JPH0627014A - Method and apparatus for monitoring contamination of water - Google Patents

Abstract

PURPOSE:To accurately grasp contamination of water in a region where concentration of microorganism is low and to predict variation thereof by concentrating floating substances in automatically sampled water, processing the image of turbid water, and then calculating features thereof. CONSTITUTION:A sampler 2 samples water stored in a closed region 1 which is then introduced to a condenser 3 where suspensoid is condensed. Condensed liquid is introduced from the condenser 3 to an imaging unit 4 incorporating a television camera where the image of the condensed liquid is converted into electrical signals (image signals). The image signals are inputted to an image processor 5 where images of suspensoid and phytoplankton are processed and extracted. A computor 8 delivers an image take-in command based on image information received from the image processor 5 thus controlling operation of the imaging unit 4. The computor 8 receives image information from the image processor 5, weather information from a weather information measuring unit 7, and water quality information from a water quality measuring unit 6 continuously thus monitoring contamination of water region.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a water pollution monitoring device and method for monitoring water pollution such as lake water, sea water, river water and dam water.

[0002]

2. Description of the Related Art When nutrient salts such as nitrogen and phosphorus flow into a closed water area such as a lake or an inner bay, they are accumulated in the water area and algae and other aquatic organisms multiply, leading to eutrophication. Eutrophication will proceed faster if urban sewage or food factory wastewater flows in. The conditions of closed water areas such as lakes and bays are water temperature, oxygen concentration, pH, transparency, CO
It can be represented by a water quality analysis value such as D or the ecology of organisms such as plankton, and in particular, microorganisms such as plankton inhabiting the water body most accurately reflect the change in water quality.
For example, in lakes, the amount of plankton is one indicator of eutrophication, and the volume of phytoplankton is 3-5 cm /
Whether or not it reaches m ³ is an approximate boundary between a hypotrophic lake and a eutrophic lake. Therefore, it is necessary to quantitatively measure the types of microorganisms and their appearance amounts in the water areas subject to pollution monitoring, and to reflect them in the water area monitoring. As a method of monitoring closed water areas, water quality, meteorological instruments, and plankton microscopic observations have been carried out. The measurement of the plankton's appearance species and its amount depends on the microscopic observation, and the measures for monitoring the abnormal breeding of plankton such as the red tide are inevitable. In the case of microscopic observation, generally, water is sampled from a ship on a regular basis and brought back at a specific location for visual observation with a microscope, which requires manpower and labor. In addition, it is difficult to constantly monitor because it requires an expert to identify the type. On the other hand, for continuous monitoring of microorganisms, a device combining an underwater camera and an image processing device (Japanese Patent Application No. 02-18263).
A method for quantifying microorganisms in 0) has been devised.

[0003]

[Problems to be Solved by the Invention] Usually, the number of plankton in lakes and rivers is as low as tens of thousands / ml at most. When the plankton is photographed at a magnification (objective lens 4 times or more) that can be classified according to the shape, the probability that plankton exists in the photographing screen is very small. An apparatus combining an underwater camera and an image processing apparatus (Japanese Patent Application No. 02-182630) is a method of photographing microorganisms in a state as natural as possible and can be used in a liquid having a high concentration of microorganisms. That is, it is effective in a state where the microorganism concentration is high such as a culture tank or a water area where eutrophication has already progressed, but the current pollution such as poor or mesotrophic lake has not progressed so much, and the water area where the microorganism concentration is low is constantly monitored. Not suitable for. Moreover, although the conventional device can monitor the current pollution of lakes and rivers, no consideration is given to predicting how the current pollution will fluctuate in the future and estimating the purification measures against this prediction fluctuation. The purpose of the present invention is to measure online the microbial state in a water area where the microbial concentration is low,
It is an object of the present invention to provide a water pollution monitoring apparatus and method suitable for continuously monitoring the pollution status, accurately grasping the water pollution status, predicting fluctuations in water pollution, and estimating purification measures.

[0004]

[Means for Solving the Problems] The above objects are to automatically sample the water quality of lake water, sea water, river water, dam water, etc., to condense suspended substances in the sample, It is provided with means for capturing turbidity, means for image-processing the captured image to extract turbidity, means for calculating the characteristics of the extracted turbidity, and means for classifying suspended matter into a plurality of shapes based on the characteristics. Then, it is achieved by measuring the appearance amount for each shape classification of suspended solids. In addition, lake water, seawater, river water,
The target water area such as dam water is divided into multiple monitoring blocks, and the convection of the entire water area is calculated based on the flow velocity, water temperature, and inflow / outflow information to / from each monitoring block. Modeling the state of diffusion between monitoring blocks, then calculating the water quality information and plankton generation amount after a certain time with the current as the initial value, and calculating the nutrient index, organic matter index, plankton index based on these calculated values However, it is achieved by simulating the condition of the whole water area after a certain time and predicting the water quality fluctuation.

[0005]

[Function] The number of plankton in lakes and rivers is usually as low as tens of thousands (cell number / ml). Therefore, the sampling means, the concentrating means, and the transferring means concentrate the diluted suspended suspended matter, and enhance the efficiency of image recognition. Further, the image pickup device including the magnifying means, the illuminating means, and the image pickup means completely holds the suspended matter in the liquid that is transferred and flows, and obtains a still screen. In this imaging device, the plankton extraction means measures the type and number of plankton, the area and particle size of suspended solids in consideration of the coordinates between the images and the feature amount of the shape based on the shape recognition image, and further calculates the shape. The distribution and mass of the plankton are measured by classifying them. The image measurement information and the water quality information are continuously measured, and the water pollution state, its range, and the cause are determined and displayed from the time series change of the information, the weather information, and the geographical information of the water area. Furthermore, the state of pollution is estimated by means of simulation, and the purification measures for the contaminated water area are estimated. In this way, it is possible to identify the water-polluted area from the continuous measurement information and to predict it in advance, and to take measures to purify the polluted water area.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is an embodiment of the present invention and shows a block diagram of the overall structure of a water quality monitoring device. The water quality monitoring device of this embodiment includes a sampling device 2, a concentrating device 3, an imaging device 4, an image processing device 5, a water quality measuring device 6, a weather information measuring device 7,
It is composed of a computer 8 and a display device 9.

Next, the details of these devices will be described. The sampling device 2 samples the stored water in the closed water area 1 such as a lake. Suspended suspended solids in the water area 1, especially phytoplankton, change from moment to moment depending on the position, time, and season, so it is necessary to carry out sampling to cope with this change. In FIG. 2, in order to sample the stored water, the position adjusting device 202 operates the conduit tube 203 in the vertical direction to change the water depth position of the inlet port 204, and the sampling pump 201 sucks the sample liquid from the inlet port 204. Water is taken and sent to the concentrator 3. Here, the water depth position operation is adjusted automatically or from a remote place in consideration of the time zone, the season, and the like. By setting the inlet port 204 near the water surface, it becomes possible to sample floating phytoplankton, which is a cause of water-bloom and red tide. The sample liquid from the sampling device 2 is guided to the concentrating device 3 to concentrate suspended suspended substances. Water area 1 of lakes and dams is several thousand ppm when it rains
However, it is usually clear at several tens of ppm or less. In addition, phytoplankton is tens of thousands (cells / ml) even in large numbers, and averages hundreds (cells / ml).
It is said that. Even if this is observed by a microscope with an objective lens of 4 times, it is determined whether or not there is one cell in one screen (observation volume of about 0.2 mm ³ ), which cannot be said to be efficient monitoring. Therefore, it is necessary to concentrate the sample liquid for efficient monitoring. FIG. 3 shows an example of the concentrating device 3. The sample liquid is sprinkled by the conduit 302, and the reticulated separation membrane 3
In 03, the suspended substance is concentrated and separated and stored in the concentration tank 304. The concentration ratio is determined by the ratio of the sample liquid amount and the concentrated liquid amount. The filtrate from which the suspended solids have been separated is drained. Further, the concentrated liquid is collected by the pump 305 from the concentration tank 304 via the extraction pipe 306. As a method for concentrating and separating the suspended substance, a centrifugal filtration method in which centrifugal force is used for separation, a centrifugal separation method or a centrifugal sedimentation method in which a difference in density or a difference in weight is used, may be adopted. The concentrated liquid from the concentrating device 3 is guided to the imaging device 4. FIG. 4 shows an example of the imaging device 4. The imaging device 4 is composed of a first housing 401 and a second housing 402 with respect to a concentrated liquid introducing tube 409. Liquid contact windows 405 and 406 made of transparent glass are arranged in a part of the introduction tube 409 facing each case. Wetted windows 405, 40
One of them has a flat shape and the other has a concave shape, and the space formed when the two are in contact with each other serves as a sample chamber and holds a part of the concentrated liquid. The first housing 401 contains a magnifying optical lens 404 and an ITV (TV camera) 403 focused on the sample chamber, while the second housing 402 has an illuminating device 402 having a condenser lens, and a liquid contact surface. A driving device 408 for adjusting the position of the window 406 is installed. In such a configuration, the driving device 408 is controlled by a command from the image processing device 5 described later or a timer from the outside, and vertically moves the liquid contact window 406. FIG. 4 shows a state in which the liquid contact window 406 is lowered when the sample chamber is opened, and the liquid in the sample chamber is replaced by the flow of the concentrated liquid. And the liquid contact window 4
By raising 06 again, a new concentrate is retained in the sample chamber. That is, the sample chamber has a liquid contact window 4
It is formed during the 06 ascending operation. The concentrated liquid in the sample chamber is illuminated by the transmitted light method, and is received by the ITV 403 via the magnifying optical lens 404 and converted into an electric signal. At this time, the retentate in the sample chamber is in a stationary state without being affected by the concentrated liquid flowing through the introduction tube 409. These operations are continuously performed and the screen is automatically updated. In addition, with the liquid contact window 406 being lowered, a wiper (not shown) is installed in the introduction tube 409 to perform cleaning of both liquid contact window surfaces and forced replacement of the sample chamber retentate. You can also The image converted into an electric signal (video signal) by the imaging device 4 is input to the image processing device 5 and image-processed and extracted the shapes of suspended suspended matter and phytoplankton.
In addition, an image capture command is issued based on the information from the image processing device 5, and the operation control of the imaging device 4 is performed.

FIG. 5 shows an embodiment of the image processing apparatus 5. The image processing device 5 includes a CPU (central processing unit) 50.
1, a main memory 502, a storage device 503, a communication interface 504, a system bus 505, and an image processing unit 506. The CPU 501 executes a program stored in the main memory 502 and controls the image processing apparatus 5 as a whole. The storage device 503 stores a program for controlling the image processing device 5 and an image processing result. The program is executed by the CPU 50 at the initial start of the image processing apparatus.
One microprogram is transferred to the main memory 502. The communication interface 504 has a data transmission / reception function with other computers and measurement devices, and transmits image measurement information from the CPU 501 to the computer 8 and control signals to the drive device 408 of the imaging device 4 via the communication interface 504. carry out. The image processing unit 506 includes an image memory 507, an image processing processor 508, and a video interface 509. The image memory 508 has a vertical direction of 25.
A grayscale image memory having 6 pixels, 256 pixels in the horizontal direction and a luminance gradation of 256 (8 bits), 256 pixels in the vertical direction, and 2 in the horizontal direction.
A plurality of binary memories each having 56 pixels and a brightness gradation of 2 (1 bit) are arranged. Image processor 508
Is composed of a plurality of LSIs that perform high-speed grayscale image processing calculation, binarization processing, shape feature amount extraction, and the like on the grayscale image and the binary image in the image memory 508.
The video interface 509 has a plurality of inputs and outputs, has an A / D converter at the input and a D / A converter at the output, and converts an analog video signal into a digital signal. Inside the image processing device 5, processing is performed using digital signals. The video signal from the image pickup device 4 is converted into a digital signal in the video interface 509 and is stored in the image memory 507 in real time as one grayscale image, and the image processor 508 outputs the grayscale image stored in the image memory 507 to the grayscale image. And executes various image processing.
The timing of image processing is controlled by the CPU 501. The monitor television 510 connected to the video interface 509 displays the image stored in the image memory 507 and the video signal of the imaging device 4. The image processor 508 drives the driving device 408 to move the liquid contact window 409 downward and upward. By this descending and ascending operation, the concentrated liquid in the sample chamber is exchanged, and a new image is captured by the imaging device 4.

Details of the processing procedure of the image processing apparatus 5 will be described with reference to FIGS. 6, 7, and 8. According to the observation by the present inventors, in the enlarged image of the transillumination type imaging device 4, the phytoplankton has a high brightness due to partial light transmission, but is darker than the liquid phase part. In addition, dust-like substances have lower brightness than plankton. The grayscale image (a) in FIG. 6 is an example of a phytoplankton image of the concentrated liquid. As shown in (a) of the same figure, the grayscale image includes a region of a bright liquid phase portion Z, phytoplankton A, B and C darker than the liquid phase portion, and a dust-like substance D. As described above, the grayscale image has grayscale information according to the brightness (brightness), and the grayscale image is binarized to separate and extract only phytoplankton.
That is, a binarization process is performed in which a region having a lower luminance than the luminance level h ₁ is set to 1 and a region having a higher luminance than h ₁ is set to 0 on the basis of the specific luminance level h ₁ . The binary image converted into the binary information of 0 and 1 is shown in FIG. First, a binarized brightness level h ₁ is set between the brightness of the liquid phase portion and the brightness of the suspended substance (plankton and dust-like substance), and the suspended substance is set to 1
Then, a binary image of the suspended substance with the liquid phase portion set to 0 is obtained. Next, a binarized luminance level h ₂ is set between the dust-like substance portion luminance and the plankton luminance, and the dust-like substance is set to 1 and the others are set to 0.
A binary image of the dust-like substance is created and is subtracted from the binary image of the suspended substance to obtain the binary image of FIG. 6B. FIG. 7 shows the procedure of this binarization processing. First, the internal system time is read out, and image processing is executed if the preset start time or start cycle matches (54).
0). The image processor 508 stores the plankton image obtained through the image interface 509 in the gray image memory in the image memory 507, then performs the gray image processing operation, binarization process, and stores it in the binary memory (54
1). Details of the binarization processing will be described with reference to FIG. Furthermore, the labeling process is executed for this binary memory, and the total number of objects and the total area of the entire image are calculated (5
42), followed by calculating the particle size distribution, average particle size, etc. (543). Next, for each plankton, the shape feature amount such as the area, the perimeter, the shape coefficient, the second moment, the number of end points, the number of intersections, and the number of holes is calculated and stored in the main memory 502 (544). The CPU 501 transmits these various image processing measurement values to the computer 8 via the communication interface 504 (545). Further, FIG. 8 shows a detailed procedure of the binarization process. The plankton video signal of the image pickup device 4 is stored in the grayscale image memory G1 (550), and the luminance information is extracted as a luminance-pixel histogram (551). The brightness-pixel histogram can be expressed as a curve of the number of pixels, with the horizontal axis representing the luminance (0 to 255) and the vertical axis representing the number of pixels. From this curve, for example, a specific brightness level h ₁ for separating and extracting only phytoplankton is calculated using a method of subtracting a fixed constant from the maximum value of the curve (552), and the brightness level h ₁ is used as a reference. , A region having a luminance lower than the luminance level h ₁ is set to 1 and a region having a luminance higher than h ₁ is set to 0, and stored in the binary memory B1 (553). In order to calculate the luminance information of only the binary extracted plankton, AND image is executed with the grayscale image memory G1 and the binary memory B1 (mask processing), and the image in which the luminance of the background liquid phase part is set to 0 is a grayscale image memory. It is stored in G2 (554). With respect to the grayscale image memory G2, the brightness-pixel histogram is extracted by the same method as the above steps 551 and 552 (555), and the binarized brightness level h ₂ for separating the dust-like substance portion and the plankton is calculated (556). ) Is performed, and ₂ according to the brightness level h 2.
The binarization is performed and stored in the binary memory B2 (557).
In the binarization processing procedure, the grayscale image (a) was directly processed, but the background image of each image (the image of only the liquid phase portion) is captured in advance, and the image is subjected to the difference processing with the image. You may perform a process. By executing this pre-processing, it is possible to eliminate the uneven brightness of the entire image, that is, the influence of the brightness due to the illumination light, and obtain a good extracted image. Further, the setting of the brightness level for the binarization process can use a fixed value and an automatic binarization method in which the brightness distribution (histogram) of each image is changed.

The computer 6 continuously receives the image information from the image processing device 5, the meteorological information from the meteorological information measuring device 7, and the water quality information from the water quality measuring device 6, and monitors the pollution state of the water area. . Here, water quality information includes water temperature, turbidity, p
H, DO (dissolved oxygen concentration), electrical conductivity, chemical oxygen demand COD, flow velocity, water level, etc., and weather information includes temperature, wind direction, wind power, solar radiation, rainfall, etc. The image processing device 5, the meteorological information measuring device 7, and the water quality measuring device 6 are installed at optimum positions for monitoring in consideration of the geographical condition of the monitored water area. It is desirable to divide the water area into multiple monitoring blocks and place each measuring device at the representative point of each block. Here, FIG. 9 shows a configuration example of the computer 6. A CPU (Central Processing Unit) 805 transfers various programs stored in the program storage device 803 to the main memory 806 and executes them. The execution timing is the video interface 80.
It depends on a start command signal from the display device 9 connected to the display device 8, a timer, information from the input / output interface 807, and the like. The measurement value database 801 stores image information, water quality information, meteorological information and simulation results received via the input / output interface 807, and the terrain database 802 stores the shape of the water area, the inflow water amount, the inflow location,
Data such as the amount of runoff water and the location of runoff is stored. The program storage device 803 stores an abnormality diagnosis program, a water quality simulation program, a data transmission / reception program, a display device operation program, and the like. The knowledge base 809 stores information necessary for knowledge engineering, fuzzy, and neural network such as rules based on past phenomena. The abnormality diagnosis program diagnoses the pollution status of the water area, its factors, and purification measures based on various measurement information and the stored rules of the knowledge base 809. First, the phytoplankton increase / decrease tendency is grasped by using the total volume and number of plankton, and if the phytoplankton is an influencing factor, its type is identified, and the growth factor is diagnosed from the rule of the knowledge base 809. The type identification is obtained from the shape feature. For example, microcystis and urogelena, which are responsible for blue-green algae and red tide, are identified using measurement information such as the circular shape factor and particle size, and the area ratio of cells to hollow (hole) portions. The filamentous or rectangular plankton is identified by the major axis / minor axis length ratio, the area / peripheral length ratio, the number of end points and the number of intersections, and the like. In addition, star-shaped and ring-shaped plankton have end points, the number of intersections, cells and hollows (holes).
It is judged from the area ratio of the parts, the number of holes, the area ratio of the cells and the hollow (hole) parts, the area ratio of the perimeter, and the like. For these judgments, a knowledge engineering method using crispy or fuzzy rules, and a neural network using various measurement information as an input layer and feature values corresponding to the measurement information as teacher data. Techniques can also be used. Various rules, feature values, and plankton names corresponding to them are stored in the knowledge base 809, which is called as needed to execute inference. In this way, the pollution status of water bodies is monitored by continuously measuring the amount of plankton and its species.

FIG. 10 shows an example of processing of this abnormality diagnosis program. The CPU 805 loads the abnormality diagnosis program into the main memory 806 and executes it at the same time when the computer 8 is started. First, the abnormality diagnosis program reads the system time, compares it with the time or cycle registered in advance, and when it determines that it is the start time, moves to step 851 (85
0). Next, the nutrient index I
k, the organic matter index Ic, and the plankton Ip index are calculated (851). The calculation procedure will be described later with reference to FIGS. 11 and 12. Next, the water quality information, the image information, and the three indexes for the next month are simulated with the water quality information, the image information, and the three indexes as initial values (852). The simulation procedure will be described later with reference to FIG. If the current value or simulation value exceeds the reference range, it is determined that an abnormality has occurred when the water quality information and the type and amount of emerging plankton do not match the annual fluctuation (85
3), the place where the abnormality occurs, the inflow water that causes the abnormality are estimated from the simulation result or the past empirical rule (85
4). Measures against abnormal factors are stored in the knowledge base and inferred from the occurrence place, occurrence factor, season, weather, etc. (855). FIG. 11 shows an example of a method of calculating the nutrient index Ik and the organic matter index Ic. From water quality information DO, turbidity Tu, phosphorus P, nitrogen TN, chemical oxygen demand COD, p
H is read (861), and the nutrient index Ik and the organic matter index Ic are calculated by the equations (1) and (2), respectively (8)
62). Ik = w ₁ × P + w ₂ × TN (1) Ic = w ₃ × DO + w ₄ × COD + w ₅ × pH (2) where w ₁ , w ₂ , ₃ w ₄ , and w ₅ are weighting factors. Rainfall, the factors such as typhoons, the fluctuation of the water quality is expected rainfall, using the coefficient w _6, w ₇ with inflow to water bodies, the turbidity Tu variable (3), (4) (863). _{Ik = w 6 × Ik (3} ) Ic = w 7 × Ic (4) above calculated index as stores the main memory 806 to the measurement value database 801 (864). In FIG.
An example of a procedure for calculating the Plankton index Ip will be shown. First,
After classifying the plankton into several (N) based on the shape information based on the image information (870), the plankton volume PV (i) and the number PN (i) of each plankton are calculated (871).
To do. The filamentous or rectangular plankton can be identified from the major axis / minor axis length ratio, the area / peripheral length ratio, the number of end points and the number of intersections, and the like.
Furthermore, a star-shaped or continuous plankton can be determined from the number of end points and intersections, the area ratio of cells to hollow (hole) parts, the number of holes, the area ratio of cells to hollow (hole) parts, and the area to perimeter ratio. .
Next, the presence / absence of a plankton of a specific species that causes the blue-green algae and red tide is determined from the shape feature (872). For example, microcystis and urogelena, which are responsible for blue-green algae and red tide, are identified using information such as the circular shape factor and particle size, and the area ratio of cells to hollow (hole) portions. When the existence is confirmed, the plankton occurrence of the abnormality index is stored in the main memory 806 and the measurement value database (874). Plankton generation amount VV and plankton index Ip are (5), (6)
It is calculated by the formula and stored in the main memory 806 and the measurement value database 801 (874, 876). However, w ₈ (i): Plankton classification i when calculating volume
Weighting coefficient C: coefficient

FIG. 13 shows the details of the simulation procedure. First, the inflow / outflow information to the water area, the water quality information for each monitoring block, the weather information, the topographic information of the water area is read from the topographic database 802 (880), and the flow velocity, the water temperature, the inflow / outflow information to the water area for each monitoring block are read. The convection of the whole water area is calculated based on (881). From this convection calculation value and water temperature, the state in which water quality information diffuses between monitoring blocks is modeled. Next, DO, turbidity Tu, phosphorus P, nitrogen TN, chemical oxygen demand COD, and pH after a certain time are calculated using the current water quality information as an initial value, and subsequently, image information and nutrient salts, water temperature, Using DO and COD as initial values, calculate plankton generation using a plankton breeding model (8
82). Based on these calculated values, the nutrient index Ik, the organic matter index Ic, and the plankton index Ip are calculated (88).
3). By repeating steps 882 and 883 for the number of monitoring blocks (884), the state of the entire water area after a fixed time can be calculated. Further, the time is updated, the calculation for about one month is repeated (885), and the result is stored in the main memory 806 and the measurement value database 801 (886). In this way, the water quality simulation program consists of image information, water quality information, meteorological information of the day and the inflow amount of rivers etc.
A simulation calculation is performed based on water quality and geographical information such as topography and size peculiar to the hydrosphere, and the location of the cause of pollution is estimated and the water quality fluctuations such as convection, diffusion, and reaction in the water area on the next day are analyzed. That is, by predicting the water quality fluctuation of the next day, the countermeasure can be taken early. When abnormality is detected by diagnosis or simulation, purification means such as aeration operation by air or oxygen blowing, agitation operation, and ultraviolet irradiation is activated. There are two methods of purification: one is fixedly installed at multiple points in the water area and the other is moved to a contaminated point. The display device 9 displays information in the computer 8 and displays measurement data, a pollution state, a simulation result, etc. in the horizontal and vertical directions of the water area. It also sends signals from the mouse, light pen, keyboard, etc. to the computer 8.

[0013]

According to the present invention, the efficiency of image recognition can be improved by concentrating suspended suspended matter even if the number of plankton in a lake or river is low. Also,
Since the pollution status of the closed water area can be measured continuously and quantitatively from the plankton appearance amount and water quality information, it becomes possible to efficiently monitor and purify the water quality of the water area. In addition, the simulation makes it possible to predict the pollution situation in advance and take measures to purify the contaminated water area.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a sample device.

FIG. 3 is a diagram showing a configuration of a concentrating device.

FIG. 4 is a diagram showing a configuration of an imaging device.

FIG. 5 is a diagram showing an image processing apparatus.

FIG. 6 is a diagram showing an example of a plankton image.

FIG. 7 is a diagram showing a procedure of binarization processing.

FIG. 8 is a diagram showing a detailed procedure of binarization processing.

FIG. 9 is a diagram showing the configuration of a computer.

FIG. 10 is a diagram showing processing of an abnormality diagnosis program.

FIG. 11 is a diagram showing an example of a calculation method.

FIG. 12 is a diagram showing an example of a calculation method.

FIG. 13 is a diagram showing details of a simulation procedure.

[Explanation of symbols]

 1 Water Area 2 Sampling Device 3 Concentrating Device 4 Imaging Device 5 Image Processing Device 6 Water Quality Measuring Device 7 Meteorological Information Measuring Device 8 Calculator 9 Display Device

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Internal reference number FI technical display location // G06F 15/20 D 7052-5L (72) Inventor Shoji Watanabe 4026 Kuji Town, Hitachi City, Ibaraki Prefecture Hitachi Research Laboratory, Hiritsu Manufacturing Co., Ltd. (72) Nobuo Yahagi 4026 Kuji Town, Hitachi City, Ibaraki Prefecture

Claims (9)

[Claims] 1. A means for automatically sampling the water quality of lake water, sea water, river water, dam water, etc., means for concentrating suspended matter in the sample, and means for imaging turbidity in the concentrate.
Each of the shape classifications of suspended matter is provided with a means for image-processing the captured image and extracting turbidity, a means for calculating the characteristics of the extracted turbidity, and a means for classifying the suspended matter into a plurality of shapes based on the characteristics. A water pollution monitoring device characterized by measuring the amount of appearance. 2. The water pollution monitoring device according to claim 1, wherein the pollution state is determined from the distribution of appearance quantity for each classification and the water quality, season, and meteorological data. 3. The water pollution monitoring device according to claim 1 or 2, wherein the appearance amount distribution measurement for each classification is performed in a horizontal direction and a depth direction of the water area. 4. The method according to claim 1, claim 2 or claim 3, comprising means for storing geographical information such as topography and area of the target water area, and means for simulating a pollution situation.
A water pollution monitoring device characterized by estimating a polluted point. 5. The water pollution monitoring device according to claim 4, further comprising means for determining a contaminated region based on a simulation result and purifying the contaminated region. 6. A water pollution monitoring device according to any one of claims 1 to 5, further comprising means for displaying the respective results. 7. A target water area such as lake water, sea water, river water, dam water, etc. is divided into a plurality of monitoring blocks, and a nutrient index, an organic matter index, a plankton index are calculated for each of these monitoring blocks, A simulation is performed based on image information obtained by image-processing water quality information and suspended solids of suspended solids.If the simulation value exceeds the reference range, it is determined that the type and amount of the appearing plankton have abnormally occurred, and abnormal diagnosis is performed. Characteristic water pollution monitoring method. 8. The water pollution monitoring method according to claim 7, wherein the place where the abnormality occurs and the inflow water that causes the abnormality are estimated from the result of the abnormality diagnosis, and the countermeasure against the abnormality factor is inferred. 9. The target water area such as lake water, sea water, river water, and dam water is divided into a plurality of monitoring blocks, and convection of the entire water area is based on the flow velocity, water temperature, and inflow / outflow information to / from each monitoring block. From the calculated convection values to model the state in which the water quality information diffuses between the monitoring blocks, and then calculate the water quality information and the plankton generation amount after a certain period of time with the present as the initial value, and based on these calculated values. A method for monitoring water pollution, characterized in that the nutrient index, the organic matter index, and the plankton index are calculated, the state of the entire water area after a certain period of time is simulated, and fluctuations in water quality are predicted.