System and method for vegetable detection

Abstract

The invention discloses a vegetable detection system and method, and relates to the technical field of agricultural product quality detection, and the method comprises the following steps: collecting an electromagnetic change signal, a sensing line noise signal and a power supply fluctuation signal at a vegetable detection site, drawing an interference change curve according to the collection result, recording the time positions of sudden rising and zero returning of the detection signal; and identifying an interference source according to the interference change curve, analyzing a propagation direction and a coupling path of electromagnetic waves, extracting an interference frequency band and a continuous time period, and forming a signal anomaly comparison table. According to the invention, through multi-source interference signal acquisition and feature recognition, real-time sensing and adaptive avoidance of electromagnetic interference are realized, and stability of vegetable detection signals is guaranteed; the detection rhythm is planned based on the interference risk list and the sampling and illumination periods are dynamically adjusted, so that signal drift and fluctuation are effectively suppressed, continuous and accurate detection in a complex environment is realized, and the detection reliability and the data consistency are improved.

Description

Technical Field

[0001] This invention belongs to the field of agricultural product quality testing technology, specifically, it relates to a system and method for testing vegetables. Background Technology

[0002] Vegetable testing refers to the process of identifying and evaluating key indicators such as quality, safety, and freshness of vegetables during production, distribution, and sales by utilizing multi-source sensing and data analysis technologies. Intelligent sensors collect information in real time on the color, odor, firmness, moisture content, temperature, and surface reflectance spectrum of vegetables. This multi-dimensional data is then input into an analysis system, which combines signal processing and pattern recognition technologies to comprehensively assess the appearance characteristics, nutritional components, and pesticide residues of vegetables, thereby achieving automated and intelligent quality testing. This process, centered on non-destructive testing and intelligent data processing, significantly improves testing efficiency and accuracy, providing technical support for agricultural product quality and safety supervision and intelligent grading.

[0003] In existing technologies, when vegetable detection devices operate near high-power detection equipment, the sensing circuits are highly susceptible to electromagnetic interference. High-power equipment generates strong electromagnetic fields and instantaneous pulse signals during operation. These interference signals can enter the sensing circuits through spatial radiation or power coupling, causing drastic fluctuations in the sensor output signal within a very short time. Such fluctuations can lead to sudden spikes or instantaneous drops in detection values, which the system may misinterpret as abnormal changes in vegetable quality, resulting in biased identification results. Especially in multi-sensor parallel detection environments, the superposition effect of interference is more likely to cause data distortion, affecting the overall detection accuracy and stability. Summary of the Invention

[0004] This invention addresses the technical problems mentioned in the background section by providing a system and method for vegetable detection. Through multi-source interference signal acquisition and feature recognition, it achieves real-time perception and adaptive avoidance of electromagnetic interference, ensuring stable vegetable detection signals. Based on an interference risk list, it plans the detection rhythm and dynamically adjusts the sampling and illumination cycles, effectively suppressing signal drift and fluctuations, enabling continuous and accurate detection in complex environments, and improving detection reliability and data consistency.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A method for detecting vegetables includes the following steps: Electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals are collected at the vegetable testing site. Interference change curves are plotted based on the collected results, and the time and position of the detection signal rise and return to zero are recorded to obtain basic data for interference analysis. The source of interference is identified based on the interference change curve, the propagation direction and coupling path of electromagnetic waves are analyzed, the interference frequency range and duration range are extracted, and a signal anomaly comparison table is formed to provide a basis for retrospective analysis of detection data. Based on the signal anomaly comparison table, the data of each batch of vegetable testing were retrospectively analyzed to find time misalignment, batch overlap and sensor saturation of the test data, determine the trigger point and duration of abnormal signals, generate an interference risk list, and provide a reference for the planning of the testing rhythm. Based on the interference risk list, the detection process is rhythmically planned to determine the sampling sequence, exposure time, and power supply quiescent phase, forming an interference suppression scheme to guide the dynamic adjustment of subsequent detection processes; During periods of high risk of interference, adaptive detection and adjustment are implemented based on the interference suppression scheme. Reverse silent sampling, intermittent sampling period and delayed sampling window are set to dynamically coordinate sampling time and illumination period, balance signal energy fluctuations, and stabilize the output signal during vegetable detection.

[0006] The following are further optimizations of the above technical solution by the present invention: The process of plotting the interference change curve is as follows: Signal acquisition units were set up at the vegetable testing site to collect electromagnetic change signals, sensor circuit noise signals and power fluctuation signals, respectively, and a reference potential benchmark was established to determine the sampling frequency, sampling period and signal channel response threshold. Based on the collected electromagnetic change signals, sensor line noise signals and power fluctuation signals, time sequence alignment and amplitude standardization are performed to unify the signals onto the same time coordinate axis to form a time series framework. Based on the time series framework, interference variation curves are plotted, and electromagnetic variation signals, sensor line noise signals, and power fluctuation signals are represented as independent curves, with interference concentration intervals marked. Based on the interference variation curve, record the time positions of the sudden rise and return to zero of the detection signal to form an interference time table and associate it with the real-time amplitude data of each signal to obtain the basic data for interference analysis.

[0007] Further optimization: When plotting the interference change curve, the order of interference propagation is determined by comparing the peaks, troughs, and abrupt changes of the electromagnetic change signal, the sensor line noise signal, and the power supply fluctuation signal. The time position is recorded synchronously when the signal amplitude suddenly increases and returns to zero, so as to distinguish the direction of interference energy propagation and the duration interval, and to establish an interference distribution mapping for subsequent interference analysis.

[0008] Further optimization: The process of forming the signal anomaly comparison table is as follows: Based on the interference change curve, the electromagnetic change signal, the sensor line noise signal and the power fluctuation signal are divided into intervals, and the layered comparison is carried out on the same time axis to determine the time period and starting order of the interference energy concentration. The synchronization of amplitude changes between electromagnetic change signals and sensing line noise signals is analyzed based on the time interval table of interference change curves in order to identify the propagation direction of electromagnetic waves and determine the coupling path. By combining the propagation direction and coupling path, the frequency components of the interference variation curve are divided into intervals, and the interference frequency segment and duration segment are extracted. Based on the analysis results of propagation direction, coupling path, frequency band, and duration, a signal anomaly comparison table is formed, recording the time and location of the interference, propagation direction, coupling path, and frequency band information, which is used for backtracking analysis of detection data.

[0009] Further optimization: When forming the signal anomaly comparison table, the start time, end time, propagation direction, coupling path, and frequency segment of the interference are correlated with the time series of electromagnetic change signals, sensor line noise signals, and power fluctuation signals. The interference energy concentration interval is marked in the comparison table to achieve a unified mapping of interference characteristics in the time, space, and frequency dimensions, which is used for interference location and comparison analysis of subsequent detection data.

[0010] Further optimization: The steps for generating the interference risk list are as follows: Using the signal anomaly comparison table as an index, time series matching was performed on the original test data of each batch of vegetables to establish a mapping relationship between batch data and interference periods; After completing the time series matching, time misalignment analysis and batch cross-analysis are performed on the detection data of each batch. The interference range is determined based on the interference frequency band and recorded in the time association table. Based on the time correlation table, saturation phenomenon analysis was performed on the amplitude changes of sensor output signals in each batch of test data to determine the interference trigger point and interference duration range; After completing the analysis of time misalignment, batch overlap and saturation phenomena, the analysis results are summarized to generate an interference risk list, which records the trigger time, duration, affected batch and sensor channel information of the interference event, and is used for detection rhythm planning.

[0011] Further optimization: The process of forming the interference suppression scheme is as follows: Using the signal anomaly comparison table as an index, time series matching was performed on the original test data of each batch of vegetables to establish the mapping relationship between batch data and interference periods and form a batch time index table; After completing the time series matching, time misalignment analysis and batch cross-analysis are performed on the detection data of each batch, and the identification results of time misalignment and batch cross-analysis are recorded in the time association table. Based on the time correlation table, the amplitude changes of sensor output signals in each batch of test data are analyzed for saturation phenomena to determine the interference trigger point and the duration of interference and mark the saturation interval. After completing the analysis of time misalignment, batch overlap and saturation phenomena, the analysis results are summarized to generate an interference risk list, which records the trigger time, duration, affected batches and sensor channel information of the interference events, and is used for detection rhythm planning.

[0012] Further optimization: Interference events in the interference risk list are arranged in order of trigger time and are classified and labeled according to the duration of interference and the affected range of the sensing channel. The interference risk list also records the detection stage information when interference occurs. The periodic characteristics of the interference source are determined by the repetition interval of the interference events, which is used to optimize and adjust the subsequent detection rhythm and formulate interference avoidance strategies.

[0013] Further optimization: During periods of high interference risk, adaptive detection adjustments are implemented based on the interference suppression scheme. The following steps are taken to dynamically coordinate sampling time with the illumination cycle, setting reverse silent sampling, intermittent sampling periods, and delayed sampling windows: Within the high-interference range identified in the interference risk list, a reverse silent sampling phase is set according to the interference suppression scheme. By setting the sampling trigger point within the attenuation range after the interference peak, sampling time avoidance is achieved. After completing the reverse silent sampling phase, an intermittent sampling period is constructed based on the risk level distribution in the interference suppression scheme. The sampling interval is adjusted to achieve time avoidance between the sampling time and the interference fluctuation period. After completing the intermittent sampling period, a delayed sampling window is established, and the sampling action is delayed after the interference peak ends in order to take advantage of the signal stability interval during the interference attenuation phase. After constructing the reverse silent sampling, intermittent sampling period and delayed sampling window, the sampling time and illumination period are dynamically coordinated and executed. The sampling and illumination are synchronized and balanced through time matching to stabilize the output of the detection signal.

[0014] The present invention also provides a vegetable detection system for implementing the above-mentioned vegetable detection method, comprising an interference sensing module, a feature parsing module, an anomaly diagnosis module, a rhythm planning module, and an adaptive control module: The interference sensing module collects electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals at the vegetable testing site. Based on the collected results, it plots the interference change curve and records the time and position of the detection signal rise and return to zero, thus obtaining basic data for interference analysis. The feature analysis module identifies the source of interference based on the interference change curve, analyzes the propagation direction and coupling path of electromagnetic waves, extracts the interference frequency segment and duration segment, and generates a signal anomaly comparison table to provide a basis for retrospective analysis of detection data. The anomaly diagnosis module performs retrospective analysis on the data of each batch of vegetable testing based on the signal anomaly comparison table, finds time misalignment, batch overlap and sensor saturation of the test data, determines the trigger point and duration of the abnormal signal, and generates an interference risk list to provide a reference for the planning of the testing schedule. The rhythm planning module plans the rhythm of the detection process based on the interference risk list, determines the sampling sequence, exposure time, and power supply static phase, and forms an interference suppression scheme to guide the dynamic adjustment of the subsequent detection process; The adaptive control module implements adaptive detection and adjustment based on the interference suppression scheme during periods of interference risk. It sets reverse silent sampling, intermittent sampling period and delayed sampling window, dynamically coordinates sampling time with light cycle, balances signal energy fluctuations, and stabilizes the output signal during vegetable detection.

[0015] The present invention, by adopting the above technical solution, has at least the following beneficial effects: 1. This invention introduces a multi-source acquisition and feature recognition mechanism for interference signals during vegetable testing, enabling the testing device to perceive interference in real time and dynamically adjust the sampling rhythm in complex electromagnetic environments. This effectively avoids signal abrupt changes caused by transient electromagnetic waves or power fluctuations. By establishing interference change curves and signal anomaly comparison tables, the invention achieves the identification of interference sources and the quantification of their impact range. This enables the testing device to have adaptive perception and avoidance capabilities during operation, improving the stability and reliability of vegetable quality testing data.

[0016] 2. This invention achieves dynamic balance between sampling time, exposure stage, and illumination cycle within the interference range by planning and adaptively adjusting the detection rhythm based on an interference risk list. The detection device can actively adjust the sampling window and the power supply standby stage according to the interference intensity and duration, weakening the coupling effect of electromagnetic energy on the sensing signal and fundamentally suppressing signal drift and fluctuation accumulation. This method enables continuous and stable detection under interference environments, providing an accurate and repeatable data foundation for vegetable quality analysis. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for detecting vegetables according to the present invention; Figure 2 This is a schematic diagram of a vegetable detection system according to the present invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings; however, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The method for detecting vegetables shown includes the following steps: Electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals are collected at the vegetable testing site. Interference change curves are plotted based on the collected results, and the time and position of the detection signal rise and return to zero are recorded to obtain basic data for interference analysis. In vegetable testing, to ensure the acquisition of fundamental data usable for subsequent interference analysis under complex electromagnetic environments, the entire process employs multi-stage acquisition and processing to systematically acquire and organize electromagnetic change signals, sensor circuit noise signals, and power supply fluctuation signals. This generates a complete interference change curve and accurately records the timing of signal spikes and zeroing. The specific implementation steps are as follows: Signal acquisition units are deployed at the vegetable testing site to address three signal sources: electromagnetic changes, sensor circuit noise, and power fluctuations. Before startup, each unit initializes the spatial location of the testing area, establishes a reference potential, and determines the acquisition frequency, sampling period, and response threshold of the signal channel. Electromagnetic change signals are acquired using a spatial electric field sensing element, which continuously monitors the electromagnetic environment at the vegetable testing site, capturing electromagnetic field fluctuations caused by the operation of high-power equipment, the start / stop of conveyor motors, or changes in lighting drive current. Sensor circuit noise signals are acquired synchronously at both ends of the vegetable testing circuit, detecting voltage disturbances and current fluctuations, with particular attention paid to short-duration pulses, weak high-frequency components, and instantaneous peaks. Power fluctuation signals are acquired through synchronous detection at the input and load ends, used to monitor the power stability of the vegetable testing equipment and identify instantaneous interference sources caused by power supply fluctuations. These three types of signals are acquired in parallel under the same time reference, forming a raw data sequence containing time, amplitude, and frequency components, providing a foundation for subsequent comprehensive processing.

[0020] After acquiring the raw signals, timing alignment and amplitude standardization are performed on signals from different sources to unify electromagnetic change signals, sensor line noise signals, and power fluctuation signals onto the same time coordinate axis, thereby eliminating the offset differences caused by sampling delay. During the processing, the integrity of the correspondence between the acquisition time marker and the signal amplitude is maintained to ensure that the correspondence between various types of signals is clear at any time. Through a unified time reference, the synchronous change trend of the three types of signals can be accurately identified within a certain time period, providing continuous signal input for subsequent plotting of interference change curves. The focus of this stage is to establish a time series framework for the interference signals, so that different signal sources can show dynamic changes in the same coordinate system, thereby revealing the correlation and amplitude change trajectory before and after the interference occurs.

[0021] After establishing the time series framework, interference variation curves were plotted for the three types of signals that had undergone unified processing. During the plotting process, time was used as the horizontal axis and signal amplitude as the vertical axis, and electromagnetic variation signals, sensor line noise signals, and power supply fluctuation signals were represented as independent curves, forming a comparative relationship in the same chart. By observing the intersection, overlap, and peak alignment of multiple curves in the time dimension, the concentrated intervals and fluctuation patterns of interference could be identified. When the amplitude of the electromagnetic variation signal rises rapidly within a certain time period, and the sensor line noise signal shows pulse amplification and the power supply fluctuation signal shows voltage transients, it can be determined that this time period is a potential interference zone. At this time, the relative positional relationship between the curve intersections and peak values ​​can determine the order of interference propagation and the duration of interference energy. After the curves were plotted, the fluctuation intervals were marked to establish a preliminary interference distribution mapping, which is used to support subsequent recording of signal surges and zeroing.

[0022] Based on the interference variation curve, the time positions of signal spikes and zeroing are recorded. This process uses the interference variation curve as a reference to analyze the local extreme points of all acquired signals one by one. When the instantaneous amplitude of the electromagnetic change signal exceeds a set threshold, this moment is marked as a spike event, and the changing trends of the sensor line noise signal and power fluctuation signal at that moment are simultaneously checked to determine the synergistic effect of the interference. For cases where the signal amplitude rapidly drops from a high value to a baseline level within a short period, this moment is defined as a zeroing event and is also time-marked. All spike and zeroing time positions are recorded in the interference time table in chronological order, and simultaneously correlated with the real-time amplitude data of the electromagnetic change signal, sensor line noise signal, and power fluctuation signal to form a complete interference event recording sequence. After completing the time position recording, the interference variation curve, interference time table, and original signal data are saved accordingly to provide data support for subsequent interference source analysis and signal stability assessment.

[0023] The source of interference is identified based on the interference change curve, the propagation direction and coupling path of electromagnetic waves are analyzed, the interference frequency range and duration range are extracted, and a signal anomaly comparison table is formed to provide a basis for retrospective analysis of detection data. After plotting the interference variation curves and recording the time locations, in order to identify the interference sources from multiple dimensions and extract the key characteristic parameters of the interference, the interference curves were subjected to hierarchical analysis and comparative processing. This was done to clarify the propagation direction, coupling path, frequency distribution, and duration range of electromagnetic waves in the vegetable testing site, and ultimately to form a signal anomaly comparison table that can be used for backtracking of testing data. The specific implementation steps are as follows: Based on the obtained interference variation curves, the fluctuation patterns of different types of signals are divided into intervals, and electromagnetic variation signals, sensor line noise signals, and power supply fluctuation signals are compared layer by layer on the same time axis. By continuously observing the distribution of peaks, troughs, and abrupt change points in the interference curves, the time periods of concentrated interference energy and the starting sequence of interference can be preliminarily identified. For these concentrated fluctuation intervals, a time-marked interval table is established to mark the main time windows of interference activity. In this process, special attention is paid to the time delay of the sensor line noise signal response after the sudden rise of the electromagnetic variation signal. By comparing the delay with the change in signal amplitude, the propagation direction and path of the interference wave can be inferred. When the sudden rise of the electromagnetic variation signal occurs before the sensor line noise signal, it indicates that the interference may originate from external electromagnetic radiation. If the sensor line noise signal occurs first and the electromagnetic variation signal lags behind, it may originate from coupling interference from the power supply line. Through this comparison of time sequence, the propagation direction of interference in space and in the line is obtained, providing a reference for subsequent coupling path analysis.

[0024] After determining the direction of interference propagation, the signal fluctuation amplitudes at different time intervals in the interference variation curve are layered and organized to identify the coupling paths of electromagnetic waves. This process uses a time interval table of the interference variation curve as a reference to refine the correspondence between the electromagnetic variation signal and the sensor circuit noise signal within each time interval. By analyzing the synchronicity between the amplitude changes of the electromagnetic variation signal and the response of the sensor circuit noise signal, it can be determined whether the interference energy enters the detection circuit through spatial radiation, ground coupling, or power input coupling. When the electromagnetic variation signal and the power fluctuation signal simultaneously exhibit peak phenomena within the same time interval, it indicates that the interference may be conducted through the power supply path. When the electromagnetic variation signal and the sensor circuit noise signal exhibit phase delay characteristics in space, it indicates that the interference energy may propagate through the air and be induced coupled through the sensor circuit. Through this process, a correspondence between the coupling paths of the interference wave and the sensor circuit can be established in the time domain, and the interference intensity of different paths can be qualitatively classified.

[0025] Based on the determined propagation direction and coupling path, the frequency components of the interference variation curve are divided into intervals to extract the main interference frequency range and duration range. At this stage, the repetitive oscillation intervals of each signal are identified using the time signature and fluctuation amplitude of the interference variation curve as a benchmark. When the electromagnetic change signal exhibits periodic rise and fall within a certain time period, this time period is determined as the effective range of the specific frequency interference. By comparing the repetition periods of the electromagnetic change signal, the sensor line noise signal, and the power supply fluctuation signal within the same time period, the range of the main interference frequency range can be determined. The duration range is extracted based on the start and end points of the fluctuation, indicating the duration of the interference energy in the time dimension. This process combines the time series and amplitude changes of the interference variation curve to form a dual description of the interference effect in the frequency and time domains, giving the interference characteristics a quantifiable and comparable basis.

[0026] After extracting interference features, a signal anomaly comparison table is generated based on the analysis results of propagation direction, coupling path, frequency band, and duration. This table is used for subsequent data retrospective analysis. The table uses time as the primary index, recording the synchronization characteristics and correspondences of electromagnetic change signals, sensor line noise signals, and power fluctuation signals within each time period. The table includes information such as the start and end times of the interference, the main frequency band, duration, propagation direction, and coupling path. To ensure consistency in subsequent analysis, the table associates each interference event with the corresponding signal change curve number and identifies the concentrated area of ​​interference energy and the type of signal affected. When retrospectively analyzing subsequent data, the time position and frequency band information in the table can be used to quickly locate the interference area, thereby determining the corresponding interference cause of abnormal fluctuations in the vegetable detection data. This signal anomaly comparison table not only reflects the temporal, spatial, and frequency dimensions of interference features but also establishes a direct mapping relationship between interference identification and data comparison, providing clear data support for subsequent detection rhythm planning and sampling adjustments.

[0027] Based on the signal anomaly comparison table, the data of each batch of vegetable testing were retrospectively analyzed to find time misalignment, batch overlap and sensor saturation of the test data, determine the trigger point and duration of abnormal signals, generate an interference risk list, and provide a reference for the planning of the testing rhythm. After establishing the signal anomaly comparison table, in order to further identify the specific impact of interference on the testing process of each batch of vegetables from the perspective of detection data, historical detection data were retrospectively analyzed to identify data time misalignment, batch overlap, and sensor saturation. The trigger points and duration ranges of abnormal signals were determined using interference characteristics as an index, thus forming a complete interference risk list to provide a basis for planning subsequent testing schedules. The specific implementation steps are as follows: Using a signal anomaly comparison table as an index, time-series matching was performed on the raw test data of each batch of vegetables. This matching process was primarily driven by the temporal location of the interference events. By comparing the time markers in the interference comparison table with the sampling time of each batch of test data, a mapping relationship between batch data and interference periods was established. In this way, it was possible to determine which moments during the testing process of each batch were within the interference range and which moments were in a relatively stable region. When the sampling time of a test batch partially overlapped with the interference time window, the data of that batch was marked as the interfered data segment. At this point, to avoid data confusion caused by time overlap between different batches, all test batches were rearranged according to the sampling start and end times to establish a batch time index table, ensuring that subsequent analysis was conducted under a unified time benchmark. Through this sub-step, a correspondence between the interference comparison table and the test batch data was formed, providing a foundation for further identification of data anomalies.

[0028] After completing time series matching, time misalignment analysis is performed on the data of each batch. The core of time misalignment analysis is to compare the consistency of time response of data from different sensor channels in the same batch of detection. Since the vegetable detection process usually involves multi-dimensional sensor signals such as color, odor, moisture and temperature, different channels should keep synchronous changes within the same sampling period. When the signal of a certain channel changes relatively delayed or earlier, it indicates that the time point of interference on that channel is different from that of other channels, which may be due to differences in electromagnetic wave propagation paths or signal coupling lag. By matching this time offset with the interference frequency segment in the signal anomaly comparison table, it can be confirmed whether the time misalignment is caused by interference. When the interference frequency segment matches the period of time misalignment, it indicates that the misalignment phenomenon is a direct result of interference. At this stage, the sampling boundary between different batches is also checked. When the time interval between the end and start of sampling of two batches is less than one sampling period, it is determined that there is a batch crossover phenomenon. This phenomenon can easily cause the residual signal of the previous batch to be superimposed on the initial data of the next batch, thereby introducing fluctuation data from other batches. The identification results of time misalignment and batch crossover are recorded in the time correlation table, providing an accurate interference range for the saturation analysis in the next stage.

[0029] Based on the time correlation table, saturation phenomenon analysis was performed on the amplitude changes of sensor output signals in each batch of test data. The saturation phenomenon is mainly manifested in the signal amplitude reaching the sensor's range limit for a short period of time and remaining unchanged, before returning to the normal range. This phenomenon often occurs during the period of concentrated interference energy, and its duration coincides with the interference duration range in the signal anomaly comparison table. During implementation, the amplitude change curve of each sensor channel was overlaid and compared with the interference time interval. When the signal continuously maintained the maximum or minimum output during the interference period, the time interval was marked as the saturation interval. By further comparing the saturation duration of each sensor channel, the degree of interference on different sensor channels and the energy distribution characteristics can be determined. For example, when the saturation duration of the color sensor signal is longer than that of the odor sensor signal, it indicates that the interference energy is more concentrated in the optical detection part, while the electrical signal is less affected. Through this combined amplitude and time analysis, the interference trigger point, i.e., the time position when the signal begins to be interfered with, and the interference duration range, i.e., the entire time interval before the signal returns to normal, can be accurately determined. This process not only reveals the strength and persistence of the interference effect, but also provides a data basis for risk classification.

[0030] After analyzing time misalignment, batch overlap, and saturation phenomena, all analysis results are summarized to generate an interference risk list. The interference risk list uses the interference event number as the primary index, recording the trigger time, duration, affected detection batches, affected sensor channels, interference intensity range, and data signal status during each interference event. To ensure the list's guiding role in subsequent detection schedule planning, the list also records the operational stage of the detection equipment when interference occurs, such as sampling, exposure, or signal conversion stages. This method allows for a one-to-one correspondence between interference risks and specific stages of the detection operation, providing directional reference for subsequent detection schedule adjustments. Simultaneously, the list includes information on recurring patterns of interference occurrences. When multiple interference events occur at similar time intervals in different batches, it indicates that the interference source has periodic characteristics, which can be avoided in future detections. The final interference risk list constitutes an important result document for vegetable detection data analysis, providing a systematic data basis for the detection schedule planning stage.

[0031] Based on the interference risk list, the detection process is rhythmically planned to determine the sampling sequence, exposure time, and power supply quiescent phase, forming an interference suppression scheme to guide the dynamic adjustment of subsequent detection processes; After obtaining the interference risk list, in order to effectively reduce the impact of interference and improve the stability of vegetable testing, the testing rhythm was planned in an orderly manner. This involved aligning the sampling sequence, exposure time, and power supply standby phase with the interference risk range, thus forming an interference suppression plan and providing a basis for subsequent dynamic adjustments. The specific implementation steps are as follows: Based on the interference trigger time and duration recorded in the interference risk list, the interference distribution is analyzed using a timeline. An interference time distribution map is created by arranging the time intervals corresponding to each interference event sequentially on a unified timeline, with the interference duration as the interval length and the interference intensity interval as the reference marker. Different colors or symbols in the interference time distribution map indicate different types of interference, such as interference caused by electromagnetic field changes, power fluctuations, and sensor line coupling. This graphical approach allows for intuitive identification of dense and relatively stable periods of interference, providing a foundation for optimizing the subsequent sampling sequence. At this stage, the focus is on identifying the concentrated areas of interference events through the interference time distribution map, thereby determining the time windows to be avoided during detection. The analysis results form an interference effect interval index table, providing input for the rhythm reconstruction stage.

[0032] After completing the interference time distribution analysis, the execution sequence of each sampling step in the vegetable testing process was reconstructed. The testing process typically includes multiple steps, such as sample identification, signal acquisition, data conversion and storage, and the execution time of each step has a fixed order within the testing cycle. To avoid overlap between testing steps and interference intervals, the interference effect interval index table was compared with the original timetable of the testing process, and the execution order and time span of each step were adjusted. When the interference frequency range identified in the interference risk list overlaps with the sampling cycle of a certain testing step, the execution time of that step was postponed to avoid the interference effect interval. When multiple testing steps overlap with interference intervals, the steps most affected were adjusted first, according to the principle of interference intensity interval from high to low. In this way, the testing rhythm can be redistributed while maintaining the stability of the total testing cycle. After the rhythm reconstruction is completed, a new sampling sequence table is generated, which clearly defines the start time and duration of each sampling step, providing a time framework for subsequent exposure time planning.

[0033] After obtaining the sampling sequence list, the exposure time and power quiescent phase are meticulously planned. In vegetable detection, exposure time is typically related to optical detection, while the power quiescent phase mainly corresponds to periods of strong electromagnetic interference. Based on the duration and frequency range of each interference event in the interference risk list, the exposure time is dynamically adjusted: when the interference range coincides with the exposure cycle, the exposure time is shortened or the exposure start point is delayed; when the interference frequency and exposure frequency have a harmonic or multiple harmonic relationship, the exposure time is appropriately delayed to avoid interference pulses; simultaneously, a power quiescent phase is set during periods of high interference intensity, i.e., power output changes are paused during this period to maintain stable power current and voltage, preventing external interference from being conducted to the detection circuit through the power supply path; during periods of weak interference, normal power supply is restored to ensure continuous operation of the detection device. Through the coordinated planning of exposure time and power quiescent phase, the impact of interference energy on the detection signal can be reduced through time control without altering the hardware structure of the detection equipment.

[0034] After planning the sampling sequence, exposure time, and power-off standby phase, the results of these time arrangements are integrated to form an interference suppression scheme. This scheme uses time as the main axis, recording in detail the execution sequence, duration, interference avoidance interval, and power control status of each detection step. By overlaying and aligning the sampling sequence table, exposure time table, and power-off standby phase table, a complete detection rhythm plan can be obtained, clearly defining the detection step and interference status corresponding to each moment. The interference suppression scheme also includes risk identification information to distinguish the interference risk level of each detection stage. For example, a low-risk stage corresponds to normal sampling, a medium-risk stage corresponds to delayed sampling, and a high-risk stage corresponds to silent sampling. In this way, the interference suppression scheme not only provides time control guidance for the detection process but also provides an operational basis for subsequent dynamic adjustments. During the vegetable detection process, sampling and exposure switching can be performed according to the time nodes set in the interference suppression scheme to ensure that the entire detection process is completed within the period of lowest interference risk.

[0035] During periods of high risk of interference, adaptive detection and adjustment are implemented based on the interference suppression scheme. Reverse silent sampling, intermittent sampling period and delayed sampling window are set to dynamically coordinate sampling time and illumination period, balance signal energy fluctuations, and stabilize the output signal in the vegetable detection process. After formulating the interference suppression scheme, in order to effectively control the stability of vegetable detection signals during periods of high interference risk, adaptive detection adjustments were implemented during the detection process to dynamically coordinate sampling actions with changes in interference energy. Through coordinated scheduling of time reconstruction and energy balance, the sampling time was dynamically matched with the illumination cycle, thereby suppressing signal fluctuations caused by electromagnetic interference. The specific implementation steps are as follows: Within the high-interference range identified in the interference risk list, a reverse silent sampling phase is set according to the time nodes of the interference suppression scheme. The core of this phase is to stagger the execution time of the detection sampling from the main action time of the interference signal. By analyzing the interference duration and peak time recorded in the interference risk list, the sampling trigger point is set within the attenuation range after the interference peak, so that the sampling process avoids the time point with the highest interference energy. During the reverse silent sampling phase, the signal acquisition channel of the detection device remains silent, only monitoring the ambient background noise level to establish a sampling baseline. After the interference energy enters the attenuation phase, the sampling channel is reactivated to capture the real detection signal in the later stage of the interference. This method, through reverse time offset, makes the sampling behavior inversely related to the change in interference energy, reducing the instantaneous impact of electromagnetic field abrupt changes on the sensing circuit, thereby maintaining the temporal continuity and waveform stability of the signal. The time setting of the reverse silent sampling is determined according to the interference duration range in the interference suppression scheme, and its duration is proportionally related to the interference attenuation rate to ensure a smooth transition between sampling trigger and interference termination.

[0036] After completing the setup of the reverse silent sampling phase, an intermittent sampling period is constructed based on the time distribution of different risk levels in the interference suppression scheme. The purpose of the intermittent sampling period is to achieve time avoidance of interference frequencies through periodic adjustment of the sampling interval. When the interference frequency overlaps with the sampling frequency, continuous sampling is prone to resonance effects, leading to signal amplification or fluctuations. To avoid this, the sampling period is adjusted to an unequal interval, so that the sampling times are distributed outside the interference fluctuation period. The planning of the intermittent sampling period is based on the interference frequency range recorded in the interference risk list. By staggering the sampling trigger time from the interference peak, it is ensured that there is at least one interference-free sampling point within an interference cycle. At the same time, during periods of frequent interference, the length of the sampling interval can be increased so that the next signal acquisition can be performed after the interference dissipates, thereby reducing the cumulative effect of interference. During the implementation of the intermittent sampling period, the sampling trigger signal and the illumination control signal are synchronized and coordinated to match the light source brightness with the sampling interval, ensuring that the sampling signal is performed during the illumination stable phase, thereby avoiding additional fluctuations introduced by illumination changes.

[0037] After planning the intermittent sampling cycle, a delayed sampling window is established for short-duration, high-intensity interference events recorded in the interference risk list. The delayed sampling window is used to delay the sampling action after the interference peak ends, fully utilizing the signal stability interval during the interference attenuation phase. The timing of this window is determined based on statistical data of interference duration and recovery time in the interference suppression scheme. By maintaining a fixed delay time after the interference event ends, the detected signal can recover to its natural fluctuation state before sampling is initiated. The duration of the delayed sampling window is layered according to the energy dissipation rate of different interference types. When the interference energy dissipates slowly, the delay time is relatively longer to ensure sampling is completed within the signal recovery interval. When the interference is a pulse-type transient event, the delay time can be shortened accordingly to maintain the overall continuity of the detection process. In actual implementation, the time of the delayed sampling window is correlated with the reverse silent sampling phase and the intermittent sampling cycle to ensure no overlap or gaps between time periods. The introduction of the delayed sampling window allows the detection process to automatically adapt to the changing rhythm of different interference types, ensuring the stability and repeatability of the sampling results.

[0038] After constructing the reverse silent sampling, intermittent sampling period, and delayed sampling window, the sampling time and illumination period are dynamically coordinated and executed. The core of this stage is to unify the sampling timing and illumination period within the same control framework, ensuring that illumination changes and sampling actions remain synchronized and balanced. During periods of high interference risk, the illumination period is automatically adjusted based on changes in the sampling rhythm. When the sampling interval is extended, the illumination period is lengthened accordingly to ensure that the light source brightness reaches a stable state before sampling; when the sampling interval is shortened, the illumination period is shortened synchronously to maintain consistency between sampling and peak illumination time. Simultaneously, a buffer period is established between illumination changes and sampling triggering. This method is used to eliminate short-term fluctuations caused during light intensity adjustment. Through this time-matching approach, the detection signal can be collected during periods of stable light energy, avoiding energy imbalances caused by changes in the light source. Furthermore, during dynamic coordination, the sampling rhythm is dynamically switched according to the risk level of the interference suppression scheme. When detection enters a low-risk period, continuous sampling mode is restored; when detection is in a high-risk period, intermittent and delayed sampling is maintained, thus forming a dynamic balance mechanism. Through the coordinated scheduling of light intensity and sampling time, the detection signal is ensured to maintain stable output throughout the entire interference-risk period, achieving comprehensive control from time avoidance to energy balance.

[0039] This invention introduces a multi-source acquisition and feature recognition mechanism for interference signals during vegetable testing, enabling the testing device to perceive interference in real time and dynamically adjust the sampling rhythm in complex electromagnetic environments. This effectively avoids signal abrupt changes caused by transient electromagnetic waves or power fluctuations. By establishing interference change curves and signal anomaly comparison tables, the invention achieves the identification of interference sources and the quantification of their impact range. This gives the testing device adaptive perception and avoidance capabilities during operation, improving the stability and reliability of vegetable quality testing data.

[0040] This invention achieves dynamic balance between sampling time, exposure phase, and illumination cycle within the interference range by planning and adaptively adjusting the detection rhythm based on an interference risk list. The detection device can actively adjust the sampling window and the power supply standby phase according to the interference intensity and duration, weakening the coupling effect of electromagnetic energy on the sensing signal and fundamentally suppressing signal drift and fluctuation accumulation. This method enables continuous and stable detection under interference environments, providing an accurate and repeatable data foundation for vegetable quality analysis.

[0041] This invention provides, for example Figure 2 The vegetable detection system shown includes an interference sensing module, a feature parsing module, an anomaly diagnosis module, a rhythm planning module, and an adaptive control module. The interference sensing module collects electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals at the vegetable testing site. Based on the collected results, it plots the interference change curve and records the time and position of the detection signal rise and return to zero, thus obtaining basic data for interference analysis. The feature analysis module identifies the source of interference based on the interference change curve, analyzes the propagation direction and coupling path of electromagnetic waves, extracts the interference frequency segment and duration segment, and generates a signal anomaly comparison table to provide a basis for retrospective analysis of detection data. The anomaly diagnosis module performs retrospective analysis on the data of each batch of vegetable testing based on the signal anomaly comparison table, finds time misalignment, batch overlap and sensor saturation of the test data, determines the trigger point and duration of the abnormal signal, and generates an interference risk list to provide a reference for the planning of the testing schedule. The rhythm planning module plans the rhythm of the detection process based on the interference risk list, determines the sampling sequence, exposure time, and power supply static phase, and forms an interference suppression scheme to guide the dynamic adjustment of the subsequent detection process; The adaptive control module implements adaptive detection and adjustment based on the interference suppression scheme during periods of interference risk. It sets reverse silent sampling, intermittent sampling period and delayed sampling window, dynamically coordinates sampling time with light cycle, balances signal energy fluctuations, and stabilizes the output signal during vegetable detection.

[0042] The vegetable detection method provided in this embodiment of the invention is implemented by the above-mentioned vegetable detection system. For details of the specific method and process of the vegetable detection system, please refer to the above-mentioned embodiment of the vegetable detection method, which will not be repeated here.

[0043] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for detecting vegetables, characterized in that, Includes the following steps: Electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals were collected at the vegetable testing site. Interference change curves were plotted based on the collected results, and the time and position of the detection signal rise and return to zero were recorded. The source of interference is identified based on the interference variation curve, the propagation direction and coupling path of electromagnetic waves are analyzed, the interference frequency range and duration range are extracted, and a signal anomaly comparison table is formed. Based on the signal anomaly comparison table, the data of each batch of vegetable testing were retrospectively analyzed to find time misalignment, batch overlap and sensor saturation of the test data, determine the trigger point and duration of abnormal signals, and generate an interference risk list. Based on the interference risk list, the detection process is rhythmically planned to determine the sampling sequence, exposure time, and power supply quiescent phase, thus forming an interference suppression scheme. During periods of high risk of interference, adaptive detection and adjustment are implemented based on the interference suppression scheme. Reverse silent sampling, intermittent sampling period and delayed sampling window are set to dynamically coordinate sampling time and illumination period, balance signal energy fluctuations, and stabilize the output signal during vegetable detection.

2. The method for detecting vegetables according to claim 1, characterized in that, The process of plotting the interference change curve is as follows: Signal acquisition units were set up at the vegetable testing site to collect electromagnetic change signals, sensor circuit noise signals and power fluctuation signals, respectively, and a reference potential benchmark was established to determine the sampling frequency, sampling period and signal channel response threshold. Based on the collected electromagnetic change signals, sensor line noise signals and power fluctuation signals, time sequence alignment and amplitude standardization are performed to unify the signals onto the same time coordinate axis to form a time series framework. Based on the time series framework, interference variation curves are plotted, and electromagnetic variation signals, sensor line noise signals, and power fluctuation signals are represented as independent curves, with interference concentration intervals marked. Based on the interference variation curve, record the time positions of the sudden rise and return to zero of the detection signal to form an interference time table and associate it with the real-time amplitude data of each signal to obtain the basic data for interference analysis.

3. The method for detecting vegetables according to claim 2, characterized in that, When plotting the interference variation curve, the order of interference propagation is determined by comparing the peaks, troughs, and abrupt changes of the electromagnetic change signal, the sensor line noise signal, and the power supply fluctuation signal. The time position is recorded synchronously when the signal amplitude suddenly increases and returns to zero, so as to distinguish the direction of interference energy propagation and the duration interval, and to establish an interference distribution mapping for subsequent interference analysis.

4. The method for detecting vegetables according to claim 2, characterized in that, The process of creating the signal anomaly checklist is as follows: Based on the interference change curve, the electromagnetic change signal, the sensor line noise signal and the power fluctuation signal are divided into intervals, and the layered comparison is carried out on the same time axis to determine the time period and starting order of the interference energy concentration. The synchronization of amplitude changes between electromagnetic change signals and sensing line noise signals is analyzed based on the time interval table of interference change curves in order to identify the propagation direction of electromagnetic waves and determine the coupling path. By combining the propagation direction and coupling path, the frequency components of the interference variation curve are divided into intervals, and the interference frequency segment and duration segment are extracted. Based on the analysis results of propagation direction, coupling path, frequency band, and duration, a signal anomaly comparison table is formed, recording the time and location of the interference, propagation direction, coupling path, and frequency band information.

5. The method for detecting vegetables according to claim 4, characterized in that, When creating the signal anomaly lookup table, the start time, end time, propagation direction, coupling path, and frequency range of the interference are correlated with the time series of electromagnetic change signals, sensor line noise signals, and power supply fluctuation signals, and the interference energy concentration interval is marked in the lookup table.

6. The method for detecting vegetables according to claim 4, characterized in that, The steps for generating the interference risk list are as follows: Using the signal anomaly comparison table as an index, time series matching was performed on the original test data of each batch of vegetables to establish a mapping relationship between batch data and interference periods; After completing the time series matching, time misalignment analysis and batch cross-analysis are performed on the detection data of each batch. The interference range is determined based on the interference frequency band and recorded in the time association table. Based on the time correlation table, saturation phenomenon analysis was performed on the amplitude changes of sensor output signals in each batch of test data to determine the interference trigger point and interference duration range; After completing the analysis of time misalignment, batch overlap and saturation phenomena, the analysis results are summarized to generate an interference risk list, recording the trigger time, duration, affected batch and sensor channel information of the interference events.

7. The method for detecting vegetables according to claim 6, characterized in that, The process of developing the interference suppression scheme is as follows: Using the signal anomaly comparison table as an index, time series matching was performed on the original test data of each batch of vegetables to establish the mapping relationship between batch data and interference periods and form a batch time index table; After completing the time series matching, time misalignment analysis and batch cross-analysis are performed on the detection data of each batch, and the identification results of time misalignment and batch cross-analysis are recorded in the time association table. Based on the time correlation table, the amplitude changes of sensor output signals in each batch of test data are analyzed for saturation phenomena to determine the interference trigger point and the duration of interference and mark the saturation interval. After completing the analysis of time misalignment, batch overlap and saturation phenomena, the analysis results are summarized to generate an interference risk list, which records the trigger time, duration, affected batches and sensor channel information of the interference events.

8. The method for detecting vegetables according to claim 7, characterized in that, Interference events in the interference risk list are arranged in order of trigger time and are classified and labeled according to the duration of interference and the affected range of the sensing channel. The interference risk list also records the detection stage information when interference occurs, and the periodic characteristics of the interference source are determined by the repetition interval of the interference events.

9. The method for detecting vegetables according to claim 7, characterized in that, During periods of high interference risk, adaptive detection and adjustment are implemented based on the interference suppression scheme. The following steps are taken to dynamically coordinate sampling time with illumination period by setting reverse silent sampling, intermittent sampling period, and delayed sampling window: Within the high-interference range identified in the interference risk list, a reverse silent sampling phase is set according to the interference suppression scheme. By setting the sampling trigger point within the attenuation range after the interference peak, sampling time avoidance is achieved. After completing the reverse silent sampling phase, an intermittent sampling period is constructed based on the risk level distribution in the interference suppression scheme. The sampling interval is adjusted to achieve time avoidance between the sampling time and the interference fluctuation period. After completing the intermittent sampling period, a delayed sampling window is established, and the sampling action is delayed after the interference peak ends in order to take advantage of the signal stability interval during the interference attenuation phase. After constructing the reverse silent sampling, intermittent sampling period and delayed sampling window, the sampling time and illumination period are dynamically coordinated and executed. The sampling and illumination are synchronized and balanced through time matching to stabilize the output of the detection signal.

10. A vegetable detection system, used to implement the vegetable detection method according to any one of claims 1-9, characterized in that, It includes an interference perception module, a feature parsing module, an anomaly diagnosis module, a rhythm planning module, and an adaptive control module: The interference sensing module collects electromagnetic change signals, sensor circuit noise signals, and power fluctuation signals at the vegetable testing site. Based on the collected results, it plots the interference change curve and records the time and position of the detection signal rise and return to zero. The feature analysis module identifies the source of interference based on the interference change curve, analyzes the propagation direction and coupling path of electromagnetic waves, extracts the interference frequency segment and duration segment, and generates a signal anomaly comparison table. The anomaly diagnosis module performs retrospective analysis on the data of each batch of vegetable testing based on the signal anomaly comparison table, finds time misalignment, batch overlap and sensor saturation of the test data, determines the trigger point and duration of the abnormal signal, and generates an interference risk list. The rhythm planning module plans the rhythm of the detection process based on the interference risk list, determines the sampling sequence, exposure time, and power supply static phase, and forms an interference suppression scheme. The adaptive control module implements adaptive detection and adjustment based on the interference suppression scheme during periods of interference risk. It sets reverse silent sampling, intermittent sampling period and delayed sampling window, dynamically coordinates sampling time with light cycle, balances signal energy fluctuations, and stabilizes the output signal during vegetable detection.

Images