, basic parameters: minute, (jump 1 time), w1 = 0.6, w2 = 0.4, D threshold =4.0.

[0136] Heavy Rain Weather Adjustment: First Sequence Minutes (shortened by 2 minutes), Δt = 2 minutes. Second sequence: (Cancel the jump cycle), Δt = 2 minutes. Difference calculation:

[0137] Control decision: D < 4.0, maintain the current adjustment parameters, but trigger local response (such as high-frequency scanning of mountain displacement areas).

[0138] Risk of branch breakage + heavy rain (extreme event): First sequence: Minutes, Δt = 1.5 minutes. Second sequence: Δt=1.5min. Difference calculation:

[0139] Control Decisions:

[0140] D>4.0 does not hold, but is close to the threshold, enabling backup resources (such as waking up the second camera).

[0141] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An artificial intelligence visual recognition and interactive control method, characterized in that: The following steps are involved: S1: Obtaining a data acquisition cycle of a camera and an intermittent acquisition cycle of a sensor in a target scene; and obtaining an intermittent working cycle of visual recognition based on the data acquisition cycle; S2: within the data collection period, dividing data features into steady-state data and transient data; and obtaining change data based on the steady-state data and the transient data; S3: adjusting the intermittent collection period of the change data within the intermittent collection period; and adjusting the data collection period of the camera and the intermittent working period of the visual recognition based on the adjusted intermittent collection period; S4: constructing a first interactive control sequence and a second interactive control sequence based on the adjusted data collection cycle and intermittent working cycle, and determining an interactive control method according to the degree of difference between the first interactive control sequence and the second interactive control sequence.

2. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The step of obtaining the data collection period of the camera and the intermittent collection period of the sensor in the target scene includes: Obtaining the data acquisition cycle of the camera and the intermittent acquisition cycle of the sensor to form N data acquisition cycles and intermittent acquisition cycles; In the N data collection cycles, interactive control nodes are collected and defined as first interactive control nodes; Collecting interactive control nodes during the N intermittent collection cycles and defining them as second interactive control nodes; The data collection cycle refers to the cyclic time unit in which the camera starts and stops regularly due to energy constraints, usually manifested as alternating activation phases and dormancy phases of fixed duration; The intermittent collection cycle refers to the discrete data acquisition rhythm designed by the environmental sensor to reduce energy consumption, which is usually manifested as a sampling trigger mechanism with discontinuous and equal time intervals.

3. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The step of obtaining an intermittent working cycle of visual recognition based on the data collection cycle includes: During the data collection period, N-1 intermittent working periods of visual recognition are formed; The intermittent working cycle refers to an intermittent operation mode set to adapt to the power supply cycle of the camera, which is usually manifested as selecting only N-1 cycles out of N device working cycles to perform image analysis tasks.

4. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The data characteristics are divided into steady-state data and transient data during the data acquisition period, including: In the data collection cycle, data characteristics are divided into steady-state data and transient data based on each intermittent working cycle; If the position and morphological characteristics of the data feature do not exceed the preset threshold value during the intermittent working cycle, the data is classified as steady-state data; If the position and morphological features of the data feature exceed a preset threshold value during the intermittent working cycle, the data is classified as transient data.

5. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The obtaining of change data based on the steady-state data and the transient data includes: In the intermittent working cycle, if the steady-state data or the transient data in the next intermittent working cycle is transformed into the transient data or the steady-state data, the steady-state data or the transient data is classified as changing data.

6. The artificial intelligence visual recognition and interactive control method according to claim 2, characterized in that: The step of adjusting the intermittent collection period of the change data within the intermittent collection period includes: If the time interval between the first interactive control node and the second interactive control node is greater than a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node exceeds a preset amount threshold, shortening the interval collection period of the change data; If the time interval between the first interactive control node and the second interactive control node is less than or equal to a preset interval threshold, and the amount of the change data existing in the time interval between the first interactive control node and the second interactive control node does not exceed a preset amount threshold, then the intermittent collection period of the change data is extended.

7. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The step of adjusting the camera's data acquisition cycle and the intermittent working cycle of visual recognition based on the adjusted intermittent acquisition cycle includes: If the interval collection period is shortened, the weather conditions in the monitoring area are obtained; If the interval collection period is extended, emergencies in the monitoring area are acquired; Based on the weather conditions, adjusting the data collection period of the camera according to a first adjustment formula; Based on the emergency event, the intermittent working cycle of the visual recognition is adjusted according to a second adjustment formula.

8. The artificial intelligence visual recognition and interactive control method according to claim 7, characterized in that: The data acquisition period of the camera is adjusted according to a first adjustment formula based on the weather conditions, including: the first adjustment formula is as follows, in, is the adjusted data collection period, is the data collection period before adjustment, W severity is the severity of the weather, and α is the weather impact coefficient.

9. The artificial intelligence visual recognition and interactive control method according to claim 7, characterized in that: The intermittent working cycle of visual recognition is adjusted according to a second adjustment formula based on the emergency event, including: the second adjustment formula is as follows, in, is the adjusted intermittent duty cycle, is the intermittent working cycle before adjustment, E level is the emergency level, and β is the event response coefficient.

10. The artificial intelligence visual recognition and interactive control method according to claim 1, characterized in that: The step of constructing a first interactive control sequence and a second interactive control sequence based on the adjusted data collection cycle and the intermittent working cycle, and determining an interactive control method based on a degree of difference between the first interactive control sequence and the second interactive control sequence, includes: Combining the data collection period, the adjusted data collection period and the time interval into a first interactive control sequence; Combining the intermittent working cycle, the adjusted intermittent working cycle and the time interval into a second interactive control sequence; Based on the difference between the first interactive control sequence and the second interactive control sequence, the interactive control method of the data feature is determined; the quantitative formula of the difference is as follows: Where D is the degree of difference, w1 is the weight coefficient of the data acquisition cycle, w2 is the weight coefficient of the intermittent working cycle, and w1+w2=1.

Images