Plant cat litter full life cycle traceability detection method and system

Abstract

This invention discloses a method and system for tracing and detecting the entire life cycle of plant-based cat litter, belonging to the field of tracing and detecting the entire life cycle of plant-based cat litter. The method includes the following steps: collecting continuous image brightness change information from the finished plant-based cat litter testing site under different low-light conditions; continuously recording the time and location of instantaneous bright spots appearing on the surface of the cat litter particles to generate an initial brightness distribution reference; based on the initial brightness distribution reference, continuously comparing the diffusion direction changes of bright spots in adjacent image frames, extracting the characteristic of brightness change from local concentration to spreading across the particle surface, and distinguishing between migrating light spot features and sudden reflective features within the initial brightness distribution reference range. This invention replaces single-frame judgment with continuous brightness evolution analysis, effectively distinguishing between fluorescence migration and instantaneous high-brightness interference, reducing the impact of low light and uneven illumination; and dynamically tightening the judgment conditions by combining the spatial distribution of migration signals, improving detection stability, data consistency, and product qualification rate.

Description

Technical Field

[0001] This invention relates to the field of plant-based cat litter full life cycle traceability and testing technology, specifically to plant-based cat litter full life cycle traceability and testing methods and systems. Background Technology

[0002] Plant-based cat litter full lifecycle traceability testing refers to a comprehensive testing system that tracks and records data on the entire process of plant-based cat litter, from raw material collection, formulation processing, packaging, storage, transportation, sales and distribution to final use and disposal. The specific process includes: collecting physicochemical indicators such as plant fiber, moisture, ash content, and heavy metal residues at the raw material stage; monitoring drying temperature, pressing density, and additive stability during processing; recording batches, storage environment, and logistics trajectory through QR codes or blockchain tags during packaging and transportation; testing changes in absorbency, deodorization performance, and decomposition rate during sales and use; and tracking biodegradability and environmental residue characteristics during disposal. Through multi-dimensional data fusion and intelligent identification, the system achieves full-process quality traceability and safety verification of cat litter products from source to end.

[0003] The existing technology has the following shortcomings: During the finished product testing of plant-based cat litter, trace fluorescent markers are prone to migration under specific conditions, especially during prolonged storage or under uneven lighting. Fluorescent molecules can slowly diffuse along the particle surface, causing localized brightness anomalies. Camera inspection systems, under low-light conditions, can easily capture these sporadic light spots and misinterpret them as foreign objects or metallic impurities, leading to the product being mistakenly identified as having a contamination risk. This type of problem is insidious and random, often occurring intermittently during batch testing, easily triggering false alarms or causing entire batches to be incorrectly rejected and returned. If not corrected promptly, traceability data will be skewed, affecting the reliability of quality analysis and product pass rates.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for tracing and detecting the entire life cycle of plant-based cat litter, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for tracing and detecting the entire life cycle of plant-based cat litter, comprising the following steps: Step 1: Collect continuous image brightness change information of the finished plant cat litter under different low light conditions at the testing site, continuously record the time and location of instantaneous bright spots on the surface of cat litter particles, and generate an initial brightness distribution reference for characterizing the fluorescence migration rhythm. Step 2: Based on the initial brightness distribution reference, continuously compare the changes in the diffusion direction of bright spots in adjacent image frames, extract the change features of brightness from local concentration to spread to the particle surface, and distinguish between migrating spot features and sudden reflection features within the initial brightness distribution reference range; Step 3: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time. The brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. Step 4: Based on the brightness evolution sequence, perform constraint analysis on the brightness growth rate of suspected impurity areas in the image, classify light spots that show slow changes in the brightness evolution sequence as migration signal sources, and remove the brightness change data corresponding to the migration signal source from the foreign object determination path; Step 5: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and the judgment is only performed in the bright areas with abrupt changes in the brightness evolution sequence, so as to reduce the probability of false alarms and ensure the accuracy of the detection results.

[0007] Preferably, the steps for collecting continuous image brightness change information under different low-light conditions at the plant-based cat litter finished product testing site include: A low-light acquisition environment was constructed at the finished product inspection site. A stable light source and a fixed acquisition angle were set up in the inspection area. Multiple lighting states were established by continuously adjusting the light intensity and exposure duration, and the sampling frequency was kept constant to form a time-continuous image sequence. The obtained continuous image sequence is compared frame by frame. The time, location coordinates and brightness value of the bright spots are recorded according to the difference in brightness changes between adjacent image frames, and stored in chronological order to form bright spot distribution data. By using bright spot time and location records to generate continuous brightness variation curves, brightness data within the same time period are superimposed to form a spatial distribution map, and integrated into an initial brightness distribution reference with three-dimensional attributes of time, brightness and space. During the detection process, based on the initial brightness distribution reference, the brightness information of the real-time image is compared in terms of time and position. The consistency between the brightness change and the diffusion trajectory is used to determine whether the light spot belongs to the fluorescence migration process.

[0008] Preferably, the step of continuously comparing the changes in the diffusion direction of bright spots in adjacent image frames based on the initial brightness distribution reference includes: Under the initial brightness distribution reference condition, adjacent image frames acquired consecutively are arranged in time order. The spatial position and brightness range of the bright spot at different time points are determined with the brightness peak point as the center. The trend of the bright spot spreading outward from the central region is identified by comparing the changes in brightness intensity. The spatial evolution path of diffusion is determined by the change in brightness profile of bright spots in adjacent image frames. The relative positions of brightness enhancement and attenuation regions are analyzed and the diffusion direction of bright spots is determined by using the initial brightness distribution as a reference. The brightness distribution maps of adjacent image frames in the time series are superimposed on the same spatial coordinates. By tracking the distribution trend of brightness from the center to the edge, the change features of brightness from local concentration to spread to the particle surface are extracted and a brightness change curve is generated. By combining the range of the initial brightness distribution reference, the characteristics of bright spot change are matched with the characteristics of brightness evolution. Based on the temporal continuity and spatial extension of brightness change, the characteristics of migrating spot and sudden reflection are distinguished.

[0009] Preferably, when distinguishing between migrating spot features and sudden reflection features, the temporal continuity of brightness change is judged by the stable increase of the brightness difference between adjacent image frames, and the spatial extensibility is judged by the consistent direction of the brightness enhancement area in consecutive frames. The spot is determined to be a migrating spot feature only when both conditions are met.

[0010] Preferably, the step of time-rearranging the exposure response sequence of the detection image based on the diffusion change characteristics corresponding to the migrating light spot includes: The continuous image frames in the detection screen are indexed and organized by time. The exposure response parameters of each frame are matched with the brightness records and sorted according to the acquisition time, exposure order and brightness change rate. The continuous diffusion area in the initial brightness distribution reference is used as the fixed reference for the time series. Based on the diffusion change characteristics of the migrating light spot, the exposure response sequence is rearranged in time, and the brightness changes that continue to spread in adjacent image frames are arranged on the same brightness evolution path, and the image frames with sudden brightness increases are identified separately as instantaneous reflection signals. The rearranged image frames were continuously compared, and the continuous diffusion trend was determined by comparing the spatial distribution of brightness changes in adjacent frames. The brightness change curves within the brightness evolution range were extracted as the evolution trajectory of the migrating light spot. The filtered brightness change data are recombined into a continuous and consistent brightness evolution sequence, and the time and spatial coordinates of the initial brightness distribution reference are kept consistent during the generation process to obtain a complete brightness evolution record.

[0011] Preferably, during the time rearrangement process, the continuous diffusion region in the initial brightness distribution reference is used as a fixed benchmark, and the time interval and spatial position of brightness changes in consecutive image frames are synchronously aligned. By removing instantaneous reflection signals and retaining only brightness data with a continuous diffusion trend, the generated brightness evolution sequence remains continuous and consistent in time and space.

[0012] Preferably, the step of constraining the brightness growth rate of suspected impurity regions in an image based on the brightness evolution sequence includes: The brightness change areas in the detection image are identified and calibrated. By comparing the brightness evolution curves in consecutive image frames frame by frame, the areas with large brightness changes are identified and the growth curve of each brightness change point is established to form a brightness growth record. Constraint analysis was performed on the temporal variation characteristics of brightness growth records in the brightness evolution sequence. By comparing continuous time periods, the changing trend of brightness growth rate was identified, and stable brightness growth regions and regions with prominent changes were marked in the brightness evolution sequence to distinguish the source of migration signals from foreign object signals. The identified migration signal source areas are classified and marked. Light spots with temporal extension and spatial traceability of brightness changes are recorded as migration signal sources, and the time starting point, initial brightness value, and spatial coordinates are kept consistent with the brightness evolution sequence. The foreign object detection path is updated by comparing the migration signal record with the brightness evolution sequence to simultaneously exclude the brightness change data corresponding to the migration signal source in both time and space dimensions, so as to preserve the true foreign object signal distribution.

[0013] Preferably, in the constraint analysis of brightness growth rate, by comparing the brightness change amplitude at consecutive time points in the brightness evolution sequence, a time continuity threshold and a spatial extension range are set. When the brightness change curve is continuous in time and the spatial distribution is consistent with the source of the migration signal, the corresponding brightness data is automatically included in the migration signal record to ensure that only the bright spot information with abrupt change characteristics is retained in the foreign object determination path.

[0014] Preferably, the step of dynamically tightening the brightness triggering conditions in the finished product detection and judgment process based on the spatial distribution results of the migration signal source includes: The detection image is divided into spatial regions. The brightness change region is divided into different brightness levels according to the spatial characteristics of the migration signal distribution, using the brightness evolution sequence as the time reference. Time evolution curves and brightness change rate data are established in each spatial unit. The brightness triggering conditions in each spatial region are reset, strict triggering conditions are set in the region dominated by the migration signal, and sensitivity is maintained in the region of high brightness change to achieve dynamic adjustment of the zones and balance the accuracy of the detection results. The detection and judgment process is executed with dual synchronization of time and space. By combining the time nodes in the brightness evolution sequence with the spatial distribution results, the corresponding triggering conditions are selected at different stages and the brightness changes are dynamically responded to. In the dynamic tightening judgment process, the bright areas with abrupt changes in brightness evolution sequence are analyzed in detail. By comparing the brightness changes at different time points, sudden reflection characteristics are identified and areas that meet the conditions are marked as foreign object judgment targets.

[0015] The plant-based cat litter full life cycle traceability and detection system includes a brightness acquisition and recording module, a light spot diffusion analysis module, a brightness sequence construction module, a signal constraint recognition module, and a dynamic judgment and control module; Brightness Acquisition and Recording Module: Acquires continuous image brightness change information of the finished plant cat litter under different low light conditions at the testing site, continuously records the time and location of instantaneous bright spots on the surface of cat litter particles, and generates an initial brightness distribution reference; Light spot diffusion analysis module: Based on the initial brightness distribution reference, the diffusion direction changes of bright spots in adjacent image frames are continuously compared, and the change characteristics of brightness from local concentration to spread to the particle surface are extracted. Within the range of the initial brightness distribution reference, the characteristics of migrating light spots and sudden reflections are distinguished. Brightness sequence construction module: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time, and the brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. Signal constraint recognition module: Based on the brightness evolution sequence, it performs constraint analysis on the brightness growth rate of suspected impurity areas in the image, and classifies light spots that show slow changes in the brightness evolution sequence as migration signal sources; Dynamic Judgment Control Module: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and judgment is only performed in the bright areas that show abrupt changes in the brightness evolution sequence.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention introduces a time-continuous brightness evolution analysis mechanism during the finished product inspection stage, transforming brightness change judgment from single-frame determination to multi-temporal correlation determination. This enables the inspection process to accurately distinguish between slow brightness diffusion caused by fluorescence migration and instantaneous high-brightness changes caused by external factors. By constructing an initial brightness distribution reference and forming a continuous and consistent brightness evolution sequence, the inspection results no longer depend on brightness anomalies at a single moment. This effectively reduces the interference of low-light environments and uneven illumination on the judgment results, ensuring a stable and consistent judgment basis for finished product inspection even in complex environments, and significantly improving the reliability of the overall inspection results.

[0017] This invention introduces a dynamic tightening mechanism based on the spatial distribution of migration signals into the detection and judgment process. This allows the brightness triggering conditions to be adjusted according to the brightness evolution characteristics, performing judgments only in high-brightness areas with abrupt changes. By removing brightness change data corresponding to migration signals from the foreign matter judgment path, it avoids misidentifying fluorescence migration as a contamination risk, reducing the probability of false alarms at the source. Simultaneously, this processing method ensures the continuity and consistency of traceability detection data, effectively preventing erroneous judgment information from entering the traceability chain, thereby improving product qualification rates and enhancing the stability and reliability of lifecycle quality analysis results. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of the method for tracing and detecting the entire life cycle of plant-based cat litter according to the present invention.

[0020] Figure 2 This is a schematic diagram of the modules of the plant-based cat litter full life cycle traceability and detection system of the present invention. Detailed Implementation

[0021] 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, they 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.

[0022] This invention provides, for example Figure 1 The method for tracing and testing the entire life cycle of plant-based cat litter, as shown, includes the following steps: Step 1: Collect continuous image brightness change information of the finished plant cat litter under different low light conditions at the testing site, continuously record the time and location of instantaneous bright spots on the surface of cat litter particles, and generate an initial brightness distribution reference for characterizing the fluorescence migration rhythm. The specific implementation method for this step is as follows: A low-light acquisition environment was constructed at the finished product testing site, with a stable light source and fixed acquisition angle set up within the testing area. By continuously adjusting the light intensity and exposure duration, multiple different lighting conditions were established, ensuring that the surface brightness of the cat litter granules was completely recorded across several ranges, from relatively dark to near-standard testing brightness. During acquisition, the sampling frequency was kept constant, ensuring that each image frame corresponded to a specific time point. In this way, a temporally continuous image sequence could be formed within the same testing area. This image sequence recorded the brightness distribution of the cat litter granule surface under different low-light conditions, thus ensuring the capture of brightness fluctuations caused by the migration of fluorescent markers on the granule surface. All acquired images were numbered chronologically and saved one-to-one with the corresponding light intensity parameters, forming a set of temporally continuous, traceable raw data on brightness changes under different lighting conditions.

[0023] After obtaining a complete continuous image sequence, the brightness distribution of each frame is compared frame by frame to determine the specific location and time of brightness changes on the surface of the cat litter particles. By analyzing the differences in brightness change areas in adjacent frames, the coordinate positions of bright spots in the images are determined, and the time of brightness surges is recorded. Each bright spot record contains three basic attributes: appearance time, location coordinates, and brightness value. To ensure the continuity of the records, a fixed sampling interval is maintained between adjacent time points, allowing the changes in bright spots on the time axis to form a continuous brightness curve. All bright spot records are stored in chronological order and superimposed in the same spatial coordinate system to form bright spot distribution data containing both time and location information. This data reflects the frequency and distribution patterns of bright spots during the detection process, providing complete basic information for the subsequent generation of brightness change curves.

[0024] Subsequently, based on the acquired bright spot time and location records, a continuous brightness variation curve was constructed, forming an initial brightness distribution reference for characterizing the fluorescence migration rhythm. Bright spot records at consecutive time points were arranged sequentially according to the acquisition order, and brightness values ​​were matched with time information. By connecting adjacent brightness data points on the same time axis, a continuous curve showing brightness changes over time was obtained. Data from bright spot locations within the same time period were then superimposed to form a dynamic variation map of brightness in spatial distribution. Through combined processing of the time and spatial dimensions, the trajectory of bright spot diffusion on the cat litter particle surface was obtained. The temporal continuity of the bright spot reflects the continuous process of fluorescence migration, while the spatial variation of brightness reflects the migration direction and range. Integrating all spatiotemporally corresponding brightness data into a dataset with three-dimensional attributes of time, brightness, and space constitutes the initial brightness distribution reference. This reference clearly expresses the process of fluorescence migration from concentration to dispersion and provides a unified benchmark for subsequently determining whether bright spots are migratory or sudden.

[0025] After establishing the initial brightness distribution reference, it is used as the basic reference framework in subsequent detection stages to ensure temporal and spatial consistency. In subsequent continuous detection, whenever new image data is acquired, its real-time brightness information is compared with the initial brightness distribution reference. Temporal matching determines whether the current bright spot's appearance time matches the migration rhythm recorded in the reference; positional matching determines whether the bright spot's diffusion direction matches the existing diffusion trajectory in the reference. If the comparison results show that the brightness change of the bright spot is consistent with the gradual process described in the initial reference, the bright spot is identified as a natural brightness evolution during fluorescence migration; if the comparison results show that the brightness change rate or spatial range exceeds the change range of the initial reference, it is classified as a burst reflection feature for further analysis in subsequent stages. By continuously referencing the initial brightness distribution reference throughout the entire detection process, each detection has stable benchmark conditions, thus forming a temporally continuous and spatially controllable brightness analysis system. This operation process ensures the consistency of brightness data between acquisition, recording, and comparison stages, enabling the complete reconstruction and tracking of the temporal and spatial characteristics of the fluorescence migration process.

[0026] By continuously implementing the above steps, the temporal changes and spatial distribution of the surface brightness of cat litter particles can be completely recorded under low-light conditions at the testing site of the plant-based cat litter product. This allows for the construction of an initial brightness distribution reference that characterizes the fluorescence migration rhythm. This implementation method, through specific operations such as stable illumination control, continuous temporal acquisition, frame-by-frame brightness comparison, and reference comparison application, establishes a continuous technical workflow encompassing acquisition, recording, analysis, and reference application. This enables the systematic identification and tracking of brightness changes in the plant-based cat litter product during the testing phase, providing a sufficient data foundation and stable analytical basis for distinguishing fluorescence migration characteristics and correcting misjudgments.

[0027] Step 2: Based on the initial brightness distribution reference, continuously compare the changes in the diffusion direction of bright spots in adjacent image frames, extract the change features of brightness from local concentration to spread to the particle surface, and distinguish between migrating spot features and sudden reflection features within the initial brightness distribution reference range; The specific implementation method for this step is as follows: Under the established initial brightness distribution reference conditions, adjacent image frames acquired consecutively are arranged chronologically to ensure that the brightness changes of the same particle region at different time points can be matched one-to-one. To accurately identify brightness diffusion characteristics, in each pair of adjacent frames, the spatial position and brightness range of the bright spot in the first and second frames are determined, centered on the brightness peak point in the initial brightness distribution reference. By comparing the brightness intensity changes of the same bright spot in adjacent frames, it can be identified whether the brightness diffuses outward from the central region. When it is found that the brightness gradually increases in the peripheral region while the brightness in the central region tends to be uniform, it indicates that the light spot has a trend of spreading outward from local concentration within that time interval. Through this continuous comparison method, a preliminary description of the diffusion direction of the bright spot in the time dimension can be formed, providing basic data support for subsequent directional change analysis.

[0028] After obtaining the initial diffusion trend of bright spots in adjacent image frames, the spatial evolution path of diffusion is determined by comparing the directional relationship of brightness changes in each frame. Using the initial brightness distribution as a reference, the changes in the brightness contour of the same bright spot in temporally adjacent images are observed, and the relative spatial positions of the brightness enhancement region and the attenuation region are analyzed. When the brightness enhancement portion gradually extends along a specific direction on the particle surface, and this direction remains consistent across consecutive time points, it can be determined that the diffusion direction of the light spot has stable migration properties. In this process, by comparing consecutive time points, it can be determined whether the bright spot spreads towards the particle surface in a certain direction or exhibits a random distribution within a short period of time. If the brightness distribution of the bright spot shows a gradual transition from the center to the periphery at multiple time points, it can be confirmed that it belongs to brightness diffusion behavior. This process, through the analysis of continuous time and space correspondence, clearly depicts the trajectory of the direction change of bright spot diffusion and provides a complete spatiotemporal data basis for subsequent feature extraction.

[0029] After obtaining the diffusion direction characteristics of the bright spot, the variation characteristics of brightness spreading from local concentration to the particle surface are extracted based on the initial brightness distribution reference. Brightness distribution maps of adjacent frames in the time series are superimposed under the same spatial coordinates. By comprehensively comparing the brightness peaks and their surrounding brightness changes, the amplitude, range, and direction of brightness changes are determined. By tracking the distribution trend of brightness from high to low within the same spatial region, the spreading pattern of the bright spot can be depicted. When the brightness gradually increases at the periphery while remaining stable in the center, it can be considered that the bright spot exhibits a diffusion characteristic from the center to the edge at this stage. By comparing consecutive frames, the time process of the bright spot transitioning from a local concentration area to a wider surface can be obtained. The key to this step is to combine the initial brightness distribution reference to unify the brightness changes in consecutive images under the same spatiotemporal benchmark for comparison, thereby fully revealing the spreading pattern of brightness in time and space and forming a brightness change curve that reflects the dynamic characteristics of fluorescence migration.

[0030] After extracting the brightness variation characteristics from local concentration to particle surface spread, the characteristics of different types of light spots are distinguished by combining the range of the initial brightness distribution reference. By matching the brightness change characteristics obtained in the current image sequence with the brightness evolution characteristics in the initial brightness distribution reference, it can be determined whether the light spot belongs to the migrating or abrupt type. When the brightness change is continuous in time, shows a pattern of expansion from the center to the outside in space, and is consistent with the diffusion trend recorded in the initial brightness distribution reference, the light spot can be classified as a migrating light spot. When the brightness change suddenly increases in a short period of time and lacks spatial diffusion continuity, the light spot is classified as an abrupt reflective feature. In this way, different types of bright spots in the detection image can be classified without changing the acquisition conditions. To ensure the continuity of classification, the initial brightness distribution reference is used stably in subsequent detections, so that each batch of detections can complete the identification and differentiation of light spot features under a unified reference framework. This comparative analysis process can effectively distinguish between the continuous brightness diffusion caused by fluorescence migration and the instantaneous brightness change caused by external reflection, thus providing a reliable prerequisite for the subsequent brightness rearrangement and judgment of the detection image.

[0031] Through the above steps, based on the initial brightness distribution reference, the diffusion direction changes of bright spots in adjacent image frames can be continuously compared, extracting the characteristic of brightness change from local concentration to spreading across the surface of cat litter particles. Within this reference range, the characteristics of migrating light spots and sudden reflections can be accurately distinguished. This implementation method, through specific operations such as temporal arrangement, spatial direction comparison, brightness change extraction, and feature type differentiation, forms a continuous and traceable brightness change analysis process. This allows fluorescence migration behavior and occasional reflection phenomena to be effectively distinguished during the detection stage, providing complete basic data and stable judgment criteria for the subsequent establishment of brightness evolution sequences.

[0032] Step 3: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time. The brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. The specific implementation method for this step is as follows: After obtaining the diffusion change characteristics corresponding to the migrating light spot, all consecutive image frames in the detection scene are time-indexed and organized, and the exposure response parameters of each frame are mapped one-to-one with the brightness records. In this way, a sequential chain of brightness acquisition can be established on the time axis. To ensure the accuracy of subsequent reordering, all image frames are comprehensively sorted according to acquisition time, exposure response order, and brightness change rate, so that each brightness change process can be traced back to a specific time node. In this process, the continuous diffusion area recorded in the initial brightness distribution reference is used as a fixed benchmark in the time series. With this benchmark as the center, the brightness change information acquired subsequently is divided into time periods. Through this division, the continuous diffusion stage and the instantaneous reflection stage of brightness can be clearly distinguished in the time dimension, establishing a complete time framework for subsequent time reordering operations.

[0033] After the time index is organized, the exposure response sequence of each brightness change process is rearranged temporally. This process is based on the diffusion characteristics of migrating light spots, treating brightness changes that continuously diffuse in adjacent image frames as a continuation of the same temporal evolution process. During the rearrangement, the continuous diffusion trend in the initial brightness distribution reference is used as the reference line for the time series. Image frames showing continuous brightness growth are rearranged onto the same brightness evolution path, while image frames with sudden increases in brightness are separated from this path and identified separately as instantaneous reflection signals. In this way, the continuous diffusion process can be separated from random reflection interference, making the temporal sequence of brightness changes exhibit coherent evolutionary characteristics. The rearranged time series can reflect the brightness evolution trajectory of migrating light spots in different time periods, providing a foundation for constructing a stable brightness evolution sequence.

[0034] After time rearrangement, the rearranged image frames are continuously compared to identify and confirm brightness changes that exhibit a continuous diffusion trend in the initial brightness distribution reference. By comparing the brightness changes of adjacent frames in the rearranged sequence one by one, it can be determined whether the spatial distribution of brightness maintains a diffusion state. When the brightness gradually transitions from the center to the outside in multiple consecutive frames, and the brightness enhancement area shows an extending relationship in space, it indicates that the brightness change belongs to a continuous diffusion trend. At this time, the continuous frame group is defined as the brightness evolution interval of the migrating spot, and the corresponding brightness change curve is extracted as the evolution trajectory of the spot in the entire time series. At the same time, frame groups with sudden brightness changes and no spatial continuity are marked as instantaneous reflection signals and removed from the brightness evolution sequence. Through this frame-by-frame comparison and separation process, the continuity of brightness changes can be clearly defined, thus retaining only spot information that conforms to the continuous diffusion law in the final brightness evolution sequence.

[0035] After separating the continuous diffusion trend from the instantaneous reflection signal, the selected brightness change data are recombined to form a continuous and consistent brightness evolution sequence. This brightness evolution sequence consists of multiple sets of image frames that have undergone temporal rearrangement and feature separation. Each frame is arranged according to temporal order and spatial diffusion relationship, so that the entire sequence presents a continuous change process from brightness concentration to brightness equalization on the time axis. To ensure the overall consistency of the brightness evolution sequence, the temporal and spatial coordinates in the initial brightness distribution reference are kept unchanged during the sequence generation process, and each set of brightness change curves is aligned with the original reference, making the brightness change continuous in time and traceable in space. In this way, a brightness evolution record that is continuous in both the temporal and spatial dimensions can be obtained, which fully reflects the diffusion rhythm and brightness evolution law of the migrating spot during the detection process. This brightness evolution sequence can provide basic data for subsequent brightness constraint analysis, while ensuring that in the subsequent judgment stage, all brightness changes are based on real migration behavior and are not affected by instantaneous reflection interference.

[0036] This implementation method uses continuous operations such as time indexing, exposure response rearrangement, continuous diffusion identification, and brightness sequence recombination to uniformly express the temporal and spatial characteristics of brightness changes, thereby achieving accurate tracking of the entire process of brightness evolution of migrating spot and providing a stable analytical basis and continuous data reference for subsequent brightness constraint and judgment steps.

[0037] Step 4: Based on the brightness evolution sequence, perform constraint analysis on the brightness growth rate of suspected impurity areas in the image, classify light spots that show slow changes in the brightness evolution sequence as migration signal sources, and remove the brightness change data corresponding to the migration signal source from the foreign object determination path; The specific implementation method for this step is as follows: After obtaining the complete brightness evolution sequence, all brightness change regions appearing in the detected image are identified and calibrated. By comparing the brightness evolution curves frame by frame in consecutive image frames, regions with large brightness changes within each time period are identified and preliminarily defined as suspected impurity regions. During the identification process, using the brightness evolution sequence as a time reference, the magnitude of brightness intensity changes in adjacent frames is quantified, and a growth curve for each brightness change point is established on the time axis. In this way, brightness changes can be extracted from the time series, forming a brightness growth record with temporal continuity. Each brightness change point contains time coordinates, brightness values, and spatial location information, thus enabling precise tracking of the brightness growth rate in subsequent analysis. The key to this step is to fully include all regions that might be mistaken for foreign object reflections within the analysis scope and to provide basic data support for subsequent brightness rate constraints.

[0038] After obtaining the brightness growth record of suspected impurity areas, a constraint analysis was performed on its temporal variation characteristics within the brightness evolution sequence. By comparing continuous time periods, the trend of brightness growth rate was identified, and regions with stable brightness growth and regions with significant changes were marked in the brightness evolution sequence. If the brightness change of a certain region shows a slow increase over multiple consecutive time points, and the brightness change curve remains consistent in the time dimension, then this region can be determined to be the source of a migrating signal. Conversely, if a region experiences a sudden increase in brightness within a very short period of time, and then quickly returns to the background brightness level, then this region is more likely to be a foreign object signal caused by transient reflection. In this process, all judgments are based on the brightness evolution sequence as the core basis, and the correspondence between the brightness growth rate and the source of the light spot is established by comparing continuous temporal changes. This constraint analysis can distinguish between continuously spreading migrating light spots and sudden reflection foreign object signals in the time dimension, thus laying the foundation for subsequent removal processing.

[0039] After completing the constraint analysis of the brightness growth rate, the identified migration signal source regions were classified and labeled. By tracking the light spots with slowly changing characteristics in the brightness evolution sequence over time, it was found that these spots maintained a stable diffusion pattern throughout the detection process, and the brightness growth curve showed a smooth and continuous upward trend. These spots were marked as migration signal sources, and the corresponding brightness change data were stored separately in the migration signal record. In this process, the definition of the migration signal source relies on the temporal continuity and spatial continuity of the brightness evolution sequence, that is, the brightness change of the light spot has extension in time and is traceable in space. To maintain consistency in subsequent analysis, the time starting point, initial brightness value, and spatial coordinates of the migration signal source were kept consistent with the original brightness evolution sequence, thereby ensuring that the corresponding data could be accurately removed in the subsequent foreign matter detection path. This step provides a stable reference basis for the identification of migration signals, preventing the brightness diffusion caused by fluorescence migration from being mistaken for a contamination signal.

[0040] After calibrating the source of the migrating signal, the foreign object detection path is updated by removing brightness change data corresponding to the source of the migrating signal from the detection path. This operation performs simultaneous exclusion in both time and space by comparing the migrating signal record with the original brightness evolution sequence. Specifically, in the time dimension, the time interval corresponding to the source of the migrating signal is removed from the foreign object detection sequence; in the spatial dimension, brightness change data overlapping or adjacent to the location of the migrating signal source is masked from the detection screen. In this way, the foreign object detection path retains only bright spot data that exhibits rapid abrupt changes and lacks temporal continuity in the brightness evolution sequence. The detection path after this removal process can more accurately reflect the true distribution of foreign object signals while avoiding false alarms caused by interference from migrating signals. This operation not only ensures the consistency of brightness change data in time and space but also makes subsequent detection and judgment more accurate and reliable.

[0041] By implementing the above steps, the brightness growth rate of suspected impurity regions in the image can be constrained based on the brightness evolution sequence. Spots exhibiting slow changes in brightness evolution are classified as migration signal sources, and brightness change data corresponding to these migration signal sources are removed from the foreign object detection path. Through a series of techniques, including continuous time tracking, brightness growth rate constraint, migration signal classification, and detection path removal, effective differentiation between fluorescence migration signals and transient reflection signals is achieved. This ensures that the entire detection process possesses temporal continuity and spatial controllability at the brightness analysis level, thereby improving the stability and accuracy of the detection data and providing a reliable data foundation and logical support for the subsequent dynamic tightening of brightness triggering conditions.

[0042] Step 5: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and the judgment is only performed in the bright areas with abrupt changes in the brightness evolution sequence, so as to reduce the probability of false alarms and ensure the accuracy of the detection results. The specific implementation method for this step is as follows: After identifying the source of the migration signal and obtaining its spatial distribution, the entire detection screen is spatially divided. Using the brightness evolution sequence as a time reference, all brightness change areas in the detection screen are divided into different brightness levels according to the spatial characteristics of the migration signal distribution. This method clearly identifies which areas belong to the migration signal influence zone and which areas maintain brightness abrupt change characteristics within a spatial range. Each spatial unit contains its corresponding time evolution curve and brightness change rate data, enabling synchronous correlation between space and time in subsequent judgments. To ensure the stability of the spatial distribution division, the center location, diffusion range, and temporal continuity of the migration signal source are kept consistent with the initial brightness distribution reference, ensuring that all region divisions are based on actual brightness change behavior. Through this operation, the entire detection screen is subdivided into migration signal-dominated areas and high-brightness abrupt change areas, providing a clear spatial basis for the dynamic adjustment of subsequent brightness triggering conditions.

[0043] After obtaining the spatial division results of the migration signal source, the brightness triggering conditions in each region are reset. In the region dominated by the migration signal, since the brightness change mainly originates from the fluorescence migration process, the triggering condition is set to a stricter constraint range, i.e., detection is only triggered when the brightness change rate exceeds the normal change range of the migration signal. In the high-brightness abrupt change region unaffected by the migration signal, a higher sensitivity is maintained, enabling the system to respond quickly to real foreign object reflections or contaminated light spots. Through this differentiated setting, the detection process achieves a "dynamic tightening" state overall. That is, in the region where the migration signal exists, the detection threshold automatically increases to avoid triggering false alarms due to slow brightness changes; while in the non-migration region, the detection threshold remains stable to ensure that sudden light spots can be identified in a timely manner. This process, by combining the spatial characteristics of the migration signal with the brightness change rate, achieves dynamic adjustment of the judgment conditions by region, maintaining a balance between sensitivity and accuracy in the detection results.

[0044] After dynamically setting the brightness trigger conditions, the detection and judgment process is synchronized in both time and space. By combining the time nodes in the brightness evolution sequence with the spatial distribution results, each judgment moment corresponds to a specific spatial region and brightness characteristic. During judgment, the system first identifies which stage in the brightness evolution sequence the current time point is in, and then selects the corresponding trigger condition based on the spatial brightness distribution characteristics of that stage. If the detection area at the current moment is within the diffusion range of the migrating signal source, the judgment is made according to the tightened trigger condition to avoid misjudgment caused by fluorescence diffusion; if the current detection area is within the brightness abrupt change range, the judgment is made using the conventional trigger condition to ensure that the foreign object spot can be accurately identified. Through this synchronous control method, the judgment process can dynamically respond to brightness changes in both time and space, achieving real-time matching and flexible adjustment of brightness trigger conditions, thereby making the detection process more stable and accurate.

[0045] During the dynamic tightening process, high-brightness regions exhibiting abrupt changes in the brightness evolution sequence are analyzed and recorded. By continuously comparing the brightness changes of these regions at different time points, it can be confirmed whether they possess sudden reflection characteristics. If the brightness abrupt change occurs only within a single time period and the rate of change is much higher than the diffusion rate of the migration signal, the region is marked as a foreign object detection target and output in the subsequent quality inspection report. If the brightness change is continuous in time and overlaps with the spatial range of the migration signal, the data corresponding to that region is automatically removed and not included in the foreign object detection path. In this way, high-brightness regions can be accurately screened during the detection stage, focusing the detection process only on actual contamination or reflection anomalies, thereby effectively reducing the frequency of false alarms. After this processing, the entire finished product inspection process maintains sensitive detection capabilities while avoiding brightness interference caused by fluorescence migration, ensuring the reliability and consistency of the detection results.

[0046] Through the above steps, the brightness triggering conditions in the finished product testing and judgment process are dynamically tightened, and judgment is only performed in high-brightness areas that exhibit abrupt changes in the brightness evolution sequence. This effectively reduces the probability of false alarms and ensures the accuracy of the test results. By employing techniques such as spatial division, dynamic setting, synchronous control, and key analysis, the spatial distribution patterns of migration signals are organically combined with the brightness triggering conditions, forming a dynamic judgment mechanism that can adapt to brightness changes in real time. This ensures that the finished plant litter testing process maintains high precision and stability even under complex lighting and fluorescence migration conditions.

[0047] Beneficial effect 1

[0048] This invention introduces a time-continuous brightness evolution analysis mechanism during the finished product inspection stage, transforming brightness change judgment from single-frame determination to multi-temporal correlation determination. This enables the inspection process to accurately distinguish between slow brightness diffusion caused by fluorescence migration and instantaneous high-brightness changes caused by external factors. By constructing an initial brightness distribution reference and forming a continuous and consistent brightness evolution sequence, the inspection results no longer depend on brightness anomalies at a single moment. This effectively reduces the interference of low-light environments and uneven illumination on the judgment results, ensuring a stable and consistent judgment basis for finished product inspection even in complex environments, and significantly improving the reliability of the overall inspection results.

[0049] Benefit 2

[0050] This invention introduces a dynamic tightening mechanism based on the spatial distribution of migration signals into the detection and judgment process. This allows the brightness triggering conditions to be adjusted according to the brightness evolution characteristics, performing judgments only in high-brightness areas with abrupt changes. By removing brightness change data corresponding to migration signals from the foreign matter judgment path, it avoids misidentifying fluorescence migration as a contamination risk, reducing the probability of false alarms at the source. Simultaneously, this processing method ensures the continuity and consistency of traceability detection data, effectively preventing erroneous judgment information from entering the traceability chain, thereby improving product qualification rates and enhancing the stability and reliability of lifecycle quality analysis results.

[0051] This invention provides, for example Figure 2 The plant-based cat litter full life cycle traceability and detection system shown includes a brightness acquisition and recording module, a light spot diffusion analysis module, a brightness sequence construction module, a signal constraint recognition module, and a dynamic judgment and control module; Brightness Acquisition and Recording Module: Acquires continuous image brightness change information of the finished plant cat litter under different low light conditions at the testing site, continuously records the time and location of instantaneous bright spots on the surface of cat litter particles, and generates an initial brightness distribution reference; Light spot diffusion analysis module: Based on the initial brightness distribution reference, the diffusion direction changes of bright spots in adjacent image frames are continuously compared, and the change characteristics of brightness from local concentration to spread to the particle surface are extracted. Within the range of the initial brightness distribution reference, the characteristics of migrating light spots and sudden reflections are distinguished. Brightness sequence construction module: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time, and the brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. Signal constraint recognition module: Based on the brightness evolution sequence, it performs constraint analysis on the brightness growth rate of suspected impurity areas in the image, and classifies light spots that show slow changes in the brightness evolution sequence as migration signal sources; Dynamic Judgment Control Module: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and judgment is only performed in the bright areas that show abrupt changes in the brightness evolution sequence.

[0052] The plant-based cat litter full life cycle traceability and detection method provided in this embodiment of the invention is implemented through the above-mentioned plant-based cat litter full life cycle traceability and detection system. For details of the specific methods and processes of the plant-based cat litter full life cycle traceability and detection system, please refer to the above-mentioned embodiment of the plant-based cat litter full life cycle traceability and detection method, which will not be repeated here.

[0053] 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 tracing and testing the entire life cycle of plant-based cat litter, characterized in that, Includes the following steps: Step 1: Collect continuous image brightness change information of the finished plant-based cat litter under different low light conditions at the testing site, continuously record the time and location of instantaneous bright spots on the surface of the cat litter particles, and generate an initial brightness distribution reference; Step 2: Based on the initial brightness distribution reference, continuously compare the changes in the diffusion direction of bright spots in adjacent image frames, extract the change features of brightness from local concentration to spread to the particle surface, and distinguish between migrating spot features and sudden reflection features within the initial brightness distribution reference range; Step 3: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time. The brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. Step 4: Based on the brightness evolution sequence, constrain the brightness growth rate of suspected impurity regions in the image, and classify the light spots that show slow changes in the brightness evolution sequence as the source of migration signals; Step 5: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and the judgment is only performed in the bright areas that show abrupt changes in the brightness evolution sequence.

2. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 1, characterized in that, The steps for collecting continuous image brightness variation information under different low-light conditions at the testing site of the finished plant-based cat litter product include: A low-light acquisition environment was constructed at the finished product inspection site. A stable light source and a fixed acquisition angle were set up in the inspection area. Multiple lighting states were established by continuously adjusting the light intensity and exposure duration, and the sampling frequency was kept constant to form a time-continuous image sequence. The obtained continuous image sequence is compared frame by frame. The time, location coordinates and brightness value of the bright spots are recorded according to the difference in brightness changes between adjacent image frames, and stored in chronological order to form bright spot distribution data. By using bright spot time and location records to generate continuous brightness variation curves, brightness data within the same time period are superimposed to form a spatial distribution map, and integrated into an initial brightness distribution reference with three-dimensional attributes of time, brightness and space. During the detection process, based on the initial brightness distribution reference, the brightness information of the real-time image is compared in terms of time and position. The consistency between the brightness change and the diffusion trajectory is used to determine whether the light spot belongs to the fluorescence migration process.

3. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 1, characterized in that, The steps for continuously comparing the changes in the diffusion direction of bright spots in adjacent image frames based on the initial brightness distribution reference include: Under the initial brightness distribution reference conditions, adjacent image frames acquired consecutively are arranged in time order. The spatial position and brightness range of the bright spot at different time points are determined with the brightness peak point as the center. The trend of the bright spot spreading outward from the central region is identified by comparing the changes in brightness intensity. The spatial evolution path of the diffusion is determined by the change in the brightness profile of the bright spot in adjacent image frames. The relative positions of the brightness enhancement and attenuation regions are analyzed and the diffusion direction of the bright spot is determined by using the initial brightness distribution as a reference. The brightness distribution maps of adjacent image frames in the time series are superimposed on the same spatial coordinates. By tracking the distribution trend of brightness from the center to the edge, the change features of brightness from local concentration to spread to the particle surface are extracted and a brightness change curve is generated. By combining the range of the initial brightness distribution reference, the characteristics of bright spot change are matched with the characteristics of brightness evolution. Based on the temporal continuity and spatial extension of brightness change, the characteristics of migrating spot and sudden reflection are distinguished.

4. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 3, characterized in that, When distinguishing between migrating spot features and sudden reflection features, the temporal continuity of brightness change is judged by the stable increase of the brightness difference between adjacent image frames, and the spatial extension is judged by the consistent direction of the brightness enhancement area in consecutive frames. Only when both conditions are met simultaneously is the spot identified as a migrating spot feature.

5. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 3, characterized in that, The steps for time-reordering the exposure response sequence of the detected image based on the diffusion change characteristics corresponding to the migrating light spot include: The continuous image frames in the detection screen are indexed and organized by time. The exposure response parameters of each frame are matched with the brightness records and sorted according to the acquisition time, exposure order and brightness change rate. The continuous diffusion area in the initial brightness distribution reference is used as the fixed reference for the time series. Based on the diffusion change characteristics of the migrating light spot, the exposure response sequence is rearranged in time, and the brightness changes that continue to spread in adjacent image frames are arranged on the same brightness evolution path, and the image frames with sudden brightness increases are identified separately as instantaneous reflection signals. The rearranged image frames were continuously compared, and the continuous diffusion trend was determined by comparing the spatial distribution of brightness changes in adjacent frames. The brightness change curves within the brightness evolution range were extracted as the evolution trajectory of the migrating light spot. The filtered brightness change data are recombined into a continuous and consistent brightness evolution sequence, and the time and spatial coordinates of the initial brightness distribution reference are kept consistent during the generation process to obtain a complete brightness evolution record.

6. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 5, characterized in that, During the time reordering process, the continuous diffusion region in the initial brightness distribution reference is used as a fixed benchmark. The time interval and spatial position of brightness changes in consecutive image frames are synchronously aligned. By removing instantaneous reflection signals, only brightness data with a continuous diffusion trend are retained.

7. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 5, characterized in that, The steps for constraining the brightness growth rate of suspected impurity regions in an image based on the brightness evolution sequence include: The brightness change areas in the detection image are identified and calibrated. By comparing the brightness evolution curves in consecutive image frames frame by frame, the areas with large brightness changes are identified and the growth curve of each brightness change point is established to form a brightness growth record. Constraint analysis was performed on the temporal variation characteristics of brightness growth records in the brightness evolution sequence. By comparing continuous time periods, the changing trend of brightness growth rate was identified, and stable brightness growth regions and regions with prominent changes were marked in the brightness evolution sequence to distinguish the source of migration signals from foreign object signals. The identified migration signal source areas are classified and marked. Light spots with temporal extension and spatial traceability of brightness changes are recorded as migration signal sources, and the time starting point, initial brightness value, and spatial coordinates are kept consistent with the brightness evolution sequence. The foreign object detection path is updated by comparing the migration signal record with the brightness evolution sequence to simultaneously exclude the brightness change data corresponding to the migration signal source in both time and space dimensions, so as to preserve the true foreign object signal distribution.

8. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 7, characterized in that, In the constraint analysis of the brightness growth rate, by comparing the brightness change amplitude at consecutive time points in the brightness evolution sequence, a time continuity threshold and a spatial extension range are set. When the brightness change curve is continuous in time and the spatial distribution is consistent with the source of the migration signal, the corresponding brightness data is automatically included in the migration signal record.

9. The method for tracing and detecting the entire life cycle of plant-based cat litter according to claim 7, characterized in that, The steps for dynamically tightening the brightness triggering conditions in the finished product inspection and judgment process based on the spatial distribution results of the migration signal source include: The detection image is divided into spatial regions. The brightness change region is divided into different brightness levels according to the spatial characteristics of the migration signal distribution, using the brightness evolution sequence as the time reference. Time evolution curves and brightness change rate data are established in each spatial unit. The brightness triggering conditions in each spatial region are reset, with strict triggering conditions set in the region dominated by the migration signal, and sensitivity maintained in the region of sudden bright changes to achieve dynamic adjustment of the zones; The detection and judgment process is executed with dual synchronization of time and space. By combining the time nodes in the brightness evolution sequence with the spatial distribution results, the corresponding triggering conditions are selected at different stages and the brightness changes are dynamically responded to. In the dynamic tightening judgment process, the bright areas with abrupt changes in brightness evolution sequence are analyzed in detail. By comparing the brightness changes at different time points, sudden reflection characteristics are identified and areas that meet the conditions are marked as foreign object judgment targets.

10. A plant-based cat litter lifecycle traceability and testing system, used to implement the plant-based cat litter lifecycle traceability and testing method described in any one of claims 1-9, characterized in that, It includes a brightness acquisition and recording module, a spot diffusion analysis module, a brightness sequence construction module, a signal constraint recognition module, and a dynamic judgment and control module; Brightness Acquisition and Recording Module: Acquires continuous image brightness change information of the finished plant cat litter under different low light conditions at the testing site, continuously records the time and location of instantaneous bright spots on the surface of cat litter particles, and generates an initial brightness distribution reference; Light spot diffusion analysis module: Based on the initial brightness distribution reference, the diffusion direction changes of bright spots in adjacent image frames are continuously compared, and the change characteristics of brightness from local concentration to spread to the particle surface are extracted. Within the range of the initial brightness distribution reference, the characteristics of migrating light spots and sudden reflections are distinguished. Brightness sequence construction module: Combining the diffusion change characteristics corresponding to the migrating light spot, the exposure response sequence of the detection image is rearranged in time, and the brightness change that shows a continuous diffusion trend in the initial brightness distribution reference is separated from the instantaneous reflection signal to form a continuous and consistent brightness evolution sequence. Signal constraint recognition module: Based on the brightness evolution sequence, it performs constraint analysis on the brightness growth rate of suspected impurity areas in the image, and classifies light spots that show slow changes in the brightness evolution sequence as migration signal sources; Dynamic Judgment Control Module: Based on the spatial distribution results of the migration signal source, the brightness triggering conditions in the finished product detection and judgment process are dynamically tightened, and judgment is only performed in the bright areas that show abrupt changes in the brightness evolution sequence.

Images