Water quality multi-parameter monitoring method, device and equipment

Abstract

The invention relates to a water quality multi-parameter monitoring method, device and equipment. According to the water quality multi-parameter monitoring method, a shell is included, and a measuring cavity is defined by the shell; the light supplementing system is used for supplementing light to the measuring cavity; the ambient light catcher is used for detecting the intensity of ambient light in the measuring cavity; the target is provided with a plurality of color blocks; the image sensor module is used for penetrating through the water body in the measuring cavity to collect an image of the target; the self-cleaning mechanism is used for scraping dirt on the inner wall of the cavity; the main control processing unit is electrically connected with the light supplementing system, the ambient light catcher, the image sensor module and the self-cleaning mechanism. According to the invention, the ambient light capturer is used for coordinating the light supplementing system to supplement light to the measuring cavity, so that the background stability of the measuring light source is ensured, the acquisition and human eye errors are reduced by assisting the multi-color target, and the stability of the measuring cavity is always ensured through the self-cleaning mechanism.

Description

Technical Field

[0001] This invention relates to the field of water quality monitoring technology, and in particular to a method, apparatus and equipment for monitoring multiple parameters of water quality. Background Technology

[0002] The description in this section provides only background information related to the disclosure of this invention and does not constitute prior art.

[0003] Water transparency and color are important and intuitive indicators for evaluating water quality. Transparency reflects the turbidity of the water and is closely related to the content of suspended solids and plankton; water color indicates algal growth, dissolved organic pollution, and other conditions. Traditional transparency measurements mainly use secchi disks or lead type, which rely on manual visual inspection, are highly subjective, and cannot achieve continuous, online monitoring. Similarly, efficient and objective automated equipment is lacking for measuring water color.

[0004] Existing technologies include some water quality sensors based on optical principles, such as those using transmission or scattering optical methods to measure turbidity. However, these sensors typically only target a single parameter (such as turbidity), and their measurements differ from human visual perception of transparency or color. Furthermore, most optical sensors are susceptible to interference from factors such as changes in ambient light and lens contamination, leading to data drift and requiring frequent manual maintenance and calibration.

[0005] Therefore, there is an urgent need in this field for an integrated device that can simulate the principle of human visual discrimination, has strong anti-interference ability, and can simultaneously and automatically monitor the transparency and color of water.

[0006] It should be noted that the above description of the technical background is only for the purpose of providing a clear and complete explanation of the technical solutions of the present invention and facilitating understanding by those skilled in the art. It should not be assumed that the above technical solutions are known to those skilled in the art simply because they have been described in the background section of this invention. Summary of the Invention

[0007] The purpose of this invention is to provide a method, device, and equipment for monitoring multiple parameters of water quality. It can coordinate the supplementary lighting system with an ambient light catcher to supplement the lighting of the measuring cavity, ensuring the stability of the measuring light source background. It also reduces the error between the acquisition and human eye through the auxiliary multi-colored target, and ensures the stability of the measuring cavity through a self-cleaning mechanism.

[0008] To achieve the above objectives, the present invention discloses a multi-parameter water quality monitoring device, the multi-parameter water quality monitoring device comprising: A housing that encloses a measuring cavity having a first end and a second end disposed opposite to each other; A supplementary lighting system is provided, which is disposed on one side of the first end of the housing, with the light source end of the supplementary lighting system facing the measuring cavity, and is used to provide supplementary lighting to the measuring cavity; An ambient light trap is provided, which is disposed on one side of the second end of the housing, with the detection end of the ambient light trap facing the measuring cavity. The ambient light trap is used to detect the ambient light intensity in the measuring cavity. A target, wherein the target is disposed on one side of the second end of the housing, and the target is provided with multiple color blocks; An image sensor module is disposed on one side of the first end of the housing, with the acquisition end of the image sensor module facing the target. The image sensor module is used to acquire images of the target through the water in the measurement cavity. The self-cleaning mechanism has its movable end disposed in the measuring cavity, and is used to scrape off dirt from the inner wall of the cavity. The main control processing unit is electrically connected to the supplementary lighting system, the ambient light capture device, the image sensor module, and the self-cleaning mechanism.

[0009] As a further description of the above technical solution, the supplementary lighting system includes an LED light source, a light guide column, and a first high-resolution lens. The LED light source is installed on one side of the first end of the housing and electrically connected to the main control processing unit. The first high-resolution lens is disposed on one side of the first end of the inner wall of the measuring cavity. The light guide column is connected between the LED light source and the first high-resolution lens.

[0010] As a further description of the above technical solution, the ambient light catcher includes a photosensitive sensor and a second high-resolution lens. The photosensitive sensor is mounted on one side of the second end of the housing and is electrically connected to the main control processing unit. The second high-resolution lens is disposed on one side of the second end of the inner wall of the measuring cavity, and the photosensitive sensor is positioned toward the second high-resolution lens.

[0011] As a further description of the above technical solution, the target is set with seven standard color blocks of red, orange, yellow, green, cyan, blue and purple, and the seven color blocks are arranged in a fixed matrix or in concentric circles.

[0012] As a further description of the above technical solution, the image sensor module includes an image sensor, an optical lens, and a third high-resolution lens. The image sensor is mounted on one side of the first end of the housing and electrically connected to the main control processing unit. The third high-resolution lens is disposed on one side of the first end of the inner wall of the measuring cavity. The optical lens is connected between the image sensor and the third high-resolution lens.

[0013] As a further description of the above technical solution, the self-cleaning mechanism includes a motor, an extension shaft, and a cleaning scraper. The motor is installed on one side of the first end of the housing and is electrically connected to the main control processing unit. The extension shaft is connected to the output end of the motor. The cleaning scraper is fixed to the end of the extension shaft. The extension shaft passes through the housing and the cleaning scraper is disposed in the measuring cavity.

[0014] The present invention also discloses a water quality multi-parameter monitoring device, including the water quality multi-parameter monitoring device as described above, as well as an extension cable, an external power supply, and a control unit for powering the water quality multi-parameter monitoring device and enabling data communication.

[0015] A method for monitoring multiple water quality parameters, applied to the water quality multi-parameter monitoring device described above, wherein the method includes the following steps: After the self-cleaning mechanism cleans the inner wall of the measuring chamber, wait for the preset water stabilization delay time. The current ambient light intensity is obtained through an ambient light sensor and compared with a preset light signal reference value; Based on the comparison results, the luminous intensity of the supplementary lighting system is adaptively controlled to stabilize the total illuminance of the measurement environment in the measurement cavity near the optical signal reference value; or a fixed supplementary lighting intensity is set with a coefficient of 1, and the coefficient can be adjusted when the supplementary lighting attenuates subsequently. Under stable total illuminance, multispectral images of a seven-color target are acquired through an image sensor module; The acquired images are processed and analyzed to extract the image feature values ​​of each color block of the seven-color target; The extracted image feature values ​​are compared with a pre-stored calibration database, which contains standard image feature data for water bodies with different transparency and different colors. Based on the comparison results, the system calculates and outputs the current water body transparency monitoring value and water body color monitoring index using a preset algorithm model.

[0016] As a further description of the above technical solution, in the step of "processing and analyzing the acquired image and extracting the image feature values ​​of each color block of the seven-color target", the image feature values ​​include, but are not limited to, the overall image contrast of the entire target area, the sharpness of the edges of each color block, and the chromaticity and saturation values ​​of each color block in the RGB or HSV color space; in the step of "adaptively controlling the luminous intensity of the supplementary light system based on the comparison results, so that the total illuminance of the measurement environment in the measurement cavity is stabilized near the light signal reference value", the adaptive light compensation algorithm adopts a closed-loop feedback mechanism including proportional-integral control.

[0017] As a further description of the above technical solution, in the step of "comparing the extracted image feature values ​​with a pre-stored calibration database, wherein the calibration database contains standard image feature data under different water transparency and different water color conditions", the pre-stored calibration database is established by correlating and fitting the actual transparency values ​​measured by the lead type method with the image feature values ​​collected under the corresponding conditions; the algorithm model includes a multispectral color analysis model, which is set to feature extraction and regression judgment based on the edge sharpness of the target image and the color shift of multiple color blocks.

[0018] By employing the above technical solutions, the beneficial effects of the present invention are as follows: The water quality multi-parameter monitoring method, device, and equipment of this invention can coordinate the supplementary lighting system with an ambient light trap to illuminate the measurement cavity, ensuring a stable background for the measurement light source. It also reduces acquisition and human eye errors through an auxiliary multi-colored target and maintains the stability of the measurement cavity through a self-cleaning mechanism. Specifically, this invention is the first to apply adaptive multispectral image analysis to water quality monitoring, simultaneously obtaining two key apparent indicators: transparency and water color, demonstrating high functional integration. Furthermore, this invention exhibits strong anti-interference capabilities, effectively overcoming measurement errors introduced by changes in ambient light such as day / night, weather, and season through a unique "ambient light sensing combined with adaptive supplementary lighting" mechanism, resulting in extremely high data stability. Furthermore, the measurement results of this invention are intuitive and reliable. Using a seven-color target as a medium, the measurement principle closely resembles human visual discrimination methods (such as the sekker disk and lead type methods), making the results more consistent with human visual perception and highly practical. Furthermore, this invention has a high degree of automation and a long maintenance cycle, with a built-in automatic cleaning mechanism that significantly reduces the need for manual maintenance, making it suitable for long-term unattended online monitoring. Furthermore, the device of the present invention is small and easy to deploy. Its compact structural design and short optical path of 1-5 cm make the device small in size and low in power consumption, making it easy to integrate into various buoys, monitoring stations or pipelines.

[0019] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an exploded schematic diagram of a water quality multi-parameter monitoring device provided in the embodiments of this specification; Figure 2 This is a three-dimensional schematic diagram of a water quality multi-parameter monitoring device provided in the embodiments of this specification; Figure 3 This is a schematic diagram of the logic unit of a water quality multi-parameter monitoring method, device and equipment provided in the embodiments of this specification; In the picture: 1. Housing; 11. Measuring cavity; 2. Supplemental lighting system; 21. LED light source; 22. Light guide column; 23. First high-resolution lens surface; 3. Ambient light catcher; 31. Photosensitive sensor; 32. Second high-resolution lens; 4. Image sensor module; 41. Image sensor; 42. Optical lens; 43. Third high lens surface; 5. Self-cleaning mechanism; 51. Motor; 52. Extension shaft; 53. Cleaning scraper; 6. Target. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0023] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0024] It should be understood that while terms such as "first," "second," and "third" may be used in this document to describe various components or signals, these components or signals should not be limited by these terms. These terms are primarily used to distinguish one component from another, or one signal from another. Furthermore, the term "or" as used herein should, as appropriate, include any combination of one or more of the related listed items.

[0025] Please see Figure 1-3 This embodiment provides a water quality multi-parameter monitoring device, which includes: A housing 1 surrounds and forms a measuring cavity 11, the measuring cavity 11 having a first end and a second end disposed opposite to each other; A supplementary lighting system 2 is provided on one side of the first end of the housing 1, with the light source end of the supplementary lighting system 2 facing the measuring cavity 11. The supplementary lighting system 2 is used to provide supplementary lighting to the measuring cavity 11. An ambient light catcher 3 is disposed on one side of the second end of the housing, with its detection end facing the measuring cavity 11. The ambient light catcher 3 is used to detect the ambient light intensity in the measuring cavity 11. Target 6, the target 6 is disposed on one side of the second end of the housing, and the target 6 is provided with multiple color blocks; Image sensor module 4 is disposed on one side of the first end of the housing 1, with the acquisition end of the image sensor module 4 facing the target 6. The image sensor module 4 is used to acquire images of the target 6 through the water in the measuring cavity 11. The self-cleaning mechanism 5 has its movable end disposed in the measuring cavity 11, and the self-cleaning mechanism 5 is used to scrape off dirt from the inner wall of the cavity. The main control processing unit is electrically connected to the supplementary lighting system 2, the ambient light capture device 3, the image sensor module 4, and the self-cleaning mechanism 5.

[0026] Specifically, in the above structure, a measuring cavity 11 with an optical path of approximately 2 cm (adjustable between 1-5 cm) is formed inside the housing 1. A seven-color target 61 is fixedly installed on the lower left side of the measuring cavity 11. Specifically, the target 61 is configured with seven standard color blocks of red, orange, yellow, green, cyan, blue, and purple, arranged in a fixed matrix or concentric rings, with the central color being a circle. An ambient light catcher 2 is located on the lower right side of the measuring cavity 11. An image sensing module 3 is located on the upper left side of the measuring cavity 11. A supplementary lighting system 4 is located on the upper right side of the measuring cavity 11, used for uniformly illuminating the target 6.

[0027] The above-mentioned device can be used with the following multi-parameter water quality monitoring method. First, the multi-parameter water quality monitoring device can be placed in the water body to be tested. Since the measuring chamber 11 is exposed to the water body, the water body flows through and fills the measuring chamber 11. Then, the self-cleaning mechanism 5 is driven to clean the inner wall of the measuring chamber 11, and a preset water body stabilization delay time is waited for. Then, the ambient light capture device 3 acquires the current ambient light intensity and compares it with a preset light signal reference value. Based on the comparison result, the luminous intensity of the supplementary lighting system 2 is adaptively controlled to stabilize the total illuminance of the measuring environment in the measuring chamber 11 near the light signal reference value; or a fixed supplementary lighting is set. The intensity is a fixed supplementary light intensity with a coefficient of 1. This coefficient can be adjusted as the supplementary light attenuates. Under stable total illuminance, multispectral images of the seven-color target 6 are acquired through the image sensor module 4. The acquired images are processed and analyzed to extract image feature values ​​for each color patch of the seven-color target 6. The extracted image feature values ​​are compared with a pre-stored calibration database, which contains standard image feature data for water bodies with different transparency and colors. Based on the comparison results, the current water transparency monitoring value and water color monitoring index are calculated and output using a preset algorithm model.

[0028] Based on the above-mentioned scheme of the present invention, the ambient light catcher 3 can coordinate with the supplementary lighting system 2 to supplement the light to the measuring cavity 11, ensuring the stability of the measuring light source background. Furthermore, the auxiliary multi-color target 6 reduces the acquisition error compared to human vision, and the self-cleaning mechanism 5 ensures the stability of the measuring cavity 11 at all times. Specifically, this invention is the first to apply adaptive multispectral image analysis to water quality monitoring, simultaneously obtaining two key apparent indicators: transparency and water color, demonstrating high functional integration. Furthermore, the present invention has strong anti-interference capabilities. Utilizing the unique "ambient light sensing combined with adaptive supplementary lighting" mechanism, it effectively overcomes measurement errors introduced by changes in ambient light such as day / night, weather, and season, resulting in extremely high data stability. Furthermore, the measurement results of this invention are intuitive and reliable. Using the seven-color target 6 as a medium, the measurement principle closely resembles human visual discrimination methods (such as the sekker disk and lead type methods), and the results are more consistent with human visual perception, making it highly practical. Furthermore, the present invention has a high degree of automation and a long maintenance cycle. The built-in automatic cleaning mechanism significantly reduces the need for manual maintenance, making it suitable for long-term unattended online monitoring. Furthermore, the device of the present invention is small and easy to deploy. Its compact structural design and short optical path of 1-5 cm make the device small in size and low in power consumption, making it easy to integrate into various buoys, monitoring stations or pipelines.

[0029] Specifically, in terms of the device's structure, the supplementary lighting system 2 includes an LED light source 21, a light guide column 22, and a first high-lens surface 23. The LED light source 21 is installed on one side of the first end of the housing and electrically connected to the main control processing unit. The first high-lens surface 23 is disposed on one side of the first end of the inner wall of the measuring cavity 11. The light guide column 22 connects the LED light source 21 and the first high-lens surface 23. Under the control of the main control processing unit, the LED light source 21 emits light in the direction of the second end. Its light is guided to the first high-lens surface 23 by the light guide column 22 and then illuminates the measuring cavity 11, uniformly illuminating the target 6. The first high-lens surface 23 ensures that the LED light source 21 and other structures remain isolated from the measuring cavity 11. The LED light source can be set as a multi-color white LED.

[0030] Furthermore, the ambient light trap 3 includes a photosensor 31 and a second high-lens surface 32. The photosensor 31 is mounted on one side of the second end of the housing and electrically connected to the main control processing unit. The second high-lens surface 32 is disposed on one side of the second end of the inner wall of the measuring cavity 11, with the photosensor 31 facing the second high-lens surface 32. The photosensor 31 in the ambient light trap 3 can be configured as a photodiode, with its end facing the measuring cavity 11, to directly monitor the light intensity in the water and send the corresponding light data to the main control processing unit. The second high-lens surface 32 also serves an isolation effect.

[0031] Furthermore, the image sensor module 4 includes an image sensor 41, an optical lens 42, and a third high-resolution lens 43. The image sensor 41 is mounted on one side of the first end of the housing and electrically connected to the main control processing unit. The third high-resolution lens 43 is disposed on one side of the first end of the inner wall of the measuring cavity 11. The optical lens 42 is connected between the image sensor 41 and the third high-resolution lens 43. The image sensor 41 can be configured as a CMOS sensor. The optical lens 42 directs light guided from one side of the second end into the CMOS sensor to achieve image acquisition, with the acquisition direction facing the target 6. The third high-resolution lens 43 also serves an isolation function.

[0032] Furthermore, the self-cleaning mechanism 5 includes a motor 51, an extension shaft 52, and a cleaning scraper 53. The motor 51 is mounted on one side of the first end of the housing and electrically connected to the main control processing unit. The extension shaft 52 is connected to the output end of the motor 51. The cleaning scraper 53 is fixed to the end of the extension shaft 52. The extension shaft 52 passes through the housing 1, and the cleaning scraper 53 is disposed in the measuring cavity 11. At the beginning of each detection process, the motor 51 operates under the drive of the main control processing unit, driving the extension shaft 52 to rotate, causing the cleaning scraper 53 at the end of the extension shaft 52 to rotate and scrape within the measuring cavity 11. Specifically, the rubber edge of the cleaning scraper 53 can contact and cover the measuring cavity 11, so that the cleaning scraper 53 can scrape the first high lens surface 23, the second high lens surface 32, the third high lens surface 43, and the outer wall of the target 6 in one go during rotation.

[0033] The main control processing unit includes a photosensitive signal processing unit, a supplementary lighting control unit, an image acquisition unit, an image processing unit, a data storage unit, and a digital transmission unit. The digital transmission unit uses the RS-485 communication protocol and connects to an external host computer via a four-core shielded cable. For details, please refer to [link to relevant documentation]. Figure 3 .

[0034] The following example illustrates a specific operating procedure.

[0035] First, initialization and cleaning steps are performed. After the equipment is powered on, it is set to intermittent working mode, which allows for measurement once per hour. At the start of each measurement, the main control unit first starts motor 51 to perform a cleaning cycle, and then waits for 5 seconds to allow the agitated water to return to calm.

[0036] Then, an ambient light adaptation step is performed. The main control processing unit reads the signal from the ambient light catcher 3. Assuming the preset "light signal reference value" corresponds to a digital value of 1000 (calibrated in clear daytime air), and the current reading is 800, it indicates that the ambient light is relatively dim. The main control processing unit then proportionally controls the LED light source 21 of the supplementary lighting system 2 to increase its brightness until the reading of the ambient light catcher 3 stabilizes around 1000. This supplementary lighting and stabilization process lasts for 2 seconds.

[0037] Then, the image acquisition and processing steps are carried out. Within 2 seconds of stable supplementary lighting, the image sensor module 4 is exposed and captures a clear image of the seven-color target 6. After that, the image data is sent to the main control processing unit.

[0038] The subsequent feature extraction and algorithm analysis steps include three parts: transparency calculation, water color determination, and data output. In transparency calculation, the main processing unit calculates the image grayscale variance of the entire target area as the overall contrast. The more turbid the water, the lower the contrast. Simultaneously, the sharpness of each color patch's edges is analyzed (e.g., edge gradients are calculated using the Sobel operator). These feature values ​​(contrast, edge sharpness) are input into a pre-stored data model. This model establishes a regression relationship (such as a polynomial fitting curve) between the feature values ​​and transparency through numerous parallel experiments conducted in the laboratory, comparing the device with transparency values ​​measured using the typesetting method. By querying this curve, the current transparency value can be obtained.

[0039] In water color determination, the main control processing unit analyzes the RGB values ​​of various color patches in the image, especially white and neutral color patches. Due to the coloration of the water (such as greenish or brownish tints), the chromaticity of these color patches will systematically shift. By analyzing the direction and degree of this shift, and combining it with the saturation changes of each color patch, the dominant color of the water can be comprehensively determined, and qualitative indicators such as "yellowish-green" or "brown" or corresponding quantitative chromaticity coordinates can be output.

[0040] In the data output, the processed transparency (unit: cm) and water color code are sent to the host computer or data platform via RS-485 bus, which can be a remote monitoring center / platform, online controller or other portable device.

[0041] Furthermore, the adaptive light compensation algorithm described above can be implemented through the following embodiment. The core of this algorithm lies in constructing a closed-loop feedback control system. By sensing the ambient light intensity in real time and dynamically adjusting the output of the supplementary lighting system, the optical environment of each measurement is "normalized" to a preset, constant reference state. This fundamentally eliminates the interference of ambient light changes on the measurement results, ensuring the consistency and comparability of the data. The specific algorithm flow and mathematical model are as follows.

[0042] Step 1: Perform baseline calibration. Before the device leaves the factory, place the ambient light trap 3 in the air under standard lighting conditions (such as a sunny day without direct sunlight). Activate the ambient light trap 3 and read its output value (denoted as...). This value serves as the reference value for all subsequent optical signals. Simultaneously, under this reference illumination, a "standard image" of the seven-color target 6 is acquired. This is used in subsequent color analysis models.

[0043] Step two involves ambient light sensing and compensation decision-making. First, data acquisition is performed. During each measurement cycle, after cleaning is completed, the current value of the ambient light sensor is read. Then, deviation calculation is performed to determine the difference between the current ambient light and the reference light. Finally, a decision is made, if... ( If a small dead zone threshold is set to prevent minor system fluctuations near the baseline, then the ambient light is considered sufficient, and no supplemental lighting is needed or only minimal supplemental lighting is required. This indicates that the ambient light is too weak and a supplemental lighting system needs to be activated to compensate. In theory, if the ambient light is too strong, this device mainly avoids it through physical structure (such as installing a light shield on the housing 1). The algorithm can be set to not compensate or record alarms.

[0044] Step three involves dynamic supplemental lighting control (PID control principle). This algorithm recommends using a proportional-integral (PI) controller to achieve fast and stable supplemental lighting, avoiding over-adjustment. The proportional term (P) provides a fast response proportional to the deviation. The integral term (I) accumulates historical deviations, eliminates static errors, and ensures precise stabilization at the reference value.

[0045] The control logic pseudocode can be written in Python, and one example is shown below: # Initialization Kp = 0.475 # Scale factor Ki = 0.103 # Integral coefficient integral = 0 setpoint = V_env_ref # During the supplemental lighting process (e.g., once every 100ms) While (lighting is in progress): current_light = Read ambient light catcher value() error = setpoint - current_light # Calculate the integral term (to prevent integral saturation) integral = integral + error * delta_time # Calculate the PID output (PI in this case) output = Kp * error + Ki * integral # Map the output value to the LED's PWM (Pulse Width Modulation) duty cycle pwm_duty_cycle = max(0, min(100, output)) # Limit to 0-100% Set the brightness of the fill light LED (pwm_duty_cycle) # Determine if it is stable: The error is within the allowable range for N consecutive tests. if abs(error) <stable_threshold: stable_count += 1 else: stable_count = 0 if stable_count>= N: break # Exit the supplemental lighting cycle; lighting has stabilized. In the above embodiments, the concept of an "air reference" can be introduced, using the reading "in air, under standard lighting" as a fixed reference rather than a dynamically changing value. This simplifies the system logic and improves long-term stability. Specifically, closed-loop PI control enables adaptive and precise supplementary lighting, rather than simple "if-else" switching control, allowing the system to quickly respond to sudden changes in ambient light and stabilize at the target value. State awareness and delayed triggering ensure consistent optical conditions during image acquisition, allowing image acquisition to be triggered only after supplementary lighting has stabilized.

[0046] Furthermore, the aforementioned multispectral color analysis model based on the seven-color target can be implemented through the following embodiment. In this model, after immersing the fixed seven-color target 6 as an "optical ruler" in water, its imaging quality (contrast, sharpness, color fidelity) will systematically attenuate due to the absorption and scattering effects of the water. This model establishes a mapping relationship from image features to water quality parameters (transparency, color) by quantifying and analyzing this attenuation. The specific algorithm flow and mathematical model are as follows.

[0047] Step one involves image preprocessing and ROI (Region of Interest) localization. First, the image is aligned and cropped. Since the target positions are fixed, the algorithm first locates the precise positions of the seven-color targets in the image. Then, ROI segmentation is performed, dividing the image into independent ROIs corresponding to the seven color blocks (red, orange, yellow, green, cyan, blue, and purple), as well as possible regions with white backgrounds or black borders. Finally, color space conversion is performed, transforming the image from the RGB color space to the HSV color space. The HSV space separates value (luminance) from hue (chroma) and saturation (saturation), which better aligns with human visual perception and is more robust to changes in lighting conditions.

[0048] Step two involves extracting transparency-related features. Transparency is related to the overall attenuation and scattering of light, primarily manifested in a decrease in image contrast and sharpness. In calculating global contrast, the Michell contrast of the entire target region 6 (or the grayscale image) is first calculated. Where σ(I) is the standard deviation of pixel intensity and μ(I) is the mean of pixel intensity, the more turbid the water, the smaller the difference between the target and the background, the smaller the standard deviation of pixel intensity, and the lower the contrast. In calculating edge sharpness, the Sobel or Canny operator is used for edge detection in the boundary region of each color patch. The average gradient magnitude of the extracted edges is calculated. The larger the gradient magnitude, the sharper the edge and the higher the water transparency. In calculating feature fusion, the global contrast and the average edge sharpness of each color patch are used as a set of feature vectors. .

[0049] Step three involves extracting water color-related features. Water color is primarily determined by dissolved substances and phytoplankton, which cause selective light absorption, leading to a shift in the target color. In the chromaticity and saturation analysis, for each color patch's ROI, the median or mean of its H (chromaticity) and S (saturation) components in the HSV space are calculated. This is then compared with a baseline image. The corresponding color block and A comparison is then made. Color difference calculations are then performed, including calculating the chromaticity shift of each color patch. And calculate the saturation decay for each color block. Of particular interest is the chromaticity shift in the white or neutral gray areas, as this most directly reflects the coloration of the water (e.g., yellowish or greenish tints).

[0050] Finally, feature vectors are constructed for each color block. and Composition of feature vectors .

[0051] Furthermore, the water quality parameter calculation model mentioned above can be implemented through the following embodiment. It primarily utilizes a transparency calculation model (regression model). In a laboratory environment, this device and the lead type method were used to simultaneously collect data from a large number of water samples with different transparency levels, resulting in a dataset. Using machine learning algorithms such as Support Vector Machine Regression or Random Forest Regression, a model is trained from feature vectors. To the actual transparency value mapping function The established model can learn the non-linear contributions of different features to transparency, making it more accurate than simple linear regression. Specifically, the water color judgment model (combining classification and quantification) includes the following steps.

[0052] First, a qualitative classification step is performed, based on feature vectors. Especially neutral color blocks A training classifier—the SVM classifier—is established to classify water color into categories such as "colorless," "green," "yellowish-brown," and "blue." Then, a quantitative characterization step is performed; for common colored water bodies (such as algal blooms), a [classifier name missing] is established. Regression relationship with chlorophyll a concentration. Output internationally recognized chromaticity coordinates using the CIE Lab color space. Value, of which The value is biased towards red / green. The value leans towards yellow / blue, and this method is used to quantify the color of water bodies.

[0053] The above embodiments have the following characteristics. First, by leveraging multi-feature fusion, it no longer relies on a single image feature but integrates global contrast, local edge sharpness, and chromaticity / saturation shifts of multiple color patches to form a more comprehensive description of the optical properties of the water body. Second, by employing data-driven modeling based on machine learning, a nonlinear, high-precision inversion model is established through extensive data comparison with standard methods (typesetting method), rather than simple empirical formulas. Finally, by utilizing the "target-water" interaction analysis, by analyzing the different attenuation modes of seven different wavelength color patches, indirect measurement of the absorption spectral characteristics of the water body is achieved, providing a physical basis for simultaneously inverting transparency and color.

[0054] This invention relates to a monitoring device that is an intelligent system integrating optics, mechanics, electronics, and algorithms. Its system composition can be divided into three main modules: a field data acquisition unit, a main control and processing unit, and a remote interaction unit, such as... Figure 3 This can be clearly demonstrated. The specific working principle and signal flow are as follows:

[0055] During the startup and initialization phase, the device is connected to a power source via an extension cable, and the main control processing unit is powered on. The system loads preset parameters from the storage unit, such as the operating mode (intermittent / continuous), cleaning cycle, optical signal reference value, hysteresis time, and supplementary lighting mode.

[0056] During the automatic cleaning and water stabilization phase, the data algorithm and execution logic of the main control processing unit issue commands to drive the motor 51 of the automatic cleaning mechanism. The motor 51, via the extension shaft 52, drives the cleaning scraper 53 to perform a single scraping cleaning of the surface of the image sensing module, the supplementary lighting system 2, the optical window of the ambient light capture device 3, and the seven-color target 6, to remove biological attachments and contaminants. After cleaning is completed, the system enters a preset hysteresis waiting period (e.g., 5 seconds) to allow the water disturbed by cleaning to stabilize, ensuring the representativeness of the measurement environment.

[0057] In the ambient light sensing and adaptive supplemental lighting stage (core closed-loop control), the analog signal from the ambient light catcher 3 is read by the photosensitive signal processing unit and converted into a digital signal. The data algorithm and execution logic compare the current light intensity value with a pre-stored light signal reference value (a standard value calibrated in air). During decision-making and execution, the algorithm calculates the required light compensation amount. Based on this calculation result, the supplemental lighting control unit precisely controls the LED luminous intensity of the supplemental lighting system 2 through a PWM (pulse width modulation) signal. This process is a dynamic closed-loop feedback; the ambient light catcher 3 continuously monitors, and the supplemental lighting system 2 continuously fine-tunes until the total illuminance within the measurement cavity 11 stabilizes near the light signal reference value and is maintained for a preset duration (e.g., 2 seconds). This step ensures that each measurement is performed under standardized lighting conditions, which is crucial for eliminating environmental interference.

[0058] During the image acquisition and feature extraction stage, within a stable illumination period, the data algorithm and execution logic trigger the image acquisition unit. The image sensing module takes a picture of the seven-color target 6 below, acquiring a clear multispectral image under standard illumination. This includes the image processing and analysis unit performing the following processing on the acquired image to accurately identify and segment each color patch region in the seven-color target 6.

[0059] Feature extraction is performed by calculating the contrast of the entire image using transparency features and analyzing the sharpness (gradient) of the edges of each color block. Color features are also analyzed by converting the image to color spaces such as HSV, analyzing the chroma and saturation of each color block, and comparing it with a reference image in the storage unit to calculate color difference and saturation attenuation.

[0060] In the water quality parameter calculation and output stage, the data algorithm and execution logic call the calibration data and algorithm model in the storage unit. Specifically, in transparency calculation, the extracted transparency features (contrast, edge sharpness) are input into a regression model (such as SVR or random forest) built using a large amount of experimental data (e.g., related to the typesetting method), directly calculating the transparency value in units of length. In water color determination, the extracted color features (color difference, saturation decay) are input into a classification or regression model, outputting the water body's color category (e.g., "yellow-green") or quantitative chromaticity coordinates. Finally, the calculated transparency values ​​and color indicators are encapsulated into data frames and uploaded to a remote monitoring center or cloud platform via a digital transmission unit (RS-485 bus, following standard industrial protocols such as Modbus).

[0061] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the scope of the patent application of the present invention.

[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0063] Although this application has been described by way of examples, those skilled in the art will know that this application has many modifications and variations without departing from the spirit of this application, and it is intended that the appended embodiments include these modifications and variations without departing from this application.

Claims

1. A multi-parameter water quality monitoring device, characterized in that, The water quality multi-parameter monitoring device includes: A housing that encloses a measuring cavity having a first end and a second end disposed opposite to each other; A supplementary lighting system is provided, which is disposed on one side of the first end of the housing, with the light source end of the supplementary lighting system facing the measuring cavity, and is used to provide supplementary lighting to the measuring cavity; An ambient light trap is provided, which is disposed on one side of the second end of the housing, with the detection end of the ambient light trap facing the measuring cavity. The ambient light trap is used to detect the ambient light intensity in the measuring cavity. A target, wherein the target is disposed on one side of the second end of the housing, and the target is provided with multiple color blocks; An image sensor module is disposed on one side of the first end of the housing, with the acquisition end of the image sensor module facing the target. The image sensor module is used to acquire images of the target through the water in the measurement cavity. The self-cleaning mechanism has its movable end disposed in the measuring cavity, and is used to scrape off dirt from the inner wall of the cavity. The main control processing unit is electrically connected to the supplementary lighting system, the ambient light capture device, the image sensor module, and the self-cleaning mechanism.

2. The water quality multi-parameter monitoring device according to claim 1, characterized in that: The supplementary lighting system includes an LED light source, a light guide column, and a first high-resolution lens. The LED light source is installed on one side of the first end of the housing and is electrically connected to the main control processing unit. The first high-resolution lens is disposed on one side of the first end of the inner wall of the measuring cavity. The light guide column is connected between the LED light source and the first high-resolution lens.

3. The water quality multi-parameter monitoring device according to claim 1, characterized in that: The ambient light catcher includes a photosensitive sensor and a second high-resolution lens. The photosensitive sensor is mounted on one side of the second end of the housing and is electrically connected to the main control processing unit. The second high-resolution lens is disposed on one side of the second end of the inner wall of the measuring cavity, and the photosensitive sensor is positioned toward the second high-resolution lens.

4. The water quality multi-parameter monitoring device according to claim 1, characterized in that: The target is set with seven standard color blocks of red, orange, yellow, green, cyan, blue and purple. The seven color blocks are arranged in a fixed matrix or in concentric circles.

5. The water quality multi-parameter monitoring device according to claim 1, characterized in that: The image sensor module includes an image sensor, an optical lens, and a third high-resolution lens. The image sensor is mounted on one side of the first end of the housing and is electrically connected to the main control processing unit. The third high-resolution lens is disposed on one side of the first end of the inner wall of the measuring cavity. The optical lens is connected between the image sensor and the third high-resolution lens.

6. The water quality multi-parameter monitoring device according to claim 1, characterized in that: The self-cleaning mechanism includes a motor, an extension shaft, and a cleaning scraper. The motor is installed on one side of the first end of the housing and is electrically connected to the main control processing unit. The extension shaft is connected to the output end of the motor. The cleaning scraper is fixed to the end of the extension shaft. The extension shaft passes through the housing and the cleaning scraper is disposed in the measuring cavity.

7. A multi-parameter water quality monitoring device, characterized in that, It includes a water quality multi-parameter monitoring device as described in any one of claims 1-6, as well as an extension cable, an external power supply, and a control unit for powering the water quality multi-parameter monitoring device and enabling data communication.

8. A method for monitoring multiple water quality parameters, applied to the multi-parameter water quality monitoring device as described in any one of claims 1-6, characterized in that, The method includes the following steps: After the self-cleaning mechanism cleans the inner wall of the measuring chamber, wait for the preset water stabilization delay time. The current ambient light intensity is obtained through an ambient light sensor and compared with a preset light signal reference value; Based on the comparison results, the luminous intensity of the supplementary lighting system is adaptively controlled to stabilize the total illuminance of the measurement environment in the measurement cavity near the optical signal reference value; or a fixed supplementary lighting intensity is set with a coefficient of 1, and the coefficient can be adjusted when the supplementary lighting attenuates subsequently. Under stable total illuminance, multispectral images of a seven-color target are acquired through an image sensor module; The acquired images are processed and analyzed to extract the image feature values ​​of each color block of the seven-color target; The extracted image feature values ​​are compared with a pre-stored calibration database, which contains standard image feature data for water bodies with different transparency and different colors. Based on the comparison results, the system calculates and outputs the current water body transparency monitoring value and water body color monitoring index using a preset algorithm model.

9. The method for monitoring multiple water quality parameters according to claim 8, characterized in that: In the step "processing and analyzing the acquired image to extract the image feature values ​​of each color block of the seven-color target", the image feature values ​​include, but are not limited to, the overall image contrast of the entire target area, the sharpness of the edges of each color block, and the chromaticity and saturation values ​​of each color block in the RGB or HSV color space; in the step "adaptively controlling the luminous intensity of the supplementary lighting system based on the comparison results to stabilize the total illuminance of the measurement environment in the measurement cavity near the light signal reference value", the adaptive light compensation algorithm adopts a closed-loop feedback mechanism including proportional-integral control.

10. The method for monitoring multiple parameters of water quality according to claim 9, characterized in that: In the step "comparing the extracted image feature values ​​with the pre-stored calibration database, which contains standard image feature data for water bodies with different transparency and different water body colors", the pre-stored calibration database is established by correlating and fitting the actual transparency values ​​measured by the lead type method with the image feature values ​​collected under the corresponding conditions. The algorithm model includes a multispectral color analysis model, which is set to feature extraction and regression judgment based on the edge sharpness of the target image and the color shift of multiple color blocks.

Images