A multi-element data-based swimming pool water quality monitoring system and method

Abstract

The application discloses a kind of based on multivariate data's swimming pool water quality supervision system, including main server, first-level server, second-level server, first-level execution unit, second-level execution unit.A kind of based on multivariate data's swimming pool water quality supervision method, comprising the following steps: step 1, a plurality of groups of microbial image data of multiple areas are collected;Step 2, the microbial image of each area is processed and analyzed, the microbial distribution amount is obtained, the microbial distribution amount is compared, whether it is judged to exceed set threshold value;Step 3, the speed of human movement is detected, and the speed is analyzed, and the time node of the speed dramatic change is screened out;Step 4, continuously shooting obtains the image data in pool in a certain period of time.The cleanliness of swimming pool is monitored by the method of microbial image acquisition, the precision of swimming pool can be monitored, and alarm is sent, reminding user to clean and change water in time.

Description

[0001] The primary server receives multiple sets of microbial images collected by microbial image collectors in multiple regions, processes the microbial images of each region, obtains the microbial distribution, determines whether the distribution exceeds a predetermined threshold, and feeds back the determination result to the main server.

[0002] The first-level execution unit receives control signals sent by the main server and issues warnings;

[0003] The secondary server receives multiple sets of human movement speed data collected by speed detectors in multiple areas, saves each set of human movement speed data in chronological order, filters out time nodes with drastic speed changes, and sends signals to the main server.

[0004] The secondary execution unit starts after receiving the control signal sent by the main server, continuously filming the entire pool to obtain data on the pool's condition over a period of time, and then saves the image data.

[0005] Furthermore, the microbial pattern collector includes a miniature camera, which is covered with a transparent microbial attachment cover, and a black background plate is placed in front of the transparent microbial attachment cover.

[0006] Furthermore, the secondary execution unit also includes multiple thermal infrared cameras, which are spaced apart in each lane and start up after receiving a signal from the main server to capture multiple frames of thermal infrared images and send the captured thermal infrared images to the secondary server in a time sequence.

[0007] A method for monitoring swimming pool water quality based on multivariate data, characterized by the following steps:

[0008] Step 1: Collect multiple sets of microbial image data from multiple regions;

[0009] Step 2: Process and analyze the microbial images of each region to obtain the microbial distribution, compare the microbial distribution, and determine whether it exceeds the set threshold.

[0010] Step 3: Detect the human body's movement speed, analyze the speed, and filter out time points where the speed changes drastically;

[0011] Step 4: Continuously capture images of the pool during a certain period of time or (and) continuously capture multiple frames of thermal infrared images of the pool, and match and confirm the suspect based on the time nodes.

[0012] Furthermore, step 2 also includes step 2A: performing binarization processing on the microbial images collected by each collector in sequence, setting a grayscale threshold, and calculating the image area of ​​bacteria and impurities in each frame image by comparing the grayscale thresholds, thereby calculating the average amount of bacteria and impurities in different areas collected by each collector, using the average to calculate the average total proportion, and judging whether the microbial distribution exceeds the threshold based on the average total proportion; step 2B: extracting several images and superimposing them to produce a new image, marking the position on the new image, and calculating the grayscale value at the marked position to obtain the blur.

[0013] Furthermore, in step 2B, the number of images extracted is 2-3.

[0014] Furthermore, in step 2B, new images are generated after overlay, producing at least 3-5 new images.

[0015] Furthermore, step 3 also includes step 3A, deploying multiple pairs of speed detectors at the beginning and end of each lane in the pool to detect the swimmer's swimming speed in real time and saving the data in chronological order; and step 3B, sending a signal when a sudden change in the swimmer's speed is detected.

[0016] Furthermore, step 4 includes step 4A, continuously capturing multiple frames of thermal infrared images; and step 4B, analyzing the contours of the thermal infrared images to determine whether predetermined data in the thermal infrared images have undergone significant changes, and locking down the suspect image based on the change results.

[0017] Furthermore, the predetermined data includes height, hip width, and leg spacing.

[0018] The beneficial effects of this invention are:

[0019] The swimming pool water quality monitoring system and method based on multi-source data provided by this invention uses microbial image acquisition to monitor the cleanliness of the swimming pool. This enables precise monitoring and alerts, reminding users to clean and change the water promptly. Cleanliness monitoring employs two indicators: bacterial count and ambiguity, improving detection accuracy. A speed detector detects swimmers' movement speed, identifying whether they are in a normal state. An alert is issued for abnormal states, triggering a monitoring system or thermal infrared camera to capture images of the swimmer. By combining video or thermal infrared images with the time period of speed change, suspected swimmers can be identified, thus achieving pool monitoring and ensuring the cleanliness of the pool water. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of a swimming pool water quality monitoring system based on multi-source data according to the present invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] Please refer to Figure 1 The present invention provides a swimming pool water quality monitoring system based on multi-source data, comprising a main server, a primary server, a primary execution unit, a secondary server, and a secondary execution unit.

[0023] The primary server receives multiple sets of microbial images collected by microbial image collectors in multiple regions, processes the microbial images of each region, obtains the distribution of microorganisms, determines whether the distribution exceeds a predetermined threshold, and feeds back the determination result to the main server.

[0024] Specifically, the microbial pattern sampler is a miniature camera. The camera's surface is covered with a transparent microbial attachment shield. This shield has multiple crisscrossing scratches, 1-5 μm deep, which can be created as needed. These scratches are for bacterial attachment, and the inner walls of the scratches are polished to maintain the shield's transparency and thus the accuracy of the microbial pattern. A black background plate is also installed in front of the transparent microbial attachment shield to filter out background interference and improve image processing accuracy.

[0025] In practice, six microbial image collectors are deployed on the walls of the swimming pool and labeled as Microbial Collector 1, Microbial Collector 2, Microbial Collector 3, Microbial Collector 4, and Microbial Collector 5, respectively. Each microbial image collector collects 5-10 frames of microbial images at the same time. Each microbial image collector sends the collected microbial images to the primary server, which then processes them.

[0026] The primary server receives image information from microbial collectors 1, 2, 3, 4, and 5, and marks the image information. For example, the first frame image collected by the first microbial collector is marked as 11, the first frame image collected by the first microbial collector is marked as 12, and so on. Please refer to Table 1 below for details. Table 1 takes the example of each microbial collector collecting 5 frames of images.

[0027] First, the first server performs binarization on the image, resulting in a black and white image. Second, the area of ​​the image containing bacteria and impurities is calculated. In the image, the grayscale of areas without bacteria and impurities is lower than that of areas with bacteria and impurities. Only a grayscale threshold needs to be set for area calculation. The calculated area is compared with the total image area to determine the area percentage. After removing the maximum and minimum values, the average is calculated to obtain the average area percentage for any microbial collector. Averaging multiple average percentages yields the total average area percentage. In this embodiment, the total average percentage is 61.6%, representing that the adhesion of bacteria and dust has reached approximately 61.6%.

[0028]

[0029] Table 1

[0030] To further ensure the accuracy of the detection results, it is necessary to further identify and calculate the blurriness of the images, and then determine whether the blurriness also exceeds the set standard threshold. If the blurriness also exceeds the threshold, it can be confirmed that the cleanliness of the pool water is poor, and the pool needs to be cleaned and the water changed. The specific operation of blurriness identification and calculation is as follows: First, the first server randomly selects 2-3 images from the above images, which were collected by different collectors. Then, the above 2-3 images are superimposed to generate a new image, which is labeled as K1. This process is repeated, as shown in Table 2, to generate at least 3-5 new images. The positions are marked on the new images, and the gray value at the position is measured. Gray values ​​≤100 and ≥160 are removed, and the average value of the remaining gray values ​​is calculated. The average value is 136.5.

[0031]

[0032] Table 2

[0033] According to the test results, the average total percentage is 61.6%, and the average gray value is 136.5. The water quality in the pool is poor, and the pool needs to be cleaned and the water replaced.

[0034] The primary server sends the detection results to the main server, which then sends control signals to the primary execution unit. The primary execution unit is a buzzer or LED light used for reminders. When staff hear the buzzer or see the LED light illuminate, they can begin the cleaning process.

[0035] The secondary server receives multiple sets of human movement speed data collected by speed detectors in multiple areas, analyzes each set of human movement speed data according to time sequence, filters out time nodes with drastic speed changes, and sends signals to the main server.

[0036] Specifically, multiple pairs of speed detectors are deployed at both ends of each lane in the pool. The pair of detectors installed in the first lane is named Q1 and Q1′, the pair of detectors installed in the second lane is named Q2 and Q2′, and so on, as shown in Table 3 below. Each speed detector in each group detects the swimmer's swimming speed and sends the detected speed to the secondary server in time sequence. The secondary server receives and saves the above data and performs real-time analysis on the data.

[0037]

[0038]

[0039]

[0040] Table 3

[0041] Based on the speed changes, swimmer A, detected in Q1, maintained a swimming speed of approximately 2 m / s, suggesting an adult, and their speed was continuously increasing, thus eliminating them as a suspect. Swimmer B, detected in Q1′, maintained a speed of approximately 1.5 m / s; although their speed changed, the change was minor, eliminating them as a suspect. Swimmer C, detected in Q2, experienced a sudden drop in speed, which lasted for about 10 seconds before suddenly increasing, making them a suspect. Swimmer D, detected in Q2′, experienced a sudden and continuous drop in speed; although their speed changed, the continuous drop lasted for more than 14 seconds, suggesting they were likely resting, thus eliminating them as a suspect. Swimmer E, detected in Q3, maintained a swimming speed of approximately 3 m / s, suggesting an adult, suggesting an adult, and their speed was approximately 1.5 m / s, suggesting an adult, and their speed was continuously increasing, thus eliminating them as a suspect. The following are examples of swimmers: A good adult swimmer (Q3') can be ruled out; the swimmer in Q4's swimmer G maintained a speed of approximately 2.1 m / s, and although the speed changed, the change was small, so they can be ruled out; the swimmer in Q4's swimmer G maintained a speed of approximately 2.7 m / s, and although the speed changed continuously, the change was small and there was no significant and sustained decrease, so they can be ruled out; the swimmer in Q4's swimmer H's speed decreased, but the decrease was too short, so they can be ruled out; the swimmer in Q5's swimmer R's speed fluctuated abruptly and did not decrease continuously, so they can be ruled out; the swimmer in Q5's swimmer G maintained a speed of approximately 2.0 m / s, and although the speed changed, the change was small, so they can be ruled out. As can be seen from the above experiment, the suspect is swimmer C detected by Q2, based on swimming speed. This swimmer's speed gradually decreased during the period from 16:00:02 to 16:00:12, a period of about 8 seconds, and suddenly increased at 16:00:12, which is consistent with normal logic and makes him a strong suspect.

[0042] After receiving the signal from the secondary server, the main server sends a control signal to the secondary execution unit, which then starts. The secondary execution unit is a camera unit mounted on the ceiling of the pool room, covering the entire pool. Upon receiving the control signal from the main server, the camera unit starts and continuously records the pool conditions for 10 seconds to acquire image data of the pool during that time period, which is then saved. In this embodiment, the pool conditions from 16:00:02 to 16:00:12 are recorded and the captured image data is saved.

[0043] The secondary execution unit also includes multiple thermal infrared cameras, which are spaced apart at the bottom of each swimming lane. These cameras activate upon receiving a control signal from the main server. The main server then determines which thermal infrared camera in the appropriate lane to activate based on the signal from the secondary server, while the remaining cameras remain in sleep mode to conserve power. After activation, the thermal infrared cameras continuously capture multiple frames of thermal infrared images and send them to the secondary server in chronological order. The secondary server reads each frame sequentially, identifies the human body contours within the images, and filters out suspicious images based on changes in the contours, sending these suspicious images back to the main server. For example, if multiple frames of images of swimmer A and swimmer B are captured, the secondary server first identifies each frame. The identification results are shown in Table 4. In Table 4, the intercrotch distance of swimmer A decreases rapidly, indicating a rapid increase in thermal image intensity, making him a suspicious swimmer. Swimmer B, however, shows less change and is not considered a suspicious swimmer.

[0044] Height (cm) Hip width (cm) Leg spacing (cm) Height (cm) Hip width (cm) Leg spacing (cm) A1 180 36 4 B1 155 30 3 A2 179.1 35.7 4 B2 154.5 31 2.8 A3 180 36 2 B3 155 30.2 2.5 A4 180 36 1 B4 155 29.7 2.7 A5 180 36.3 0 B5 155 30 3 A6 180.2 36 3 B6 155 30.1 3

[0045] Table 4

[0046] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A swimming pool water quality monitoring system based on multi-source data, characterized in that: include The master server is used to receive result signals from the primary server and the secondary server, and to start the corresponding primary execution unit and the secondary execution unit to execute the corresponding instructions based on the result signals. The primary server receives multiple sets of microbial images collected by microbial image collectors in multiple regions, processes the microbial images of each region, obtains the microbial distribution, determines whether the distribution exceeds a predetermined threshold, and feeds back the determination result to the main server. The first-level execution unit receives control signals sent by the main server and issues warnings; The secondary server receives multiple sets of human movement speed data collected by speed detectors in multiple areas, saves each set of human movement speed data in chronological order, filters out time nodes with drastic speed changes, identifies whether the swimmer is in a normal state, and sends a signal to the main server to issue an early warning based on abnormal states. The secondary execution unit includes multiple thermal infrared cameras. After receiving the control signal sent by the main server, the thermal infrared cameras start to continuously film the entire pool, capturing multiple frames of thermal infrared images. The captured thermal infrared images are then sent to the secondary server in sequence. The secondary server reads each frame of the image in turn and identifies the human body outline in the image. Based on the video image or thermal infrared image combined with the time period of speed change, the suspect can be identified. The swimming pool water quality monitoring system based on multi-source data monitors water quality using the following steps: Step 1: Collect multiple sets of microbial image data from multiple regions; Step 2: Process and analyze the microbial images of each area to obtain the microbial distribution. Compare the microbial distribution to determine whether it exceeds the set threshold, and issue a cleaning reminder if it exceeds the threshold. Specifically, step 2A involves performing binarization processing on the microbial images collected by each collector, setting a grayscale threshold, and using the comparison of the grayscale threshold to calculate the image area of ​​bacteria and impurities in each frame of the image. Then, the average amount of bacteria and impurities in different areas collected by each collector is calculated, and the average total proportion is obtained using the average total proportion. Based on the average total proportion, it can be determined whether the distribution of microorganisms exceeds the threshold. Step 2B: Extract several images and overlay them to produce a new image. Mark the positions on the new image and calculate the gray values ​​at the marked positions to obtain the blur level. Step 3: Detect the human body's movement speed, analyze the speed, and filter out time points where the speed changes drastically; Specifically, step 3A involves placing multiple pairs of speed detectors at the beginning and end of each lane in the pool to detect the swimmers' swimming speed in real time and saving the data in chronological order. Step 3B: Send a signal when a sudden change in the swimmer's speed is detected; Step 4: Continuously capture images of the pool during a certain period of time or / and continuously capture multiple frames of thermal infrared images of the pool, and match and confirm the suspect based on the time nodes; Specifically, this includes step 4A: continuously capturing multiple frames of thermal infrared images; Step 4B: Analyze the contour of the thermal infrared image, determine whether the predetermined data in the thermal infrared image has changed significantly, and lock the suspect image based on the change results; the predetermined data are height, hip width, and leg spacing.

2. The swimming pool water quality monitoring system based on multi-source data according to claim 1, characterized in that: The microbial pattern collector includes a miniature camera, which is covered with a transparent microbial attachment cover, and a black background plate is placed in front of the transparent microbial attachment cover.

3. The swimming pool water quality monitoring system based on multi-source data according to claim 2, characterized in that: The thermal infrared cameras are spaced apart in each swimming lane and are activated upon receiving a signal from the main server.

4. A method for monitoring swimming pool water quality based on multivariate data, characterized in that: Includes the following steps: Step 1: Collect multiple sets of microbial image data from multiple regions; Step 2: Process and analyze the microbial images of each area to obtain the microbial distribution. Compare the microbial distribution to determine whether it exceeds the set threshold, and issue a cleaning reminder if it exceeds the threshold. Specifically, step 2A involves performing binarization processing on the microbial images collected by each collector, setting a grayscale threshold, and using the comparison of the grayscale threshold to calculate the image area of ​​bacteria and impurities in each frame of the image. Then, the average amount of bacteria and impurities in different areas collected by each collector is calculated, and the average total proportion is obtained using the average total proportion. Based on the average total proportion, it can be determined whether the distribution of microorganisms exceeds the threshold. Step 2B: Extract several images and overlay them to produce a new image. Mark the positions on the new image and calculate the gray values ​​at the marked positions to obtain the blur level. Step 3: Detect the human body's movement speed, analyze the speed, and filter out time points where the speed changes drastically; Specifically, step 3A involves placing multiple pairs of speed detectors at the beginning and end of each lane in the pool to detect the swimmers' swimming speed in real time and saving the data in chronological order. Step 3B: Send a signal when a sudden change in the swimmer's speed is detected; Step 4: Continuously capture images of the pool during a certain period of time or / and continuously capture multiple frames of thermal infrared images of the pool, and match and confirm the suspect based on the time nodes; Specifically, this includes step 4A: continuously capturing multiple frames of thermal infrared images; Step 4B: Analyze the contour of the thermal infrared image, determine whether the predetermined data in the thermal infrared image has changed significantly, and lock the suspect image based on the change results; the predetermined data are height, hip width, and leg spacing.

5. The swimming pool water quality monitoring method based on multi-source data according to claim 4, characterized in that: In step 2B, the number of images extracted is 2-3.

6. The swimming pool water quality monitoring method based on multi-source data according to claim 4, characterized in that: In step 2B, new images are generated by overlaying, producing at least 3-5 new images.

Images