TW201329434A - Water turbidity detection system and the method thereof - Google Patents

Abstract

A water turbidity detection system for detecting turbidity of water under test (WUT), including a light emitter, an image capturing device, and a processor, is illustrated. The light emitter emits coherent light to the WUT. The image capturing device captures a plurality of reflection picture shown on the surface of the WUT. Each reflection picture includes a blend image resulted from the coherent light confronting with suspending material of the WUT. The processor receives the reflection pictures and defines a scattering block of each picture based on a brightness threshold. The processor calculates area parameters of the scattering blocks, and filers the scattering block via a filtering threshold to generate a statistical value. The statistical value is compared with a turbidity table to determine the turbidity of the WUT, wherein the greater the statistical value the higher the turbidity of the WUT.

Description

Water quality testing system and method thereof

The invention relates to a detection system and a method thereof, and a calibration method of the detection system, and in particular to a water quality detection system and method thereof for detecting water source turbidity, and a calibration method for the water quality detection system.

Nowadays, some hydropower projects still need to generate electricity from the mountain water diversion channel. For the control of mountain water gates, it is often necessary to determine the turbidity of the raw water on the site, or to determine the opening and closing of the gate of the dam by downstream sampling analysis, such as spectrum analyzer. Measurement method, aqua spectrum analysis method or turbidity mapping method.

However, due to the dangerous terrain on the mountain, the method of judging the turbidity by the manpower site is not only vulnerable to the safety of the surveying personnel, but also unable to reflect the state of the raw water turbidity of the mountain in the first time. Once the water is too turbid, it will affect the quality of the cooling water and power water, which will cause the safety of power generation and damage to the equipment of the generator set, causing great losses. Most of the existing detection technologies have high cost, need to clean and calibrate equipment frequently, or inconvenience in collecting samples. Therefore, how to design a raw water turbidity detection system to instantly reflect the upstream water quality to complement the aforementioned defects will be an important issue.

Embodiments of the present invention provide a water quality detecting system for detecting turbidity of a water source under test, the system comprising an illuminator, an image capturing device, and a processor. The illuminator is configured to emit the same dimming light from the water source under test to the water source under test, and the image capturing device continuously captures a plurality of water surface images of the surface of the tested water source. Each surface image includes a gradation image that is generated by the dimming light striking the suspended particles in the water source under test to the surface of the water source under test. The brightness of the gradation image is enhanced from the edge to the center. The processor receives the water surface image and defines a scattering block for each water surface image according to the brightness threshold, wherein the scattering block corresponds to a high brightness area of the water surface image. The processor separately calculates an area parameter of the scattering block, and compares the area parameter according to the screening value to filter the scattering block, and obtains the parameter statistical value according to the area parameter of the scattering block retained after the screening. Then, according to the turbidity comparison table, the parameter statistics are compared to determine the turbidity of the tested water source, wherein the larger the parameter statistical value, the higher the turbidity of the tested water source.

In addition, the embodiment of the invention further provides a water quality detecting method, which uses an illuminator, an image capturing device and a processor to detect the turbidity of the water source under test. The method includes: emitting, by the illuminator, the same dimming light from the surface of the tested water source to the water source under test, and then continuously capturing a plurality of water surface images of the surface of the tested water source by the image capturing device. Each of the water surface images includes a gradation image generated by the same dimming light that collides with the suspended particles in the water source under test to generate a surface of the water source, and the brightness of the gradation image increases from the edge to the center. Then, the water surface image is received, and a scattering block of each water surface image is defined according to at least one brightness threshold, and the scattering block corresponds to a high brightness area of the water surface image. Then, the area parameters of the scattering block are calculated separately, and the scattering blocks are screened according to the screening values. Then, according to the area parameter of the scattering block retained after the screening, the parameter statistical value is obtained, and finally the parameter statistical value and the turbidity comparison table are compared to determine the turbidity of the tested water source, wherein the larger the parameter statistical value, the tested water source The higher the turbidity.

In addition, the embodiment of the invention further provides a calibration method for the water quality detection system for maintaining the correctness of the detection of the water quality detection system.

In summary, the water quality detecting system and the method provided by the embodiments of the present invention can be used by the water quality monitoring personnel to monitor and detect the water quality changes at the tested water source in real time without the need to collect the water sample in person, so as to facilitate the water quality. The monitoring personnel can respond immediately when the water turbidity changes, so as to protect the various electromechanical equipments at the tested water source and avoid damage caused by sediment deposition in the tested water source.

[Water Quality Testing System Example]

Please refer to FIG. 1. FIG. 1 is a block diagram of an embodiment of a water quality detecting system provided by the present invention. The water quality detecting system 1 of the present embodiment includes an illuminator 10, an image capturing device 12, a processor 14, and a memory unit 16. The illuminator 10 and the image capturing device 12 are erected above the surface of the water source 2 under test. The illuminator 10 and image capture device 12 are particularly adapted to be placed in locations where there is little or no environmental light source interference, such as a waterway beam well of a reservoir. The illuminator 10, the image capturing device 12, and the memory unit 16 are connected to the processor 12, respectively. The processor 12 and the memory unit 16 can be disposed at a remote control box or a control center. The processor 12 is configured to control the illuminating power of the illuminator 10 and receive the image captured by the image capturing device 12 and The analysis process is performed to determine whether the water of the tested water source 2 is clear or turbid. Therefore, the water quality monitoring personnel do not need to sample the water intake port of the water source 2 to detect or visually check the water quality, and the image 14 captured by the image capturing device 12 can be analyzed and calculated by the processor 14 from the far end. The turbidity of the water source 2 to be tested is determined quickly and conveniently.

The illuminator 10 is configured to emit the same dimming light, such as laser light, from the measured water source 2 from the upper side of the water source 2 under test. The same dimming has the same characteristics of direction, phase and frequency, so the same dimming does not interfere with each other, avoiding the problem that the light source diffuses and the light is weakened as the emission distance becomes longer. The illuminator 10 can be a single beam or multi-beam array of laser light emitting devices, such as a laser pen. In particular, the illuminator 10 in this embodiment is disposed at an angle that is perpendicular or nearly perpendicular to the horizontal plane of the water source 2 under test to facilitate the dimming of light into the water surface at a vertical or near vertical angle.

The image capturing device 12 is configured to capture an image of the region where the illuminator 10 emits the same dimming light to the water source 2 to be tested. Therefore, the image capturing device 12 and the illuminator 10 can be disposed adjacent to each other to facilitate capturing the image. The image capturing device 12 can be specifically a device such as a camera or a camera that can continuously capture still photos or moving images.

Since the illuminator 10 emits the same dimming light to the water source 2 under test, if the water source 2 under test contains suspended particles (such as sediment or dust), the light beam will hit the suspended particles on the way through the water source. When the beam encounters suspended particles, except for a small part of the light source that can continue to penetrate downward by diffraction, most of the light source will be substantially concentric according to the scattered light energy on the surface of the water source 2 under test due to the scattering effect. A gradation image having gradual color saturation and brightness is formed, and its color saturation and brightness are gradually increased from the periphery of the concentric circle toward the center. Under the same dimming energy irradiation, according to the amount of suspended particles contained in the tested water source 2 (that is, the turbidity of the water source is different), the degree of gradual diffusion of the gradation image reflected will also be different. The schematic diagrams of the water surface images 20a to 20c shown in Fig. 2 show the degree of gradual diffusion of the gradation images of the water surface (200a, 200b, and 200c, respectively) when the water quality is relatively clear, moderately turbid, and severely turbid. Will gradually converge and concentrate the light energy. In other words, as the water quality becomes more turbid, the radius of the entire concentric circle decreases, but the area of the same brightness in the concentric circles expands due to the concentration of light energy.

Continuing to refer to FIG. 2, as described above, the color saturation and brightness of the gradation image are increased from the outside to the inside, so that the central region of each of the gradation images 200a to 200c is high due to the concentration of light scattering energy. The luminance regions (202a, 202b, and 202c, respectively) exhibit a saturated, homochromatic block even at the center. The saturated color-matching block is because the scattering energy concentrated in the central region is too high under the same dimming light emitted by the illuminator 10, and the photo-sensing device 12 has a limited photo-sensing capability, and the true brightness of the central portion cannot be recognized. The layer thus forms a saturated, homochromatic block of uniform brightness and color saturation. It appears from the image screen that the high-brightness area appears as an image that is centered on a saturated, homochromatic block and is substantially circular. As mentioned above, the block size in the high-brightness region will vary with the turbidity of the water quality. When the suspended particles of the tested water source 2 are more, the turbidity of the water is higher, and the homogenized light enters the water source 2 after the test. It is easy to cause scattering, and the higher the scattering energy, the larger the area of high luminance is formed, and the range of the high-luminance regions 202a to 202c in FIG. 2A increases with the corresponding water quality from clear to severe turbidity.

Referring back to FIG. 1 , the image capturing device 12 is configured to capture a water surface image containing the high brightness region and transmit the water surface image to the processor 14 one by one. After the processor 14 processes and analyzes the water surface image, and filters out the noise in the image, the range of the high-luminance region in each water surface image is defined as a scattering block, and then according to a plurality of consecutively captured water surface images. The size of the scattering block is judged to determine whether the amount of suspended particles contained in the water source 2 to be measured is excessive or small, thereby confirming the turbidity of the water source 2 to be tested. In other words, when the processor 14 calculates the range of the scattering block according to the same condition, the water quality of the water source 2 under test at the time of capturing the water surface image is turbid.

The processor 14 disposed in the remote control center or the control box sequentially receives the water surface image through wired or wireless transmission, in order to separate the scattering blocks generated by the reflection and the same dimming (that is, the system is to be defined to determine The range of the turbidity block and the range of noise (eg, the reflected water image generated by the flowing water source impacting the waterway wall), the processor 14 may perform one or more pre-processing on the received water surface image. For example, when the original water surface image appears as a red or green color image due to the illuminator 10 emitting red or green laser light, the processor 14 may first grayscale the color water surface image to make the water surface image gray. Order image.

The scattering block corresponds to a high-luminance region in the water surface image. Therefore, after being converted into a gray-scale image, the region corresponding to the scattering block in the gray-scale image is still a part with higher brightness and is biased toward white.

The processor 14 can further normalize the pixel values of each pixel in the grayscale image. In other words, for example, 256 gradations in the grayscale image are normalized to a value between 0 and 1. And according to a predetermined brightness threshold, each pixel of the normalized gray scale image is divided into two pixel values according to the brightness threshold, for example, a first pixel value and a second pixel value to generate a binary image.

The brightness threshold used to draw the grayscale image pixels is used to define a criterion for defining a high brightness region in each grayscale image. If the brightness threshold is too low, it may be easy to cover the noise in the image and overestimate the size of the scattering block, so that the turbidity of the tested water source 2 is misjudged. Therefore, the selection of the brightness threshold needs to be properly calculated. The recommended value is between 0.6 and 0.8, but the actual selection is still based on the actual design and calculation of the system. The above suggested values are only one of the examples.

Please refer to the schematic diagram of the binary images 30a to 30c illustrated in FIG. 3 . For example, assuming that the luminance threshold is set to 0.8, the processor 14 normalizes the grayscale image and the pixel with the pixel value greater than or equal to 0.8 is designated as the first pixel value, which may be specified as 1 in this example; A pixel whose pixel value is less than 0.8 is designated as the second pixel value, which is 0 in this example. Thereby, the binary image can display only two colors of white (pixel value 1) and black (pixel value 0), thereby distinguishing high-brightness regions reflecting light scattering energy and others. section. In the binary image, a plurality of adjacent pixels group designated as the first pixel value are grouped together to present a white block, that is, a block corresponding to a high-brightness performance in the water surface image, which also includes a corresponding In the high-brightness area of the scattering block.

However, the white blocks in each binary image, that is, the blocks belonging to the pixel group of the first pixel value, may be more than one place. Since the detected water source 2 may be a flowing water source, as described above, when the image capturing device 12 captures a water surface image, the image image may highly contain a reflection pattern generated by the water flowing against the waterway wall, the reflection. The graphics will also exhibit high brightness in the water surface image. The reflection patterns 204a and 204b shown in the water surface image 20c of FIG. 2 will appear as white blocks in the corresponding binary image, as shown in FIG. The white blocks 302a and 302b shown by the binary image 30c corresponding to the water surface image 20c of Fig. 2 . However, this type of white block is not the data that the processor 14 needs to evaluate the turbidity of the water, but belongs to the noise in the image. Therefore, in this embodiment, the image including the noise can be excluded first to avoid the processor. 14 misjudgment.

Therefore, after the processor 14 generates the binary image according to the grayscale image, a labeling process may be performed to calculate a block of the white block formed by the plurality of first pixel values among the same binary image. Quantity. If a white block formed by a plurality of first pixel value clusters is found, the block is marked as 1; if a second white block formed by the first pixel value clustering is found, the flag is marked 2, and so on. Finally, the number of blocked blocks in the same binary image is counted. In this embodiment, when the processor 14 determines that the number of blocks in a binary image is greater than 1, the representative binary image includes other noises in addition to the block including the corresponding scattering block. At this time, the processor 14 may decide to directly exclude the binary image and its corresponding water surface image, and not include the subsequent turbidity calculation, such as the water surface image 20c shown in FIG. 2 and the binary image shown in FIG. 3. 30c. The binary image with a block number of 1 can determine that the marked block is the white block corresponding to the scattering block. The processor 14 can select a binary image with a block number of 1 as a candidate image from a plurality of binary images, such as the binary images 30a and 30b shown in FIG. 3 in this example.

The processor 14 may perform morphological erosion and pre-expansion processing on the binary image before performing the labeling process to eliminate the burr phenomenon of the white block in the binary image, so as to accurately calculate the area of the white block. .

The processor 14 further performs an area parameter calculation on the candidate image including the only one white block, for example, respectively calculating the number of pixels of the white blocks 300a, 300b of the respective candidate images 30a, 30b to generate corresponding to the candidate image. The area of the white block is an area parameter and can be temporarily stored in the memory unit 16.

After calculating the area parameters of all candidate images, the processor 14 may average all the area parameters of the candidate images for detection to generate parameter statistics, that is, corresponding to the area average of the scattering blocks, as an evaluation. The basis for comparison of the turbidity of the water source of the tested water source 2.

However, it is worth noting that even if the number of blocks formed by the clustering of the first pixel values in the binary image is 1, it is possible that noise may affect the pixels of the scattering block, as shown in FIG. A schematic diagram of a candidate image 40, although only one white block 400 is included in the candidate image 40, a portion 404 of the white block 400 corresponds to a scattering block for evaluating water quality and is connected to the scattering block. The other portion 402 is actually a noise that should be excluded. At this time, the candidate image 40 shown in FIG. 4 is calculated to have a significantly larger area parameter than other candidate images. Therefore, in this embodiment, the processor 14 may first filter each candidate image by using a specific screening value. The screening value applicable in this embodiment may also be an average value of the area parameters of each candidate image to be screened.

The processor 14 may compare the area parameters of each candidate image with the screening values to screen the candidate images that should be retained. For example, the candidate image whose parameter area is larger than the screening value is screened, thereby discarding the candidate image that may contain the noise, and retaining the area parameter of the candidate image that is not screened to calculate the parameter statistics.

The method of filtering the candidate images and then calculating the parameter statistics by filtering the values, in addition to eliminating the misjudgment that may be caused by the noise, also helps to converge and concentrate the values of the parameter statistics. Therefore, the processor 14 can perform screening of candidate images more than once, and the number of screenings suggested in this embodiment is twice to obtain more accurate parameter statistics. For example, the image capturing device 12 continuously captures 900 water surface images, and filters 750 candidate images after graying, binarizing, erasing and labeling procedures. After calculating the area parameters of 750 candidate images and the average of the area parameters of the 750 candidate images as the screening values, the 750 candidate images are screened according to the screening values. Assuming that the area parameter of the 250 candidate images is larger than the screening value, the processor 14 may select the remaining 500 candidate images whose area parameters are not greater than the screening value, and then calculate the 500 candidates by using the 500 candidate images as samples. The average of the area parameters of the image. Finally, the 500 candidate images are screened according to the average of the area parameters of the 500 candidate images. It is assumed that among the 500 candidate images, 350 candidate images are retained after screening, and the processor 14 side averages the area parameters of the 350 candidate images to calculate the final parameter statistics.

After the processor 14 calculates the parameter statistics, the preset turbidity comparison table can be compared according to the parameter statistics. The turbidity comparison table may be stored in the memory unit 16 and record the range or threshold of the average area of the plurality of different scattering blocks, and the turbidity level corresponding to each average area interval or threshold. . For example, if the statistic value of the recordable parameter in the turbidity comparison table is 300 pixels or less, the turbidity level is "clear", the parameter statistic is 300 pixels to 850 pixels, and the turbidity level is "mild turbidity", and The parameter statistic value is "moderate turbidity" from 850 pixels to 1350 pixels, and "severe turbidity" is higher than 1350 pixels. The turbidity level can distinguish the water quality into a type of clear, mild turbidity, moderate turbidity or heavy turbidity, instead of the specific turbidity measurement value.

The water quality result calculated and analyzed by the processor 14 can be output to a display interface of the control center (not shown in FIG. 1), such as a webpage type water quality monitoring system screen. By directly displaying the turbidity level, the water quality monitoring personnel can quickly and intuitively understand the water quality status of the tested water source 2, so that when the water quality reaches the turbidity level (such as moderate turbidity) that should be alert, it can be quickly performed. Related treatments, such as closing the reservoir inlet, to prevent damage to the generator due to accumulation of excessive suspended particles (such as sediment) in the water source.

[Another water quality testing system embodiment]

Considering that the gradation image in the water surface image captured by the image capturing device 12 may be affected by the intensity of the ambient light source, the water source 2 of the same turbidity exhibits different sizes under different ambient light sources. In the case of a gradation image, different implementation means can be employed in the same architecture as shown in FIG. In this embodiment, the processor 14 uses the two-stage method to obtain the scattering pattern defined by the different threshold values, and compares the area ratio between the two scattering patterns to determine the turbidity of the tested water source 2. Degree, in order to improve the tolerance of the test results to the ambient light source.

The process of extracting the water surface image from the light source of the illuminator 10 to the image capturing device 12 and graying the surface is the same as that of the foregoing embodiment. For details, refer to the description of the above embodiment. It is worth mentioning that, in this embodiment, the brightness threshold for generating a binary image from the normalized grayscale image includes a first threshold and a second threshold. The second threshold is higher than the first threshold, for example, the first threshold is 0.6 and the second threshold is 0.8. The processor 14 firstly divides the pixels of the grayscale image into two pixel values of the first pixel value and the second pixel value according to the first threshold value, thereby generating the first binary image, such as the first binary image as shown in FIG. 5A. 50a to 50d. Wherein, in each of the first binary images, the white block formed by the plurality of adjacent pixels corresponding to the first pixel value is a block having high brightness performance in the water surface image, which also includes corresponding to The high-brightness area of the scattering block. The white blocks included in the first binary image are referred to as the first block in this embodiment, such as 500a, 502a, 500b, 500c, and 500d shown in the first binary image 50a to 50d, respectively.

The first binary image 50a to 50d may also contain more than one first block. Therefore, the processor 14 may also perform labeling processing to eliminate the number of blocks in the block. The first binary image, and the first binary image with the number of blocks of 1 is reserved as the candidate image. For example, the first binary image 50a includes two white blocks 500a and 502a, such that the number of blocks exceeds 1. Therefore, the processor 14 discards the first binary image 50a whose number of blocks is greater than 1, and retains the rest. The first binary image 50b to 50d is used as a candidate image. Referring to FIG. 5B, the processor 14 can further divide the pixel values of the grayscale images respectively generating the candidate images (ie, the first binary images 50b to 50d in this example) according to the higher second threshold. The third pixel value and the fourth pixel value are two types to generate second binary images 52b to 52d, respectively.

The third pixel value is the same as the first pixel value, and the fourth pixel value is the same as the second pixel value. In other words, in this example, the third pixel value is 1 and the fourth pixel value is 0, such that the pixel corresponding to the third pixel value is white, and the pixel corresponding to the fourth pixel value is black. Among the divided second binary images 52b to 52d, the white blocks formed by the plurality of third pixel value clusters are referred to as second blocks, as shown by 520b, 520c, and 520d in FIG. 5B. Since the second threshold value is higher than the first threshold value, among the same grayscale image, pixels that are divided into pixel values according to the first threshold value, and some of the pixels are scored according to the second threshold value, It is judged that the pixel value is 0. Therefore, the position of the second block in the image necessarily overlaps the first block included in the corresponding candidate image, and the area of the second block is not greater than the area of the first block in the corresponding candidate image.

Since the image of the candidate image is selected and the image having two or more high-luminance regions is removed, only the second block of the second binary image generated for the candidate image has only one second block. The program for re-charging can be omitted.

After the processor 14 obtains the first block of each candidate image and the second block of the second binary image corresponding to the candidate image, the processor can separately calculate the area of the first block and the second block of each candidate image. And calculating the ratio of the area of the second block to the area of the first block as the area parameter in the present embodiment. Specifically, as shown in FIG. 5A, the first binary image 50b, 50c, 50d and the second binary image 52b, 52c, 52d corresponding to FIG. 5B respectively, the processor 14 can calculate the second block 520b. Comparing the area ratio of the first block 500b, the area ratio of the second block 520c to the first block 500c, and the area ratio of the second block 520d to the first block 500d are calculated. The area parameters can be temporarily stored in the memory unit 16.

After the processor 14 completes the calculation of the area parameters of the corresponding first block and the second block in all candidate images, the candidate image may be filtered by using the screening value to exclude candidate images that still contain noise. Specifically, it is used to exclude candidate images of white blocks containing noise as shown in FIG. In this embodiment, referring to the first block and the second block shown in FIGS. 5A and 5B, it can be seen that the area ratio of the second block 520b to the first block 500b may be excluded from the screening value. For details, please refer to FIG. 4 and its corresponding content, which will not be repeated here. The screening value of the embodiment may be an average value of the area parameters of the plurality of candidate images used for screening, and the processor 14 may screen the candidate image whose area parameter is greater than the screening value, and the reserved area parameter is not greater than the area parameter. The candidate image of the average value, and then calculating the parameter statistical value of the area parameter of the retained candidate image, that is, the average of the area ratio corresponding to the scattering block, for comparison with the turbidity comparison table in the pre-existing memory unit 16. Yes, to determine the turbidity level of the tested water source 2.

The turbidity table in this example records the range or threshold of the area ratio of a plurality of different scattering blocks, and the turbidity level corresponding to each area ratio interval or threshold, and the processor 14 calculates the parameters. After the statistical value, the preset turbidity comparison table can be compared according to the parameter statistics.

In particular, the means for determining the water quality using the area ratio can exert a better effect in the specific environment average brightness (EAL) range. The ambient average brightness refers to the average brightness of the tested water source 2 excluding the coherent light source illuminated by the illuminator 10. For example, the image capturing device 12 captures the water surface image when the light source of the illuminator 10 is not turned on, and calculates an average of the luminance values of all the pixels of the water surface image that does not include the light source of the illuminator 10. The brightness value of each pixel can range from 0 to 255 (ie, the darkest to the brightest), so the average brightness value of the entire water surface image without the same source is also between 0 and 255. According to the result of the experiment, when the average brightness value (that is, the ambient average brightness) of the water surface image without the homology light source is between about 20 and 80, the above method for judging the water quality by the area ratio can be accurately obtained. result.

Therefore, the illuminator 10 can be controlled by the control center to stop transmitting the same dimming signal for a short time, and the image capturing device 12 captures the water surface image without the coherent light source for the processor 14 when the light emitting device 10 pauses to emit light. Calculate the average ambient brightness at that time and determine whether it is still suitable to continue the practice of assessing water quality by area ratio as described in this example.

In summary, the use of the area ratio to determine the water quality can make the water quality detection system 1 more resistant to interference from ambient light sources other than the dimming emitted by the illuminator 10. Even if the water source 2 of the same turbidity interferes with the ambient light source of different intensities, it may affect the total area of the high-brightness area, but the area ratio of the high-brightness area drawn by different threshold values in the same water surface image can be maintained. Consistent. In other words, the area of the first block of the first binary image derived from the original water surface image is larger than the area where the interference level of the ambient light source is high when the interference level of the ambient light source is low, but the first area derived from the same water surface image The area ratio of the block and the second block will remain the same regardless of the strength of the ambient light source, that is, the area of the first block and the second block when the interference of the ambient light source is weakened or weakened. Correspondingly, the synchronization is reduced or synchronized, so the ratio is still consistent. Thereby, even if the illuminator 10 and the image capturing device 12 cannot be disposed in a place with no or almost no ambient light source, the water quality detecting system 1 can correctly detect the turbidity of the source water source 2.

For the same parts as the previous embodiment, refer to the description of the previous embodiment, and details are not described in this embodiment.

[Water quality testing method example]

Next, please refer to FIG. 6. FIG. 6 is a flow chart of an embodiment of a water quality detecting method provided by the present invention. The water quality detecting method can be implemented by using the water quality detecting system shown in FIG. 1, so please refer to the block diagram shown in FIG. 1 for illustration.

The illuminator 10 can receive the command from the remote control center to start emitting the same dimming light to the water source 2 to be tested (S601), and the image capturing device 12 continuously captures the water surface image of the water source 2 to be tested (S603). The water surface image includes a gradation image in which the light source emitted from the illuminator 10 collides with the suspended particles in the water source 2 to be measured, and the brightness of the gradation image gradually increases from the edge toward the center, thereby forming a high brightness in the central portion of the gradation image. region.

The remote processor 14 receives the captured plurality of water surface images via a wired or wireless network and performs image processing to obtain a scattering block corresponding to the high luminance region in each of the water surface images (S605). The processor 14 further calculates the area parameters of the respective scattering blocks, and compares the respective area parameters with the screening values to screen out the water surface image containing the noise that affects the water quality detection result (S607), and then according to the retained water surface. The area parameter of the image calculates the parameter statistic (S609), for example, the average of the area parameters of the retained water image for evaluating the turbidity of the water. Finally, the processor 14 may compare a preset turbidity comparison table according to the parameter statistics (S611), according to a plurality of different turbidity levels recorded in the turbidity comparison table and corresponding parameter statistical value intervals or threshold values, The turbidity level to which the water source 2 to be tested belongs is determined.

[Another Embodiment of Water Quality Detection Method]

Please refer to the flowchart of another embodiment of the water quality detecting method of the present invention shown in FIG. 7. This embodiment further describes the processing process of the water surface image captured by the image capturing device by the processor, and is obtained for evaluating the water quality. For the steps of the scattering block of the condition, please refer to the block diagram shown in FIG. 1 and its components for understanding.

The illuminator 10 emits a coherent light source such as laser light to the water source 2 under test at an angle substantially perpendicular to the water source 2 to be tested (S701), so that the water surface image of the water surface is captured by the image capturing device 12 (S703), and the water surface image includes A gradation image of light scattered on the surface of the water due to the blocking of suspended particles in the water source 2 under test. The brightness of the gradation image gradually increases from the edge toward the center, and a high-luminance region is formed in the central portion of the gradation image.

After receiving the captured water surface image, the processor 14 can perform grayscale processing on the water surface image (S705), convert the color water surface image into a grayscale image, and further normalize the pixel values in the grayscale image. (S707), the pixel values of the grayscale image originally including 256 levels are distributed between 0 and 1.

The processor 14 can then classify the grayscale image according to the brightness threshold to generate a binary image (S709). The brightness threshold may be a predetermined pixel value, for example, 0.8, for comparing each pixel value in the normalized grayscale image, and specifying a pixel value greater than or equal to the brightness threshold as the first pixel value, and will be smaller than The pixel value of the luminance threshold is specified as the second pixel value. The first pixel value may be 1, representing white and high brightness pixels, and the second pixel value may be 0, representing black and low brightness pixels. Thereby, a binary image corresponding to the water surface image can be generated. It is worth mentioning that the brightness threshold may also be a pixel value suitable for a grayscale image that is not normalized, for example, 204, for dividing a grayscale image with a pixel value ranging from 0 to 255. If the pixel value is greater than or equal to 204, the first pixel value is less than 204, and the second pixel value is less than 204. Therefore, the above step S707 may be omitted depending on the setting of the brightness threshold.

The binary image may include one or more white blocks formed by a plurality of adjacent phase clusters belonging to the first pixel value. To facilitate subsequent processing, the processor 14 may first perform the shape on the white block. The expansion or erosion treatment is performed to eliminate the burrs of the block (S711). The one or more white blocks include a scattering block corresponding to a high-luminance region of the water surface image, but may also include noise generated when the water source 2 is tested to impinge on the water channel wall, in order to The noise is excluded, and the processor 14 can perform labeling processing on the binary image, calculate the number of blocks of the white block in the binary image (S713), and determine whether the number of blocks is 1 (S715). If the number of blocks of the binary image is 1, it means that the only one white block in the binary image corresponds to the scattering block for evaluating the water quality, and the processor 14 can then calculate the white color of the binary image. The area parameter of the block is, for example, the total area (number of pixels) of the white block, and is stored in the memory unit 16 (S717). If the number of blocks is greater than 1, it means that the binary image includes noise other than the scattering block, and the area parameter of the binary image is not calculated and stored.

Regardless of whether the area parameter of the binary image is calculated or not, the processor 14 can determine whether the scattered block data of all captured water images has been calculated (S719). If the data of all the water surface images has not been calculated yet, Then, the next water surface image can be selected (S721) and the process returns to step S705 to continue the same process until all the data of the water surface image has been analyzed and calculated (ie, the determination result in step S719 is YES), and then the processor 14 The selected binary images are further screened.

In order to avoid that the white block belonging to the noise is connected to the white block belonging to the scattering block and is not correctly rejected during the labeling process, the processor 14 can compare each stored area parameter with a specific screening value. The binary image whose area parameter exceeds the screening value is excluded (S723). The filter value may be an average value of the area parameters of all the binary images to be screened. Therefore, the processor 14 may first calculate the average value of all the area parameters stored in the memory unit 16 as the filter value in step S723. Then, the area parameters of each binary image are compared according to the screening value, and the binary image whose area parameter is larger than the screening value and its area parameter are discarded, for example, the area parameter is deleted from the memory unit 16, and the value is not greater than the screening value. The area parameter of the binary image is used to calculate the parameter statistics, and the retained binary image is the candidate image.

In order to converge and concentrate the values of the parameter statistics used for the final evaluation of the water quality, the processor 14 may filter the area parameters of the binary image more than once by the screening value, for example, a screening process that is preset twice. Therefore, after the screening process of step S723, it is judged whether or not the predetermined number of times of screening has been performed (S725). If the number of times of screening has not reached the preset number of times, the processor 14 may return to step S723 to filter the area parameters still retained after the previous screening by the screening value. Although the screening value at this time is still the average value of the area parameters, the number of area parameters used for screening has been reduced by the previous screening, so the value of the screening value used for each screening will be slightly different.

After the area parameter of the binary image is filtered by a preset number of times (ie, the result of the determination in step S725 is YES), the processor 14 calculates the area parameter according to the area parameter of the binary image that is still retained in the memory unit 16. Parameter statistics (S727). The parameter statistical value in this example may also be an average value of the reserved area parameters, representing the size of the average scattering block generated by the suspended particles in the tested water source 2, and according to the calculated parameter statistical value and turbidity. The comparison table is compared (S729). The turbidity comparison table records the turbidity levels of the plurality of water quality, and the interval or critical value of the parameter statistical values corresponding to each turbidity level, thereby determining the turbidity level of the tested water source 2.

[Another embodiment of the water quality detecting method]

Please refer to FIG. 8. FIG. 8 is a flow chart of another embodiment of a water quality detecting method provided by the present invention. The same as the embodiment shown in FIG. 7, the illuminator 10 first emits the same dimming light to the water source 2 to be tested (S801), and then the image capturing device 12 captures the water surface image including the gradation image (S803), and After receiving the water surface image by the processor 14, the water surface image is gray-scaled to generate a grayscale image (S805), and the pixel values of the grayscale image are normalized so that the pixel values are distributed between 0 and 1 (S807). deal with.

The difference from the embodiment shown in FIG. 7 is that, in this embodiment, the brightness threshold of the processor for drawing the pixel values of the grayscale image to generate the binary image includes the first threshold and the second threshold, and the second The threshold is higher than the first threshold. The processor 14 firstly specifies, according to the first threshold value, the pixels of the grayscale image that are higher than the first threshold value as the first pixel value, and the pixels that are not greater than the first threshold value, The second pixel value, thereby generating the first binary image (S809). The white block formed by the clustering of the plurality of adjacent first pixel values is referred to as a first block, and the processor 14 may further perform morphological expansion and erosion processing on the first block to eliminate the first block. The raw edges.

Then, the processor 14 can also perform a labeling processing procedure on the first binary image, calculate the number of blocks of the first block included in the first binary image (S811), and determine the first binary value currently processed. Whether the number of blocks of the image is 1 (S813). If the number of blocks of the first binary image is 1, the first block in the image is a scattering block corresponding to the high-brightness area in the water image, and the first binary image can be selected as a candidate image, and The processor 14 further divides the pixel value of the grayscale image corresponding to the first binary image by the second threshold to generate another binary image, which is called a second binary image (S815).

In step S815, the processor 14 compares the pixel values of the pixels of the grayscale image according to the second threshold value, and specifies the pixel whose pixel value is greater than the second threshold value as the third pixel value, and is not greater than the second threshold value. Specified as the fourth pixel value. The third pixel value is the same as the first pixel value, and the fourth pixel value and the second pixel value are the same as 0. A white block formed by a plurality of adjacent third pixel value clusters corresponding to the scattering block is referred to as a second block. Since the second threshold is higher than the first threshold, the area of the second block will be smaller than the first block.

After the processor 14 generates the first binary image and the second binary image and the corresponding first block and the second block from the same grayscale image, the processor 14 can calculate the area of the second block separately. The ratio of the area of the first block is compared and stored in the memory unit 16 as the area parameter in the present embodiment (S817).

After the processor 14 calculates the area parameter (ie, step S817), or the number of blocks of the first binary image is determined to be more than one and is not selected as the candidate image (ie, the determination result in step S813 is no) After that, the processor 14 determines whether the scattered block data of all the captured water surface images has been calculated (S819), and if there is still uncalculated water surface image data, the next water surface image (S821) can be selected and returned. In step S805, the same processing is continued until all the data of the water surface image has been analyzed and calculated (ie, the determination result in step S819 is YES), and then the processor 14 filters the stored area parameters.

After calculating and storing all the area parameters filtered by the brightness threshold, the processor 14 further continues to filter the area parameters by using the filter values (S823), and the filter values in this example may be stored in the memory unit 16. The average of the area ratios to exclude area parameters greater than the filter value. After screening, the processor 14 further determines whether the number of times of screening by the screening value has reached a preset number of times (S825). If not, the process returns to step S823 to filter the area parameters again, thereby converging and concentrating the values of the area parameters; When the number of times of screening has met the requirements of the preset number of times, the processor 14 calculates the parameter statistics according to the area parameters still reserved (S827), which in this example may also be the average of the area ratio values. Finally, the parameter statistical value is compared with the data in the turbidity comparison table (S829), and the water quality of the tested water source 2 is determined according to the corresponding turbidity level in the turbidity comparison table according to the parameter statistical value.

The detailed operations of the above steps S823 to S829 are similar to the steps S723 to S729 in the flow shown in FIG. 7, so please refer to the related description corresponding to FIG. 7, and details are not described herein again.

[Water quality detection system correction method example]

In addition to the turbidity of the water quality, the image on the water surface image captured by the image is not stable. If the illuminating power of the illuminator 10 itself is unstable, the water surface image will produce different image representation under the same water quality conditions. For example, when the illuminator 10 is used for a long time to attenuate the power, even if the water quality of the water source 2 is not changed, the high-luminance region of the water surface image is reduced by the reduction of the light energy, causing the back end to be misjudged. In order to avoid erroneous judgment due to the deterioration of the illuminating power of the illuminator 10, it is necessary to detect whether the detecting system 1 has an error due to the fading of the illuminating power of the illuminator 10, and to correct it when it is found that the detecting system 1 is erroneously judged. Since the remote water quality detecting means is adopted, the illuminator 10 may be separated from the control center or the processor 14 where the manager is located by more than the visual distance. Therefore, if the image captured by the image capturing device 12 can be used to determine the illuminator It is one of the most convenient means to generate a power failure condition. Therefore, the present embodiment provides a method for discriminating and correcting the water quality detecting system 1 referred to in the foregoing embodiments by using an image.

Please refer to FIG. 9 for a flow chart of a method for correcting a water quality detecting system. The method can be used to correct a detection result error of the illuminator 10 of the water quality detecting system 1 shown in FIG. 1 due to power degradation. In the present embodiment, when the illuminator 10, the image capturing device 2, and the like of the water quality detecting system 1 are just installed at the water source 2 to be tested (S901), the illuminator 10 has not been subjected to power degradation, and is continuously subjected to the power luminescence. The water source 2 is detected to calculate the water turbidity (S903). Since the ambient light source at night is less, the interference to the detection can be reduced, so the above continuous detection can be performed at night, for example, continuously capturing the water surface image for several hours, and performing the processing described in the foregoing embodiment to obtain the parameter statistical value. , such as obtaining the average area of the scattering block.

The processor 14 can determine the water turbidity measured by the water quality detecting system 1 just after the device is set according to the parameter statistical value comparison turbidity comparison table. At this time, the processor 14 can determine whether the measured water turbidity belongs to the "clear" The level (S905), in other words, is the case where the image captured by the image capturing device 12 is analyzed to find out that the minimum scattering energy is reflected on the water surface image.

If the turbidity of the water analyzed by the processor 14 is not a clear level, the same dimming is continued by the illuminator 10 and the water surface image is continuously captured by the image capturing device 12 (return to step S903) until the processor 14 is aligned. The water quality of the water source 2 to be tested is a clear grade for a predetermined period of time. For example, one week after the installation of the detection system 1 is measured, the water quality judged to be a clear grade for 10 consecutive hours is measured. After the processor 14 measures the data that belongs to the clear water quality for a continuous period of time, the average value of the parameter statistical value determined to be the clear water quality for the continuous period of time is calculated and recorded as the corrected turbidity value, and the preset error is added. Values are formed to form a corrected turbidity range of the detection system 1 (S907). The parameter statistic value may be, for example, an average of the total number of pixels corresponding to the scatter block as described in the foregoing embodiments, so the corrected turbidity value may be the total average of the number of pixels of the scatter block that is continuously determined to be clear water quality.

When the processor 14 adds the corrected turbidity value to the preset error value, a range of intervals is provided, where the same group of illuminators 10 and the image capturing device 12 emit the same dimming and the captured water surface image. The statistical value of the parameter obtained after being processed by the processor 14 is within the range of the corrected turbidity, and can be regarded as the same clear water quality measured when the detection system 1 is just installed. . For example, after the installation of the detection system 1 is completed, the corrected turbidity value is 100 pixels (the corrected turbidity value naturally falls within the range of the "clear" level in the turbidity comparison table), and the preset The error value is 30 pixels. Therefore, in the case of the detection system 1 of the group, when the illuminator 10 is just installed and the illuminating power is optimal, if the water quality of the water source 2 under test is "clear", The parameter statistics of the scattering blocks detected and calculated by the detection system 1 of this group should be between 70 pixels and 130 pixels.

After obtaining the initial corrected turbidity range of the water quality detecting system 1 of the group, the water quality detecting system 1 can be officially started to operate, and the water quality testing described in the above embodiments is performed to determine the turbidity change of the water source of the tested water source 2 (S909) ). During the operation of the detection system 1, the processor 14 can monitor whether the data generated by the analyzed water surface image also has the data of the clear level after a continuous period of time (such as several hours) of analysis and calculation. (S911). If the water quality continuously classified as "clear" has not been encountered during the detection, the process returns to step S909 to continue the detection. When the information of the same level of clearness is continuously obtained during the detection of the water quality detecting system 1, the processor 14 may calculate the average of the parameter statistical values among the consecutive times as the reference turbidity value (S913), for example, all of the time The average of the number of pixels of the scattering block used to calculate the parameter statistics.

The processor 14 may compare whether the reference turbidity value is less than the lower limit of the corrected turbidity range (S915). In the above illustration, it is determined whether the number of pixels of the reference turbidity value is less than 70 pixels. If the reference turbidity value is not less than the lower limit (ie, the minimum value) of the corrected turbidity range, the processor 14 can determine that the illuminating power of the illuminator 10 is normal and not attenuated, so it can return to step S909 to continue the water quality detection and determine whether it is re-occurred. The water quality is in a clear grade for a continuous period of time.

However, if the processor 14 compares the reference turbidity value with the corrected turbidity range and confirms that the reference turbidity value is less than the lower limit of the corrected turbidity range, it is determined that the illuminating power of the illuminator 10 has decayed. The above principle is that when the illuminator 10 is started when the detection system 1 is just installed, it is in a state with the strongest luminous power, and when the same dimming light is emitted to the water source 2 whose water quality belongs to a clear level, The reflected scatter energy will at least exhibit a scattering block with a minimum size equivalent to 70 pixels. If the illuminator 10 maintains the same luminous power or even increases the luminous power during the operation of the detecting system 1, the calculated reference turbidity value should be not less than 70 pixels. Therefore, if the reference turbidity value is less than the lower limit of the corrected turbidity value after the comparison, that is, when the reference turbidity value corresponds to the number of pixels less than 70 pixels, it can be determined that the illuminating power of the illuminator 10 is attenuated, resulting in the same Under the condition of water turbidity, the average area of the scattering blocks of the reactive scattering energy is reduced.

After the processor 14 finds the power attenuation of the illuminator 10 through calculation and comparison, the external signal control device (not shown) can be controlled to adjust the output voltage output to the illuminator 10 according to a certain compensation ratio to illuminate The illumination module of the device 10 (such as the laser module of the laser pointer) performs adjustment adjustment of the corresponding power ratio (S917). For example, the output voltage is increased in accordance with the ratio of the corrected turbidity range lower limit value to the reference turbidity value to increase the luminous power of the illuminator 10.

The processor 14 can again retrieve and analyze the newly received plurality of water surface images after the illumination power of the illuminator 10 is corrected, and the measured water source 2 is still in a clear state (S919). After grayscale, binarization, and labeling, the new parameter statistics are calculated again, and it is verified whether the new parameter statistics are not less than the lower limit of the corrected turbidity range (S921). If the parameter statistical value is still less than the lower limit of the corrected turbidity range, the processor 14 may re-adjust the output voltage according to the difference between the new parameter statistical value and the lower limit of the corrected turbidity range to finely adjust the luminous power of the illuminator 10 (S923). And the verification procedure is completed (S925) by repeated instant verification (ie, returning to step S917) and judgment (ie, step S921) until the new parameter statistic is not less than the lower limit of the corrected turbidity range.

It is worth mentioning that the error tolerance range defined by the error value also affects the accuracy of the calibration method. If the error tolerance range defined by the error value is small, the power attenuation of the illuminator 10 is relatively easy to find. Conversely, if the margin of error defined by the error value is large, the illuminator 10 needs to be detected by the processor 14 when there is a significant power degradation. When the reference turbidity value or the newly calculated parameter statistic value compared by the processor 14 has been greater than or equal to the lower limit of the corrected turbidity range, and it is still known whether the illuminating power of the illuminator 10 is attenuated, then Another set of illuminators and image capturing devices can be additionally added as a control group when the water quality detecting system 1 is initially installed, and the water surface images captured by the two sets of illuminators and corresponding image capturing devices under the same conditions are compared. Whether the generated parameter statistics are similar or identical, or have significant numerical differences, thereby determining whether the lighting power of one of the illuminators 10 is maintained in the normal range or has been attenuated and needs to be adjusted.

[Possible efficacy of the embodiment]

According to an embodiment of the invention, the water quality detecting system and method use the illuminator to actively project a light source to the water source under test, so that the water surface image reflects the pattern of increasing brightness from the outside to the inside according to the turbidity of the water source, and uses the image analysis and processing through the image. The technical calculation of the area of the pattern in the water surface image can instantly determine the water quality of the water source under test, so that the water quality monitoring personnel can detect the turbidity of the water source under test without collecting the water sample in person, thereby effectively protecting the water source under test. Electromechanical equipment at the office.

In addition, according to at least one embodiment of the present invention, the water quality detecting system and method use the ratio of the area of the scattering block in the water surface image to determine the turbidity of the water source under test, and is resistant to the light source of the illuminator except the illuminator. External ambient light interference prevents the correctness of water quality test results from being affected by the strength of external light sources.

Furthermore, according to an embodiment of the present invention, when the water quality detecting system and method described above perform image analysis and processing, multiple screening and filtering methods are adopted, including a labeling process and at least one time to filter the concentration and convergence. Analysis of the area parameters of the image can effectively eliminate the noise in the image, so as to improve the correctness of the detection.

Moreover, in accordance with an embodiment provided by the present invention, a correction method for correcting the water quality detection system to maintain the accuracy of the detection system is further illustrated. By using the water quality detection system to complete the initial installation and the state in which the illuminator power attenuation problem has not occurred, the calibration reference is obtained, so that the data detected during the operation of the water quality detection system can be compared and corrected, and the entire remote monitoring can be achieved. The effect of the correctness of the operation and testing of the water quality testing system.

However, the above description is only an embodiment of the present invention, and is not intended to limit the scope of the invention.

1. . . Water quality testing system

10. . . Illuminator

12. . . Image capture device

14. . . processor

16. . . Memory unit

2. . . Water source under test

20a-20c. . . Water image

200a-200c. . . Gradual image

202a-202c. . . High brightness area

204a-204b. . . Reflection pattern

30a-30c. . . Binary image

300a-300c. . . White block

302a-302b. . . White block

40. . . Candidate image

400. . . White block

402. . . Partial block

404. . . Partial block

50a-50d. . . First binary image

500a-500d. . . First block

502a. . . First block

52b-52d. . . Second binary image

520b-520d. . . Second block

S601-S611. . . Process step

S701-S729. . . Process step

S801-S829. . . Process step

S901-S925. . . Process step

Figure 1 is a block diagram of an embodiment of a water quality detecting system provided by the present invention;

2 is a schematic view showing a water surface image exemplified in the embodiment of the present invention;

FIG. 3 is a schematic diagram of a binary image illustrated in an embodiment of the present invention; FIG.

FIG. 4 is a schematic diagram of candidate images exemplified in the embodiment of the present invention; FIG.

FIG. 5A is a schematic diagram of a first binary image illustrated in an embodiment of the present invention; FIG.

FIG. 5B is a schematic diagram of a second binary image exemplified in the embodiment of the present invention; FIG.

6 is a flow chart of an embodiment of a water quality detecting method provided by the present invention;

7 is a flow chart of another embodiment of a water quality detecting method provided by the present invention;

8 is a flow chart of another embodiment of a water quality detecting method provided by the present invention; and

9 is a flow chart of an embodiment of a method for correcting a water quality detecting system provided by the present invention.

1. . . Water quality testing system

10. . . Illuminator

12. . . Image capture device

14. . . processor

16. . . Memory unit

2. . . Water source under test

Claims (24)

A water quality detecting system for detecting the turbidity of a water source to be tested, the system comprising: an illuminator for transmitting a dimming light from the water source to be tested to the measured water source; and an image capturing device continuously Taking a plurality of water surface images of the surface of the tested water source, each of the water surface images including a gradation image generated by the same dimming light colliding with the suspended particles in the tested water source on the surface of the tested water source, the gradation image The brightness is enhanced from the edge to the center; and a processor receives the water surface images and defines a scattering block of each of the water surface images according to at least one brightness threshold, the scattering block corresponding to the high brightness of the water surface image a region, the processor separately calculating an area parameter of the scattering blocks, and screening the scattering blocks according to a screening value ratio, and according to the scattering blocks retained after the screening The area parameters obtain a parameter statistical value, and the turbidity of the measured water source is determined according to a turbidity comparison table to determine the turbidity of the measured water source; wherein the larger the statistical value of the parameter, the measured water The higher the turbidity. The water quality detecting system of claim 1, wherein the brightness threshold includes a first threshold value, and the processor divides each pixel of each of the water surface images into a first one according to the first threshold value. One of a pixel value and a second pixel value to generate a first binary image, wherein the pixel values of the pixels corresponding to the scattering block are the first pixel value. The water quality detecting system of claim 2, wherein the processor performs grayscale processing on the water surface images to form a plurality of grayscale images, and the processor compares each of the first threshold values according to the first threshold value. The pixels of the grayscale images are respectively assigned to the pixel values of the pixels as one of the first pixel value and the second pixel value, thereby generating the first binary image. The water quality detecting system of claim 2, wherein each of the first binary images comprises at least one first block formed by a plurality of pixel groups corresponding to the first pixel value, The processor calculates a number of blocks of the first block of each of the first binary images, and reserves the first binary image of the number of blocks as a plurality of candidate images, wherein the The first block included in the candidate image corresponds to the scattering block. The water quality detecting system of claim 4, wherein the processor calculates a first block area of the first block of each of the candidate images as the area parameter. The water quality detecting system of claim 4, wherein the brightness threshold further comprises a second threshold value, and the processor, according to the second threshold value, corresponds to the grayscale images of the candidate images. The pixel values of the pixels are respectively designated as one of a third pixel value and a fourth pixel value to generate a plurality of second binary images, wherein one of the plurality of the third pixel values is formed The second block corresponds to the scattering block. The water quality detecting system of claim 6, wherein the processor separately calculates a second block area of the second blocks, and calculates the second block area of each of the candidate images. A ratio relative to the area of the first block is the area parameter. The water quality detecting system of claim 7, wherein the second threshold value is greater than the first threshold value, and the second block area in each of the candidate images is smaller than the first block area. The water quality detecting system of claim 5, wherein the processor filters the area parameters according to the screening value, and retains the candidate images whose area parameters are not greater than the screening value, and according to The parameter values of the candidate image images that are retained are calculated. The water quality detecting system of claim 9, wherein the screening value is an average of the area parameters of the candidate images retained by the processor. The water quality detecting system of claim 9, wherein the processor performs at least two screenings according to the screening value. The water quality detecting system of claim 2, wherein the processor erodes and expands the binary images to eliminate the burr effect of the binary images. A water quality detecting method for detecting turbidity of a water source to be tested by using an illuminator, an image capturing device and a processor, the method comprising: transmitting, by the illuminator, a homology from the surface of the tested water source to the water source under test The image capturing device continuously captures a plurality of water surface images of the surface of the tested water source, and each of the water surface images includes the same dimming light colliding with the suspended particles in the tested water source to generate on the surface of the tested water source a gradation image, the brightness of the gradation image is enhanced from the edge to the center; receiving the water surface images, and defining a scattering block of each of the water surface images according to at least one brightness threshold, the scattering block corresponding to a high-luminance region of the water surface image; respectively calculating an area parameter of the scattering blocks, and screening the scattering blocks according to a screening value ratio; according to the scattering blocks retained after the screening The area parameters obtain a parameter statistical value; and compare the statistical value of the parameter with a turbidity comparison table to determine the turbidity of the tested water source; wherein the larger the statistical value of the parameter, the The higher the turbidity of the water. The water quality detecting method of claim 13, wherein the step of defining the scattering block of each of the water surface images according to the brightness threshold comprises: according to a first threshold value of the brightness threshold, Dividing a plurality of pixels of each of the water surface images into one of a first pixel value and a second pixel value to generate a first binary image; wherein the pixels corresponding to the scattering block The pixel value is the first pixel value. The water quality detecting method according to claim 14, wherein before the generating the first binary image according to the water surface image, the method further comprises: grayscale processing the water surface images to form a plurality of grayscale images. And determining, according to the first threshold, the pixels of each of the grayscale images as one of the first pixel value and the second pixel value to generate the first binary image. The water quality detecting method of claim 14, wherein after the generating the first binary image, the method further comprises: calculating, by each of the first binary images, a plurality of pixels corresponding to the first pixel value The number of blocks of the at least one first block formed by the clustering; and the first binary image with the number of the blocks being one of the plurality of candidate images; wherein the candidate image includes the first A block corresponds to the scattering block. The method for detecting water quality according to claim 16 , wherein after selecting the binary images of the number of blocks as the candidate images, the method further comprises: calculating each of the selected candidate images. A first block area of the first block is the area parameter. The water quality detecting method of claim 14, wherein the selecting the first binary image of the number of blocks to be a plurality of candidate images further comprises: following a second threshold of the brightness threshold a value, the pixel values of the grayscale images corresponding to the candidate images are respectively designated as one of a third pixel value and a fourth pixel value, to generate a plurality of second binary images; One of the second pixel value clusters forms a second block corresponding to the scattering block. The water quality detecting method of claim 18, wherein after generating the second binary images, the method further comprises: separately calculating a second block area of the second blocks; and calculating each The ratio of the area of the second block of the candidate image to the area of the first block is the area parameter; wherein the second threshold is greater than the first threshold, the first of each candidate image The area of the second block is smaller than the area of the first block. The water quality detecting method of claim 17 or 19, wherein after calculating the area parameters of the candidate images, the method further comprises: screening the area parameters according to the screening value, and retaining the area parameter The candidate image is greater than the screening value; and the parameter statistics are calculated according to the area parameters of the candidate images that are retained; wherein the screening value is the area parameters of the candidate images that are retained average value. The water quality detecting method of claim 14, wherein the generating the first binary image further comprises: etching and expanding the binary image to eliminate the burr effect of the binary image . A method for correcting a water quality detecting system according to claim 1, comprising: continuously detecting the measured water source to determine the turbidity of the water source under test when the illuminating power of the illuminator is not attenuated; recording the illuminating The illuminating power of the device is not attenuated and the turbidity of the measured water source for a period of time is one of the corrected turbidity values, and a corrected turbidity range is generated according to the corrected turbidity value and an error value; the correction is generated After the turbidity range, the water surface images of the tested water source are continuously captured, and the statistical values of the measured parameters are analyzed and calculated to determine the turbidity of the tested water source; and the turbidity of the tested water source is determined to last for the length of time When the level is clear, calculating the parameter statistic values within the length of time to generate a reference turbidity value; comparing the reference turbidity value with the corrected turbidity range to determine whether the reference turbidity value is less than the corrected turbidity value The lower limit of the range; and when the reference turbidity value is less than the lower limit of the corrected turbidity range, the illuminating power of the illuminator is increased according to a compensation ratio. The method of claim 22, wherein, after the illuminating power of the illuminator is increased according to the compensation ratio, the method further includes: capturing the water surface images of the water source to be measured to calculate another parameter statistic value. And comparing whether the statistical value of the other parameter is not less than a lower limit of the corrected turbidity range; and when the statistical value of the other parameter is not less than a lower limit of the corrected turbidity range, completing the correction of the water quality detecting system . The method of claim 22, wherein the corrected turbidity value is an average of the statistical values of the parameter calculated over the length of time.