Floor antibacterial detection method and system

Abstract

The application relates to the technical field of antibacterial detection, and discloses a floor antibacterial detection method and system, which can effectively solve the influence of the image quality problem on the detection result in the prior art and the problem of inaccurate identification of microbial colonies by performing quality evaluation on multispectral image data, dynamically adjusting image acquisition parameters, and accurately distinguishing microbial colonies from non-biological foreign matters based on dynamic changes of spectral responses, and has the advantages of being capable of effectively solving the influence of the image quality problem on the detection result in the prior art, accurately distinguishing microbial colonies from non-biological foreign matters, and thus improving the accuracy and reliability of floor antibacterial detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of antibacterial detection, and in particular to a floor antibacterial detection method and system. BACKGROUND

[0002] In the conventional detection of floor antibacterial performance, although the operator strictly follows the established process and operates in a controlled experimental environment, multiple challenges still exist, which leads to deviation in microbial growth evaluation and inconsistency between the final antibacterial performance evaluation result and the actual situation.

[0003] In the prior art, due to the cumulative effects of a series of physical factors, including microscopic defects during sample preparation, fiber embedding during the cleaning process, sedimentation of tiny substances in the air, changes in the behavior of the cleaning solution, unevenness of the culture medium wettability, and complex disturbances of the instantaneous fluctuations of the temperature sensor on the microbial growth, combined with the neglect of these subtle abnormalities by the operator during image interpretation, the floor antibacterial effect calculated according to the microbial count rules and antibacterial performance determination standards has an unexplainable difference with the actual antibacterial performance of the floor. This difference further causes inconsistency in historical detection records, which brings challenges to subsequent product quality evaluation and improvement.

[0004] In particular, in this complex background, how to accurately distinguish the true microbial colonies from various non-biological foreign matters (such as fibers, particulate matters, etc.) from the multispectral image data and exclude the influence of image quality problems (such as optical distortion, feature confusion) on the detection result has become a key problem that needs to be solved in the current technology. SUMMARY

[0005] The present application provides a floor antibacterial detection method and system to accurately evaluate the antibacterial performance of the floor in a complex environment and ensure the consistency of the detection result with the actual situation. In a first aspect, to solve the above technical problems, the present application provides a floor antibacterial detection method, comprising:

[0006] Preferably, a floor antibacterial detection method comprises:

[0007] Obtaining first multispectral image data of a to-be-detected area;

[0008] Performing quality assessment on the first multispectral image data and obtaining a quality assessment result;

[0009] Adjusting image acquisition parameters according to the quality assessment result;

[0010] Obtaining second multispectral image data according to the image acquisition parameters, and determining microbial colonies and non-biological foreign matters according to the second multispectral image data.

[0011] Preferably, the adjusting the image acquisition parameters according to the quality assessment result comprises:

[0012] According to the quality assessment result, identifying an abnormal region in the first multi-spectral image data, the abnormal region comprising a region where optical distortion or feature confusion exists;

[0013] Adjusting optical parameters of the abnormal region until the image quality of the abnormal region reaches a preset standard.

[0014] Preferably, the determining the microbial colony and the non-biological foreign matter according to the second multi-spectral image data comprises:

[0015] According to the second multi-spectral image data, identifying a complex formed by the microbial colony and the non-biological foreign matter superimposed;

[0016] Identifying spectral response characteristics of the complex at different wavelengths;

[0017] Obtaining dynamic changes of the spectral response characteristics;

[0018] According to the dynamic changes, distinguishing the growth spectral response of the microbial colony from the non-growth spectral response of the non-biological foreign matter;

[0019] According to the distinguishing result, determining the microbial colony and the non-biological foreign matter from the complex.

[0020] Preferably, the distinguishing the growth spectral response of the microbial colony from the non-growth spectral response of the non-biological foreign matter according to the dynamic changes comprises:

[0021] Monitoring spectral response changes of the complex at a plurality of preset wavelengths;

[0022] Identifying change rates of the spectral response changes;

[0023] Judging whether there is a biological correlation between the spectral response changes;

[0024] According to the biological correlation and the change rates, distinguishing the growth spectral response of the microbial colony from the non-growth spectral response of the non-biological foreign matter.

[0025] Preferably, the judging whether there is a biological correlation between the spectral response changes comprises:

[0026] According to the spectral response changes, identifying whether there is a non-linear growth trend;

[0027] According to the geometric morphology of the complex, identifying whether there is a sustained diffusion trend;

[0028] According to the nonlinear growth trend and the continuous diffusion trend, it is determined whether there is a biological correlation between the spectral response changes.

[0029] Preferably, the identifying whether there is a nonlinear growth trend according to the spectral response changes comprises:

[0030] Monitoring the amplitude of the spectral response of the complex at multiple preset wavelengths over time;

[0031] According to the amplitude, the instantaneous change rate of the amplitude in a continuous time interval is calculated;

[0032] According to the instantaneous change rate, the fluctuation characteristics of the instantaneous change rate over time are analyzed;

[0033] According to the fluctuation characteristics, it is determined whether there is a continuous acceleration or deceleration trend of the fluctuation characteristics, so as to identify whether there is a nonlinear growth trend of the spectral response changes.

[0034] Preferably, the identifying whether there is a continuous diffusion trend according to the geometric shape of the complex comprises:

[0035] Monitoring the change of the pixel area or the boundary perimeter of the complex over time;

[0036] According to the change, it is determined whether there is the continuous diffusion trend.

[0037] Preferably, the determining whether there is a continuous acceleration or deceleration trend of the fluctuation characteristics according to the fluctuation characteristics comprises:

[0038] Trend fitting is performed on the data of the instantaneous change rate over time to obtain a fitting curve;

[0039] Analyzing the characteristic parameters of the fitting curve;

[0040] According to the characteristic parameters, it is determined whether there is a continuous acceleration or deceleration trend of the fluctuation characteristics.

[0041] Preferably, the determining whether there is a biological correlation between the spectral response changes according to the nonlinear growth trend and the continuous diffusion trend comprises:

[0042] Monitoring the correlation of the nonlinear growth trend and the continuous diffusion trend over time;

[0043] Evaluating the correspondence of the nonlinear growth trend and the continuous diffusion trend in a spatial region;

[0044] According to the correlation and the correspondence, it is determined whether there is a biological correlation between the spectral response changes.

[0045] In a second aspect, the present application provides a floor antibacterial detection system, comprising:

[0046] a detection end configured to obtain first multi-spectral image data of a to-be-detected area;

[0047] a processing end configured to perform quality assessment on the first multi-spectral image data and obtain a quality assessment result, and adjust image acquisition parameters according to the quality assessment result;

[0048] an output end configured to obtain second multi-spectral image data according to the image acquisition parameters, and determine microbial colonies and non-biological foreign matters according to the second multi-spectral image data.

[0049] Compared with the prior art, the present application effectively solves the problems of the influence of image quality on the detection result and inaccurate identification of microbial colonies by performing quality assessment on multi-spectral image data and dynamically adjusting image acquisition parameters, and accurately distinguishing microbial colonies from non-biological foreign matters based on dynamic changes in spectral response, thereby improving the accuracy and reliability of floor antibacterial detection. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 is a floor antibacterial detection method flowchart provided by an embodiment of the present application;

[0051] Figure 2 is a floor antibacterial detection system structure diagram provided by an embodiment of the present application. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0053] Referring to Figure 1 , the first embodiment of the present application provides a floor antibacterial detection method flowchart, comprising the following steps:

[0054] S1, obtaining first multi-spectral image data of a to-be-detected area;

[0055] S2, performing quality assessment on the first multi-spectral image data and obtaining a quality assessment result;

[0056] S3, adjusting image acquisition parameters according to the quality assessment result;

[0057] S4, acquiring second multi-spectral image data according to the image acquisition parameters, and determining microbial colonies and non-biological foreign matter according to the second multi-spectral image data.

[0058] Wherein, the multi-spectral image data refers to spectral information containing multiple discrete wavebands, each waveband corresponds to the light intensity in a specific wavelength range, which can be obtained by using multi-spectral cameras, spectrometers combined with scanning devices and other equipment, and its main purpose is to capture the optical response of the detection area under different wavelengths, and to provide more comprehensive spectral feature information than traditional monochrome or tricolor images, and to lay a data foundation for subsequent material identification and differentiation.

[0059] Quality evaluation refers to analyzing image data to determine whether it meets the pre-set quality standards or whether there are defects affecting subsequent processing, which can be achieved by analyzing the signal-to-noise ratio, clarity, uniformity, whether there is optical distortion or feature confusion, and other indicators of the image, and its main purpose is to identify possible problems in the image data, ensure the accuracy and reliability of subsequent analysis, and avoid misjudgment caused by low-quality data.

[0060] Image acquisition parameters refer to various settings that affect the image data acquisition process, such as exposure time, gain, light source intensity, focal length, aperture, white balance, etc., which can be adjusted by software control, hardware adjustment, etc., and its main purpose is to optimize the quality of image data, making it more suitable for subsequent analysis and processing, such as improving image contrast, reducing noise, correcting color deviation, and thus enhancing the distinguishability of target features.

[0061] Determining microbial colonies and non-biological foreign matter refers to identifying and distinguishing microbial colonies and non-biological impurities in the detection area based on image data, which can be achieved by using spectral analysis, morphological feature recognition, time series dynamic change tracking and other image processing and pattern recognition techniques, and its main purpose is to accurately identify whether there is microbial growth on the floor surface, and to distinguish it from dust, fibers and other non-biological impurities that may exist in the environment, which is a key step in evaluating the antibacterial performance of the floor.

[0062] In the present application, the quality evaluation of multi-spectral image data and the dynamic adjustment of image acquisition parameters are combined in a feedback optimization manner, thereby solving the technical problem that microbial colonies and non-biological foreign matter are difficult to accurately distinguish due to image quality problems in traditional detection, and achieving the effect of improving the reliability of detection data and the accuracy of antibacterial performance evaluation.

[0063] In step S1, first multi-spectral image data of the detection area is acquired.

[0064] Specifically, a multispectral camera equipped with a tunable filter array can be used to scan and image the floor surface in a predetermined wavelength range, for example, visible to near-infrared band, thereby obtaining the first multispectral image data of the detection area.

[0065] In step S2, the first multispectral image data is quality evaluated and a quality evaluation result is obtained.

[0066] Specifically, quality evaluation of multispectral image data can identify defects or abnormalities present in the image, such as optical distortion, noise interference, etc., providing a basis for subsequent parameter adjustment and avoiding the impact of image quality problems on the final detection results.

[0067] In step S3, the image acquisition parameters are adjusted according to the quality evaluation result.

[0068] Specifically, by adjusting the image acquisition parameters, the image quality can be optimized and errors can be reduced. For example, if the quality evaluation result shows that the image brightness is insufficient, the exposure time or light source intensity can be adjusted to improve the signal-to-noise ratio of the image and make the differentiation between microbial colonies and non-biological foreign objects more clear.

[0069] Preferably, the adjusting of the image acquisition parameters according to the quality evaluation result comprises:

[0070] According to the quality evaluation result, an abnormal area in the first multispectral image data is identified, the abnormal area including an area where optical distortion or feature confusion exists.

[0071] The optical parameters of the abnormal area are adjusted until the image quality of the abnormal area reaches a predetermined standard.

[0072] First, according to the quality evaluation result, an abnormal area in the first multispectral image data is identified. The "abnormal area" is defined as an area where optical distortion or feature confusion exists, i.e., the area in the image where the information is inaccurate or blurred due to various factors such as uneven lighting, lens distortion, uneven sample surface, etc. By identifying these areas, unnecessary adjustments to the good quality parts of the image can be avoided, thereby reducing the possibility of introducing new errors.

[0073] Then, for the identified abnormal regions, their optical parameters are adjusted individually until the image quality of these regions reaches the preset standard. This targeted adjustment strategy can more effectively correct defects in the image and improve the clarity and accuracy of the image. For example, if there is optical distortion in a certain region, the focal length or angle of the lens can be adjusted; if there is feature confusion in a certain region, the light intensity or spectral range can be adjusted. Through iterative adjustment, until the image quality of the abnormal region meets the preset standard, so as to ensure that the subsequent identification of microbial colonies and non-biological foreign matter can be based on high-quality image data, and the accuracy and reliability of the identification are improved.

[0074] In step S4, second multi-spectral image data is acquired according to the image acquisition parameters, and microbial colonies and non-biological foreign matter are determined according to the second multi-spectral image data.

[0075] For example, spectral feature analysis is performed on the second multi-spectral image data, by comparing the spectral response curves of pixels at different wavelengths, regions with typical absorption or reflection characteristics of microorganisms are identified, and they are distinguished from regions with spectral characteristics of non-biological foreign matter, such as dust or fibers, so as to determine microbial colonies and non-biological foreign matter.

[0076] Preferably, the determination of microbial colonies and non-biological foreign matter according to the second multi-spectral image data comprises:

[0077] According to the second multi-spectral image data, a composite formed by the superposition of the microbial colonies and the non-biological foreign matter is identified;

[0078] The spectral response characteristics of the composite at different wavelengths are identified;

[0079] The dynamic changes of the spectral response characteristics are acquired;

[0080] According to the dynamic changes, the growth spectrum response of the microbial colonies is distinguished from the non-growth spectrum response of the non-biological foreign matter;

[0081] According to the distinction result, the microbial colonies and the non-biological foreign matter are determined from the composite.

[0082] First, according to the second multi-spectral image data, a composite formed by the superposition of the microbial colonies and the non-biological foreign matter is identified, which is the basis for subsequent differentiation. Through image processing technology, the region where the microbial colonies and the non-biological foreign matter may exist is preliminarily locked, avoiding the blindness of subsequent analysis and improving the efficiency.

[0083] Then, the spectral response characteristics of the composite at different wavelengths are identified, which is a key step in distinguishing the two. Different substances will exhibit different spectral responses at different wavelengths, and by analyzing these spectral response characteristics, data support can be provided for subsequent differentiation of microbial colonies and non-biological foreign objects. The reason for identifying the spectral response characteristics at different wavelengths is that the information at a single wavelength may not be sufficient to distinguish the two, while multi-wavelength information can provide more comprehensive characteristics.

[0084] Next, the dynamic changes in the spectral response characteristics of the composite at different wavelengths over time are tracked, because the time-varying behavior of microbial colonies and non-biological foreign objects is different. Microbial colonies will grow and reproduce, and their spectral response will change over time, while non-biological foreign objects will not change significantly. By tracking this dynamic change, the two can be more accurately identified.

[0085] After that, according to the dynamic changes, the growth spectrum response of microbial colonies is distinguished from the non-growth spectrum response of non-biological foreign objects. Growth spectrum response is unique to microbial colonies, while non-biological foreign objects do not have it. By distinguishing between the two responses, the two can be effectively distinguished. The reason for distinguishing according to the dynamic change is that the static spectral response may be affected by many factors, while the dynamic change can better reflect the essential characteristics of microbial colonies.

[0086] Finally, according to the results of the distinction, the non-biological foreign objects are stripped from the composite to obtain the information of the microbial colonies, which is the ultimate goal. Through the analysis and distinction of the above steps, the influence of non-biological foreign objects can be removed from the composite, so that more accurate information of microbial colonies can be obtained, providing reliable data support for subsequent evaluation of antibacterial performance.

[0087] For example, after obtaining the second multi-spectral image data, image processing algorithms such as pixel intensity thresholding, morphological operations, or deep learning models (such as U-Net) can be used to perform preliminary segmentation on the image to identify the composite formed by the superposition of microbial colonies and non-biological foreign objects. For example, a threshold of brightness or specific wavelength reflectance can be set to mark areas above the threshold as potential composite areas, or a convolutional neural network can be trained to identify composites with specific textures and shapes in the image.

[0088] Further, for each identified complex region, its spectral response features at multiple preset wavelengths can be extracted. Specifically, wavelengths such as 450nm, 550nm, 650nm, 750nm and 850nm can be selected from the multispectral image data, and the average reflectance or absorbance of each pixel within the complex region at these wavelengths can be calculated, thereby constructing the spectral curve of the complex. These spectral curves can serve as the unique "fingerprint" of the complex. On this basis, in order to track the dynamic changes of the spectral response features of the complex at different wavelengths over time, a time interval can be set, for example every 2 hours or 4 hours, and multispectral image acquisition of the same complex region can be repeated, and its spectral response features can be repeatedly extracted. By comparing the spectral curves of the same complex at different time points, the intensity change trend at each wavelength can be observed. For example, the difference or change rate of spectral intensity between adjacent time points can be calculated.

[0089] Subsequently, according to these dynamic changes, the growth spectrum response of the microbial colony and the non-growth spectrum response of the non-biological foreign matter can be distinguished. Specifically, it can be analyzed whether the spectral intensity of the complex at a specific wavelength presents a nonlinear growth trend, for example, if the spectral intensity presents an exponential growth over a period of time, it can indicate the presence of microbial growth. At the same time, the geometric morphological changes of the complex, such as whether its pixel area or boundary perimeter continuously expands, can be combined to assist in judgment. Non-biological foreign matter usually does not exhibit such biologically related nonlinear growth or continuous expansion. For example, a classifier such as support vector machine (SVM) or random forest can be trained, which inputs the spectral change data and morphological change data over time, and outputs whether the complex is a microbial colony or a non-biological foreign matter.

[0090] Finally, according to the distinction result, the non-biological foreign matter is stripped from the complex, and the microbial colony information is obtained. Once the microbial part in the complex is determined, image mask technology can be used to remove the non-biological foreign matter part from the complex image, and only the pixel information of the microbial colony is retained. For example, if a complex is classified as containing microorganisms, it can be marked as a microbial colony according to the area indicated by its growth spectrum response and expansion trend, and the remaining part is regarded as non-biological foreign matter and is removed, thereby obtaining a pure microbial colony image or its quantitative data, such as the number, area or biomass of colonies.

[0091] Preferably, the distinguishing, according to the dynamic changes, the growth spectrum response of the microbial colony and the non-growth spectrum response of the non-biological foreign matter comprises:

[0092] Monitoring the spectral response changes of the complex at multiple preset wavelengths;

[0093] identifying a rate of change of the spectral response changes;

[0094] determining whether there is a biological correlation between the spectral response changes;

[0095] distinguishing the growth spectral response of the microbial colonies from the non-growth spectral response of the non-biological foreign matter based on the biological correlation and the rate of change.

[0096] First, the spectral response changes of the complex at multiple preset wavelengths are monitored to obtain more comprehensive spectral information. The spectral response at a single wavelength may be disturbed by certain substances, while the spectral response at multiple wavelengths can provide more rich feature information, thus more accurately reflecting the composition and state of the complex.

[0097] Second, the rate of change of the spectral response changes is identified, which is a key step to distinguish between growth and non-growth responses. The growth of microbial colonies is usually accompanied by rapid changes in spectral response, while non-biological foreign matter usually exhibits relatively stable spectral response. By identifying the rate of change, the two types of substances can be preliminarily distinguished.

[0098] Then, it is determined whether there is a biological correlation between the spectral response changes, which is to exclude interference caused by non-biological factors. The growth of microbial colonies is a complex biological process, and there is usually some correlation between the spectral responses at different wavelengths. For example, the absorption peak at certain wavelengths may be enhanced with the enhancement of the absorption peak at other wavelengths. If there is no such biological correlation between the spectral response changes, it is likely that these changes are caused by non-biological factors.

[0099] Finally, based on the biological correlation and the rate of change, the growth spectral response of the microbial colonies and the non-growth spectral response of the non-biological foreign matter are comprehensively distinguished. By comprehensively considering the biological correlation and the rate of change, the microbial colonies and the non-biological foreign matter can be more accurately distinguished, thus improving the accuracy of the information of the microbial colonies.

[0100] Specifically, in distinguishing the growth spectral response of the microbial colonies from the non-growth spectral response of the non-biological foreign matter, a multi-spectral imaging system can be first used to collect images of the complex at different time points of microbial cultivation, such as every 2 hours, and obtain spectral response data at multiple preset wavelengths. These preset wavelengths can include wavelength bands sensitive to microbial growth, such as 450 nm, 550 nm, 680 nm, and 800 nm, etc.

[0101] Subsequently, time series analysis can be performed on the spectral response data at each preset wavelength to calculate the rate of change. For example, the difference in spectral response intensity at a specific wavelength (e.g. 450 nm) between adjacent time points (e.g. T1 and T2) can be divided by the time interval (T2-T1) to obtain the instantaneous rate of change at that wavelength. For all preset wavelengths, a corresponding series of rates of change can be calculated.

[0102] Further, it can be determined whether there is a biological correlation between the changes in spectral response. This can include analyzing whether the rates of change at different wavelengths exhibit a coordinated non-linear growth trend, for example, if the rates of change of spectral response at 450 nm and 680 nm wavebands exhibit accelerated growth simultaneously, this can be related to the accumulation of pigments or metabolites within the microbial cells. At the same time, the geometric morphological changes of the complex can be combined, for example, whether the pixel occupancy area or the boundary perimeter of the complex exhibits a sustained diffusion trend is monitored, because the growth of microbial colonies is usually accompanied by the expansion of the colonies in space. If the non-linear growth trend of the spectral response and the sustained diffusion trend of the geometric morphology exhibit consistency in time and space, it can be determined that there is a biological correlation.

[0103] Finally, according to the biological correlation and the rate of change identified above, the growth spectrum response of the microbial colony and the non-growth spectrum response of the non-biological foreign matter can be comprehensively judged and distinguished. For example, if the spectral response of the complex exhibits a significant non-linear growth trend at multiple wavelengths with a biological correlation, and the rate of change is high and sustained, it can be determined as the growth spectrum response of the microbial colony; on the contrary, if the rate of change of the spectral response is low and lacks biological correlation, it can be determined as the non-growth spectrum response of the non-biological foreign matter. In this way, the small changes of non-biological foreign matter can be effectively avoided from being misjudged as microbial growth, thereby improving the accuracy of the distinction.

[0104] Preferably, the determining whether there is a biological correlation between the changes in spectral response comprises:

[0105] According to the spectral response changes, it is identified whether there is a non-linear growth trend;

[0106] According to the geometric morphology of the complex, it is identified whether there is a sustained diffusion trend;

[0107] According to the non-linear growth trend and the sustained diffusion trend, it is determined whether there is a biological correlation between the changes in spectral response.

[0108] The nonlinear growth trend refers to the fact that the magnitude of the spectral response changes over time presents a nonlinear growth pattern, such as exponential growth or S-shaped growth. Specifically, it can be identified by curve fitting on the spectral response data and analyzing the growth rate change of the fitted curve. The purpose is to reflect the dynamic change of cell number or metabolic product accumulation in the growth process of microbial colonies.

[0109] The continuous diffusion trend refers to the fact that the complex exhibits a continuous area expansion or boundary extension in the spatial dimension. Specifically, it can be tracked by image processing techniques such as edge detection or region growing algorithm to track the morphological changes of the complex at consecutive time points. The purpose is to reflect the spreading process of microbial colony population on the surface or inside of the culture medium.

[0110] The biological relevance refers to the fact that there is an inherent correlation between the change in spectral response and the change in complex geometry driven by biological activity. Specifically, it can be analyzed by establishing a mathematical model or logical rule to correlate the nonlinear growth characteristics of the spectral response with the continuous diffusion characteristics of the geometry. The purpose is to distinguish the real signal caused by microbial growth from the interference signal caused by non-biological factors.

[0111] Firstly, the growth of microbial colonies usually presents an exponential growth trend, so its spectral response change should also present a nonlinear growth feature. By identifying whether the spectral response change has a nonlinear growth trend, it can be preliminarily judged whether the change is related to the growth of microbial colonies. Secondly, the growth of microbial colonies will cause changes in their geometric morphology, such as the gradual increase in pixel occupancy area or boundary perimeter. By identifying whether the complex has a continuous diffusion trend in its geometric morphology, it can be further judged whether the change is related to the growth of microbial colonies.

[0112] Finally, it is determined whether there is a biological relevance between the changes in spectral response. If the spectral response change presents a nonlinear growth trend and the geometric morphology of the complex presents a continuous diffusion trend, it can be considered that there is a biological relevance between the two, i.e. the spectral response change is likely to be caused by the growth of microbial colonies. Conversely, if the spectral response change does not present a nonlinear growth trend, or the geometric morphology of the complex does not present a continuous diffusion trend, it can be considered that there is no biological relevance between the two, i.e. the spectral response change is likely not caused by the growth of microbial colonies, but by non-biological foreign objects or other factors.

[0113] By comprehensively considering the characteristics of spectral response change and the geometric morphology of the complex, and determining whether there is a biological relevance between the two, the growth spectrum of microbial colonies and the non-growth spectrum of non-biological foreign objects can be more accurately distinguished, thereby improving the accuracy of floor antimicrobial detection.

[0114] Preferably, the identifying whether there is a non-linear growth trend according to the change in the spectral response comprises:

[0115] monitoring the magnitude of the change in the spectral response of the complex over time at a plurality of preset wavelengths;

[0116] calculating the instantaneous change rate of the magnitude within a continuous time interval according to the magnitude;

[0117] analyzing fluctuation characteristics of the instantaneous change rate over a time series according to the instantaneous change rate;

[0118] judging whether there is a sustained acceleration or deceleration trend in the fluctuation characteristics to identify whether there is a non-linear growth trend in the change in the spectral response according to the fluctuation characteristics.

[0119] Wherein, monitoring the magnitude of the change in the spectral response of the complex over time at a plurality of preset wavelengths refers to obtaining the change in the intensity of reflected, absorbed or transmitted light of the complex over time within a specific wavelength range, which can be achieved by continuously collecting data with a spectrometer or by extracting pixel value changes through timed imaging with a multispectral camera, and the purpose is to quantify the changes in the optical properties of the complex at different growth stages.

[0120] Calculating the instantaneous change rate of the magnitude within a continuous time interval refers to determining the change speed of the spectral response magnitude within a very short time period through mathematical methods such as difference, derivative or sliding window average, which can be achieved by numerical differentiation algorithm or regression analysis-based method, and the purpose is to capture the dynamic characteristics of the change in the spectral response and more sensitively reflect its growth trend.

[0121] Analyzing the fluctuation characteristics of the instantaneous change rate over a time series refers to performing time series analysis on the instantaneous change rate data to identify its ups and downs, periodicity, trend or abnormal points at different time points, which can be achieved by Fourier transform, wavelet analysis or statistical methods (such as autocorrelation analysis), and the purpose is to reveal the possible internal physiological or environmental response mechanisms in the growth process of microorganisms.

[0122] Judging whether there is a sustained acceleration or deceleration trend in the fluctuation characteristics refers to evaluating whether the fluctuation pattern of the instantaneous change rate presents a sustained, non-linear growth or decay tendency, rather than random fluctuations or linear changes, which can be achieved by curve fitting, trend line analysis or machine learning algorithms (such as support vector machines, neural networks), and the purpose is to accurately identify the non-linear growth behavior of microbial colonies and distinguish them from non-biological foreign matter.

[0123] First, the magnitude of the spectral response of the complex at multiple preset wavelengths is monitored over time, which is a preliminary quantification of the growth state of the complex. By selecting multiple preset wavelengths, more comprehensive spectral information can be obtained, avoiding the limitations that may exist with a single wavelength, and providing a richer data basis for subsequent analysis.

[0124] Then, according to the magnitude, the instantaneous rate of change of the magnitude in consecutive time intervals is calculated, which realizes the transformation from static magnitude to dynamic rate. The instantaneous rate of change can more sensitively reflect the dynamic changes of the spectral response, capturing subtle growth trends. The way the consecutive time intervals are calculated ensures the continuity and stability of the rate of change, avoiding errors that may occur at discrete time points.

[0125] Next, according to the instantaneous rate of change, the fluctuation characteristics of the instantaneous rate of change over time are analyzed, which is a further analysis of the dynamic rate. By analyzing the fluctuation characteristics, the pattern and regularity of the rate of change can be identified, such as whether there is a periodic fluctuation, whether there is an abnormal mutation, etc. These fluctuation characteristics may reflect the internal mechanisms in the microbial growth process, such as the consumption of nutrients, the accumulation of metabolic products, etc.

[0126] Finally, according to the fluctuation characteristics, it is determined whether there is a sustained acceleration or deceleration trend in the fluctuation characteristics to identify whether there is a nonlinear growth trend in the spectral response change, which is the final judgment of the nonlinear growth trend. The sustained acceleration or deceleration trend is a key indicator for judging nonlinear growth, which reflects the change trend of the microbial growth rate. By judging whether there is such a trend in the fluctuation characteristics, nonlinear growth can be more accurately identified, excluding the interference of other factors.

[0127] In some preferred embodiments, specifically, when monitoring the magnitude of the spectral response of the complex at multiple preset wavelengths over time, a multispectral imaging system can be used, which is configured with multiple narrowband filters, such as image acquisition at 450 nm, 550 nm and 650 nm wavelengths. The system images the complex region every 10 minutes, and extracts the average pixel intensity value of the complex region from each wavelength image, which represents the magnitude of the spectral response. Then, when calculating the instantaneous rate of change of the magnitude in consecutive time intervals according to the magnitude, the central difference method can be used to calculate the time series data of the magnitude at each wavelength. For example, for the spectral response magnitude at time point t , the instantaneous rate of change can be approximated as , where The sampling time interval is, for example, 10 minutes. Subsequently, in analyzing the fluctuation characteristics of the instantaneous change rate on the time series according to the instantaneous change rate, the calculated instantaneous change rate series can be subjected to moving average processing to smooth short-term noise, and further, its second derivative can be calculated or wavelet decomposition can be performed. For example, by calculating the local variance or kurtosis of the rate series, the fluctuation degree and pattern thereof are quantified. Finally, in judging whether there is a sustained acceleration or deceleration trend according to the fluctuation characteristics, the processed instantaneous change rate series can be subjected to exponential function fitting or polynomial fitting. If the coefficients of the fitted curve indicate that there is a sustained positive or negative acceleration, and the goodness of fit reaches a preset threshold, it can be judged that there is a nonlinear growth trend.

[0128] Through the above technical solutions, the present application can effectively exclude the interference of environmental light changes, sensor noise and other non-biological factors on the spectral response through step-by-step in-depth analysis of the spectral response amplitude, instantaneous change rate and its fluctuation characteristics, thereby more accurately identifying the nonlinear growth trend specific to the microbial colony in the complex spectral response. This significantly improves the accuracy and reliability of judging the growth nature of the spectral response of microorganisms, provides a solid foundation for subsequent accurate separation of non-biological foreign matter from the complex, and further improves the accuracy of the floor antibacterial detection result.

[0129] Preferably, the judging whether there is a sustained acceleration or deceleration trend according to the fluctuation characteristics comprises:

[0130] trend fitting the data of the instantaneous change rate on the time series to obtain a fitted curve;

[0131] analyzing the characteristic parameters of the fitted curve;

[0132] judging whether there is a sustained acceleration or deceleration trend according to the characteristic parameters.

[0133] Wherein, trend fitting refers to modeling the time series data through mathematical models to reveal the potential laws or trends of data changes over time, which can be achieved by various mathematical methods such as linear regression, polynomial regression, exponential fitting, logarithmic fitting or spline fitting, and its purpose is to filter out random noise and short-term fluctuations in the data, thereby more clearly showing the long-term or medium-term change trend of the data.

[0134] The fitted curve refers to a mathematical curve obtained by trend fitting methods that can reflect the overall change trend of the data, which can take the form of a straight line, a parabola, an exponential curve or an S-shaped curve, etc., and its purpose is to provide a smooth and continuous function representation to facilitate subsequent quantitative analysis of the data trend.

[0135] The characteristic parameters refer to numerical indicators used to describe the shape and properties of the fitted curve, which can include but are not limited to the slope, intercept, curvature, inflection point position, goodness of fit (such as the coefficient of determination R²) or function value at a specific time point of the fitted curve, etc. The purpose is to accurately capture and express the trend information represented by the fitted curve in a quantitative manner, providing a basis for subsequent trend judgment.

[0136] Firstly, by fitting, noise and errors in the instantaneous change rate can be effectively filtered out, resulting in a more smooth and stable trend change curve. This fitting can be linear fitting or nonlinear fitting, and the specific choice of fitting method depends on the actual change of the instantaneous change rate. Through fitting, the overall trend of the instantaneous change rate can be observed more clearly, without being disturbed by local fluctuations.

[0137] Secondly, by extracting the characteristic parameters of the fitted curve, the trend change of the instantaneous change rate can be described more quantitatively. These characteristic parameters can include the slope, curvature, intercept, etc. of the fitted curve. For example, if the slope of the fitted curve is positive, it means that the instantaneous change rate presents an accelerating trend; if the slope of the fitted curve is negative, it means that the instantaneous change rate presents a decelerating trend; if the curvature of the fitted curve is large, it means that the change amplitude of the instantaneous change rate is large. By analyzing these characteristic parameters, the trend change of the instantaneous change rate can be more accurately judged.

[0138] Finally, by comprehensively considering the various characteristic parameters of the fitted curve, the trend change of the instantaneous change rate can be more comprehensively judged. For example, if the slope of the fitted curve is positive and the curvature is also large, it means that the instantaneous change rate presents a continuous accelerating trend; if the slope of the fitted curve is negative and the curvature is also large, it means that the instantaneous change rate presents a continuous decelerating trend. In this way, whether the spectral response change has a nonlinear growth trend can be more accurately identified, thereby providing a more reliable basis for the subsequent differentiation of microbial colonies and non-biological foreign matter.

[0139] In some preferred embodiments, determining whether the fluctuation feature has a persistent acceleration or deceleration trend can be implemented as follows: first, a set of data points of the instantaneous change rate over a time sequence is obtained, for example, the instantaneous change rate is recorded every 1 hour within a continuous 24 hours, obtaining 24 data points. Then, polynomial trend fitting is performed on these data points, for example, a quadratic polynomial or a cubic polynomial can be used for fitting to obtain a fitting curve. During the fitting process, the least squares method can be used to determine the polynomial coefficients, so as to minimize the deviation of the fitting curve from the original data points. Next, the characteristic parameters of the fitting curve are analyzed. For example, for a quadratic polynomial fitting curve. If the slope of the fitting curve is continuously positive and its absolute value gradually increases within a period of time, or the second derivative is continuously positive, it can be determined that the instantaneous change rate has a persistent acceleration trend. Conversely, if the slope is continuously negative and its absolute value gradually increases, or the second derivative is continuously negative, it can be determined that the instantaneous change rate has a persistent deceleration trend. In addition, the goodness of fit of the fitting curve can also be analyzed to evaluate the representativeness of the fitting curve to the original data, and according to the mathematical characteristics of the fitting curve, the persistent acceleration or deceleration trend of the fluctuation feature can be quantitatively and objectively determined.

[0140] Through the above technical solutions, by performing trend fitting on the data of the instantaneous change rate over a time sequence, the influence of environmental noise and measurement error on the data can be effectively filtered out, so that the obtained fitting curve can more accurately reflect the real change trend of the instantaneous change rate. Further, by analyzing the characteristic parameters of the fitting curve, the acceleration or deceleration trend of the instantaneous change rate can be quantitatively described, thereby avoiding the inaccuracy that may be caused by subjective judgment. This method significantly improves the accuracy and robustness of determining whether the fluctuation feature has a persistent acceleration or deceleration trend, and further makes it more reliable to identify whether the spectral response change has a nonlinear growth trend. Ultimately, this helps to more accurately distinguish the growth spectrum response of microbial colonies from the non-growth spectrum response of non-biological foreign matter, and provides a more reliable basis for floor antibacterial detection.

[0141] Preferably, the identification of whether there is a persistent diffusion trend according to the geometric shape of the complex includes:

[0142] Monitoring the change of the pixel occupation area or the boundary perimeter of the complex over a time sequence;

[0143] Determining whether there is the persistent diffusion trend according to the change.

[0144] The pixel occupation area refers to the number of pixels occupied by the complex in the image, which can be realized by using an image segmentation algorithm to identify the complex region and then counting the total number of all pixel points in the region, and the purpose is to quantify the actual coverage range of the complex in the two-dimensional plane.

[0145] The boundary perimeter refers to the total length of the contour line of the complex in the image, which can be identified by using an edge detection algorithm to identify the boundary of the complex, and the total distance between the pixel points on the boundary is calculated to achieve the purpose of reflecting the extension degree and morphological complexity of the edge of the complex.

[0146] The continuous diffusion trend refers to the pixel occupation area or the boundary perimeter of the complex showing a stable or accelerated growth trend at consecutive time points, which can be achieved by using trend analysis or regression model to fit and predict the time series data, and the purpose is to distinguish the real growth and diffusion of microbial colonies from the accidental morphological fluctuations caused by non-biological factors.

[0147] Specifically, monitoring the changes of the pixel occupation area or the boundary perimeter of the complex in the time series can reflect the expansion of the complex in the plane. The pixel occupation area directly reflects the size of the image area occupied by the complex, while the boundary perimeter reflects the extension degree of the edge of the complex. By continuously monitoring these parameters, the growth or diffusion trend of the complex over time can be observed.

[0148] Then, according to these changes, it is judged whether there is a continuous diffusion trend. If the pixel occupation area or the boundary perimeter continues to increase over time, it indicates that the complex is diffusing; on the contrary, if these parameters remain stable or decrease, it indicates that the complex has no obvious diffusion trend. This judgment method based on quantitative indicators is more objective and accurate than directly observing the morphological changes of the complex, and can effectively eliminate the interference of non-biological factors on the judgment result, so as to more accurately identify the growth and diffusion behavior of microbial colonies, and provide more reliable basis for subsequent biological correlation judgment.

[0149] Preferably, according to the nonlinear growth trend and the continuous diffusion trend, whether there is a biological correlation between the spectral response changes is judged, comprising:

[0150] Monitoring the correlation of the nonlinear growth trend and the continuous diffusion trend in the time series;

[0151] Evaluating the correspondence of the nonlinear growth trend and the continuous diffusion trend in the spatial region;

[0152] According to the correlation and the correspondence, it is judged whether there is a biological correlation between the spectral response changes.

[0153] The correlation in time series refers to the synchronous change or mutual influence relationship of the nonlinear growth trend and the continuous diffusion trend at different time points, which can be realized by cross-correlation analysis, Granger causality test or dynamic time warping, and the purpose is to reveal the collaborative evolution pattern of the two trends in the time dimension. In addition, the correspondence in spatial region refers to the overlap or consistency of the nonlinear growth trend and the continuous diffusion trend in the physical space distribution, which can be realized by image registration, region overlap calculation or spatial correlation analysis, and the purpose is to confirm whether the two trends occur in the same or closely adjacent region, so as to exclude the interference factors irrelevant in space.

[0154] First, monitor the correlation of the nonlinear growth trend and the continuous diffusion trend in time series, which means analyzing the synchronism of the two trends over time. If the nonlinear growth trend and the continuous diffusion trend show consistency in time, for example, the nonlinear growth accelerates at the same time, the diffusion speed also accelerates, it is more likely to indicate that there is a biological correlation, because biological growth usually shows such a synergistic effect. On the contrary, if the changes of the two have no obvious correlation in time, it may indicate that these changes are caused by non-biological factors.

[0155] Second, evaluate the correspondence of the nonlinear growth trend and the continuous diffusion trend in spatial region, which means investigating whether the two trends occur in the same region. If the nonlinear growth mainly occurs in the center of the diffusion region, and the expansion of the diffusion region coincides with the expansion of the growth region, it more strongly supports the hypothesis of biological correlation. This is because the growth of microorganisms usually leads to the expansion of the region they occupy. If growth and diffusion occur in different spatial regions, it may indicate that these phenomena are caused by different reasons, such as chemical reaction or physical diffusion.

[0156] Finally, according to the correlation and correspondence, comprehensively judge whether there is a biological correlation between the changes in spectral response. Only when the nonlinear growth trend and the continuous diffusion trend show strong correlation in time and space, it can be more reliably judged that these changes are caused by the growth of microorganisms, so as to distinguish the growth spectral response of microbial colonies from the non-growth spectral response of non-biological foreign matter. This comprehensive judgment method reduces the risk of misjudgment and improves the accuracy of antibacterial detection.

[0157] In some preferred embodiments, in order to determine whether there is a biological correlation between the spectral response changes, the nonlinear growth trend and the persistent diffusion trend can be first pre-processed, such as smoothing or normalizing the time series data. Subsequently, time series analysis methods can be employed to monitor their correlation in time series. For example, the Pearson correlation coefficient or dynamic time warping distance between the instantaneous rate of change series of the nonlinear growth trend and the instantaneous rate of change series of the persistent diffusion trend can be calculated to quantify their synchronicity. If the correlation coefficient is higher than a preset threshold, or the dynamic time warping distance is smaller than a preset threshold, it indicates a strong correlation. Meanwhile, image processing and spatial statistics methods can be employed to evaluate their correspondence in spatial regions. For example, the region where the spectral response nonlinear growth is most significant can be identified, as well as the region occupied by the persistent diffusion of the complex. Then, the overlapping area ratio or the centroid distance of the two regions can be calculated. If the overlapping area ratio is higher than a preset threshold, or the centroid distance is smaller than a preset threshold, it indicates a strong spatial correspondence. Finally, according to the correlation score in time series and the correspondence score in spatial regions, a weighted summation model or a decision tree model can be employed to comprehensively determine whether the spectral response changes have biological correlation. For example, a comprehensive determination threshold can be set, and only when both the correlation score and the correspondence score reach or exceed the threshold, it is determined that there is a biological correlation. This method can effectively distinguish the real changes caused by microbial growth from the false appearance caused by other non-biological factors, thereby improving the accuracy of detection.

[0158] In summary, the present application discloses a floor antibacterial detection method, which can effectively solve the problems of image quality affecting the detection results and inaccurate identification of microbial colonies in the prior art by performing quality assessment on multispectral image data and dynamically adjusting image acquisition parameters, and accurately distinguishing microbial colonies from non-biological foreign matter based on dynamic changes in spectral response. The present application has the advantages of effectively solving the problem of image quality affecting the detection results in the prior art, and accurately distinguishing microbial colonies from non-biological foreign matter, thereby improving the accuracy and reliability of floor antibacterial detection.

[0159] Reference Figure 2 The embodiment of the present application provides a system structure diagram for floor antibacterial detection, which comprises:

[0160] A detection end is configured to acquire first multispectral image data of a to-be-detected region.

[0161] A processing end is configured to perform quality assessment on the first multispectral image data and acquire a quality assessment result, and adjust image acquisition parameters according to the quality assessment result.

[0162] an output terminal configured to acquire second multi-spectrum image data according to the image acquisition parameter, and determine the microbial colony and the non-biological foreign matter according to the second multi-spectrum image data.

[0163] It should be noted that the floor antibacterial detection system provided by the embodiments of the present application is used to execute all process steps of the floor antibacterial detection method provided by the above embodiments, and the working principles and beneficial effects of the two are one-to-one correspondence, thus not repeated here.

[0164] The embodiments of the present application further provide a terminal device. The terminal device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor. The processor implements the steps in the above various floor antibacterial detection method embodiments when executing the computer program, for example Figure 1 The processor implements the functions of each module / unit in the above various system embodiments when executing the computer program.

[0165] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the terminal device.

[0166] The terminal device can be a desktop computer, a notebook computer, a palm computer, a smart tablet and the like. The terminal device can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the terminal device and do not constitute a limitation on the terminal device, and can include more or less components than the above, or combine certain components, or different components, for example, the terminal device can also include an input / output device, a network access device, a bus, etc.

[0167] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The processor is a control center of the terminal device, and connects all parts of the terminal device through various interfaces and lines.

[0168] The memory can be used to store the computer programs and / or modules, and the processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), etc.; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory device.

[0169] The modules / units integrated in the terminal device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or system, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium, etc. that can carry the computer program code. It should be noted that the computer readable medium can include or exclude contents according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signals and telecommunication signals.

[0170] It should be noted that the above-described system embodiments are only illustrative, and the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. In addition, the connection relationship between the modules in the system embodiments provided by the present application indicates that there is a communication connection between them, which can be realized as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0171] The above-described specific embodiments further illustrate the purpose, technical solutions, and beneficial effects of the present application. It should be understood that the above-described specific embodiments are only examples of the present application and are not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for detecting antibacterial properties of flooring, characterized in that, include: Acquire the first multispectral image data of the region to be detected; Perform a quality assessment on the first multispectral image data and obtain the quality assessment results; Adjust the image acquisition parameters based on the quality assessment results; The second multispectral image data is obtained based on the image acquisition parameters, and the microbial colonies and non-biological foreign matter are determined based on the second multispectral image data. The step of determining microbial colonies and abiotic foreign matter based on the second multispectral image data includes: Based on the second multispectral image data, identify the complex formed by the superposition of the microbial colony and the non-biological foreign matter; Identify the spectral response characteristics of the composite at different wavelengths; The dynamic changes of the spectral response characteristics are obtained; Based on the dynamic changes, the growth spectral response of the microbial colony is distinguished from the non-growth spectral response of the non-biological foreign matter; Based on the differentiation results, the microbial colonies and the non-biological foreign matter are identified from the complex; The step of distinguishing the growth spectral response of the microbial colony from the non-growth spectral response of the abiotic foreign substance based on the dynamic changes includes: Monitor the spectral response changes of the composite at multiple preset wavelengths; Identify the rate of change of the spectral response; Determine whether there is a biological correlation between the changes in the spectral response; Based on the biological relevance and the rate of change, the growth spectral response of the microbial colony is distinguished from the non-growth spectral response of the abiotic foreign substance.

2. The method according to claim 1, characterized in that, The step of adjusting the image acquisition parameters based on the quality assessment results includes: Based on the quality assessment results, abnormal regions in the first multispectral image data are identified, including regions with optical distortion or feature confusion. Adjust the optical parameters of the abnormal area until the image quality of the abnormal area reaches the preset standard.

3. The method according to claim 1, characterized in that, Determining whether there is a biological correlation between the changes in the spectral response includes: Identify whether there is a nonlinear growth trend based on the changes in the spectral response; The presence of a continuous diffusion trend can be identified based on the geometric morphology of the complex. Based on the nonlinear growth trend and the continuous diffusion trend, determine whether there is a biological correlation between the changes in the spectral response.

4. The method according to claim 3, characterized in that, The step of identifying whether there is a nonlinear growth trend based on the change in the spectral response includes: Monitor the magnitude of the spectral response of the composite at multiple preset wavelengths over time; Based on the amplitude, calculate the instantaneous rate of change of the amplitude over a continuous time interval; Based on the instantaneous rate of change, analyze the fluctuation characteristics of the instantaneous rate of change over a time series; Based on the fluctuation characteristics, it is determined whether the fluctuation characteristics have a continuous acceleration or deceleration trend, so as to identify whether the spectral response change has a nonlinear growth trend.

5. The method according to claim 3, characterized in that, The step of identifying whether there is a continuous diffusion trend based on the geometry of the complex includes: Monitor the changes in the pixel-occupied area or boundary perimeter of the composite over a time series; Determine whether the described continuous diffusion trend exists based on the changes.

6. The method according to claim 4, characterized in that, The step of determining whether the fluctuation characteristics have a continuous acceleration or deceleration trend based on the fluctuation characteristics includes: The instantaneous rate of change is trend-fitted onto the time series data to obtain a fitted curve; Analyze the characteristic parameters of the fitted curve; Based on the characteristic parameters, determine whether the fluctuation characteristic has a continuous acceleration or deceleration trend.

7. The method according to claim 3, characterized in that, The step of determining whether there is a biological correlation between the changes in the spectral response based on the nonlinear growth trend and the continuous diffusion trend includes: Monitor the correlation between the nonlinear growth trend and the continuous diffusion trend over time series; Evaluate the correspondence between the nonlinear growth trend and the continuous diffusion trend in spatial regions; Based on the correlation and correspondence, determine whether there is a biological correlation between the changes in the spectral response.

8. A floor antibacterial detection system for detecting the antibacterial properties of floors, the system comprising: The detection end is used to acquire the first multispectral image data of the area to be detected; The processing unit is used to perform quality assessment on the first multispectral image data and obtain the quality assessment result; and adjust the image acquisition parameters according to the quality assessment result. The output terminal is used to acquire second multispectral image data based on the image acquisition parameters, and to determine microbial colonies and non-biological foreign matter based on the second multispectral image data.

Images