CN116933042B - Water ecology evaluation method and system based on deep learning algorithm - Google Patents
Water ecology evaluation method and system based on deep learning algorithm Download PDFInfo
- Publication number
- CN116933042B CN116933042B CN202311190762.8A CN202311190762A CN116933042B CN 116933042 B CN116933042 B CN 116933042B CN 202311190762 A CN202311190762 A CN 202311190762A CN 116933042 B CN116933042 B CN 116933042B
- Authority
- CN
- China
- Prior art keywords
- individual
- data sets
- worst
- point
- distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 46
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 30
- 238000013135 deep learning Methods 0.000 title claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 58
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000010606 normalization Methods 0.000 claims description 14
- 239000000356 contaminant Substances 0.000 claims description 12
- 239000002689 soil Substances 0.000 claims description 5
- 230000004044 response Effects 0.000 claims description 3
- 241000894007 species Species 0.000 description 70
- 238000004364 calculation method Methods 0.000 description 5
- 241000195493 Cryptophyta Species 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 241000902900 cellular organisms Species 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a water ecology evaluation method and a system based on a deep learning algorithm, wherein the water ecology evaluation method based on the deep learning algorithm comprises the following steps: acquiring aquatic organism monitoring data sets, and dividing the aquatic organism data sets to acquire three different data sets; adding optimal individuals and worst individuals into the three divided different data sets to obtain three new data sets; performing data standardization processing on three new data sets; establishing a variable self-encoder for three new data sets after data standardization is completed; obtaining the distance between each individual point and the optimal individual, and the distance between each individual point and the point corresponding to the worst individual; obtaining distance weights of different species; obtaining a projection distance; and obtaining comprehensive evaluation scores according to the distance weights and the projection distances of different species. The method and the device consider the monitoring data of all biological groups, consider the influence of different species, and enable the obtained evaluation result of the aquatic organisms to be more accurate.
Description
Technical Field
The application relates to the field of data processing, in particular to a water ecology evaluation method and system based on a deep learning algorithm.
Background
In the prior art, elements for evaluating the quality of the water ecological environment of a river mainly comprise physical and chemical parameters of a water body, physical habitat and biological groups. The water physical and chemical parameters and physical habitat can be scored by unified relevant standards or the same scoring system established by an environmental supervision organization, so that no deviation problem exists in calculating the final score. There are different index systems for biota groups, including BMWP scoring system, chandler biological index, shannon-Winner diversity index, hilsenhoff index, palmer algae pollution index, and biological integrity index. Different index systems are different for biological species, and the emphasis of index calculation is different, so that different index systems are selected for calculation for different watercourses. When the aquatic organism index is calculated, an index system adapting to the characteristic is selected according to the characteristics of the river basin, or the index system suitable for the river basin is selected purely according to the advice of a river basin environment responsible department or the experience of an environmental protection expert, and the index system with a higher score is promoted to be selected in the condition, so that unified, objective and reasonable evaluation and ranking of all the river basins in the whole country cannot be realized. Aquatic organisms in a body of water include phytoplankton, zooplankton, benthonic animals, fish and large aquatic weeds, but the above-mentioned index system does not perform comprehensive calculation on all aquatic organisms, for example, a BMWP integral system and a Chandler biological index are only suitable for evaluating large benthonic animals, but the types of data are different, the former is used for scoring and evaluating qualitative monitoring data, and the latter is used for quantitatively monitoring data. While Palmer evaluates the results of qualitative monitoring of algae. It is contemplated that the same scoring style will also tend to be different for several basins that select the same scoring style. And for the water body in the flow domain, the scoring mode also causes the phenomenon of unfair evaluation.
Therefore, how to provide a unified, reasonable and objective evaluation method for the quality of the water ecological environment in all the watercourses is an urgent problem to be solved in the field.
Disclosure of Invention
The application provides a water ecology evaluation method based on a deep learning algorithm, which comprises the following steps:
acquiring aquatic organism monitoring data sets, and dividing the aquatic organism data sets to acquire three different data sets; adding optimal individuals and worst individuals into the three divided different data sets to obtain three new data sets; performing data standardization processing on three new data sets; establishing a variable self-encoder for three new data sets after data standardization is completed; in response to completion of the establishment of the variation self-encoder, obtaining distances between the point of each individual in the new data set and the optimal individual, and between the point of each individual and the point corresponding to the worst individual; obtaining distance weights of different species according to the distances between each individual point in the new data set and the optimal individual and the distances between each individual point and the point corresponding to the worst individual; obtaining projection distances according to the distances between the point of each individual in the new data set and the optimal individual and the distances between the point of each individual and the point corresponding to the worst individual; and obtaining comprehensive evaluation scores according to the distance weights and the projection distances of different species.
As above, wherein an aquatic organism monitoring dataset is obtained which is to be evaluatedDividing the aquatic organism monitoring data set into common species, contaminant-resistant species and invasive species data sets according to different attributes of aquatic organisms, respectivelyRepresentation of->For n evaluation of the water body->A data set composed of different aquatic organisms, and。
as above, adding the optimal individual and the worst individual in the three different partitioned data sets to obtain three new data sets includes adding the optimal individual and the worst individual in the common species, contaminant-resistant species and invasive species data sets, respectively.
As above, three new data sets were each data normalized for aquatic organisms according to the following equation:
，
where d represents the old dataset element, s is the new dataset element,monitoring data for aquatic organism identification i for individual j identified as k for dataset,/>For normalized data, +.>A monitoring data sequence for all individuals identified as k for the dataset with aquatic organisms identified as i,/>The structured standardized dataset is defined as +.>A, b, c are symbols of species, a is common, b is stain resistant and c is intrusion.
As above, the variational self-encoder is built for three new data sets after data normalization is completed, and the point of each individual on the new two-dimensional space is acquired.
The water ecology evaluation system based on the deep learning algorithm comprises an original data set acquisition unit, a new data set acquisition unit, a standardized processing unit, a variation self-encoder establishment unit, a corresponding point distance acquisition unit, a distance weight acquisition unit, a projection distance acquisition unit and a comprehensive evaluation score acquisition unit; the original data set acquisition unit is used for acquiring aquatic organism monitoring data sets and dividing the aquatic organism data sets to acquire three different data sets; the new data set acquisition unit is used for adding an optimal individual and a worst individual into the three different divided data sets to obtain three new data sets; the standardized processing unit is used for carrying out data standardization processing on three new data sets; the variable self-encoder establishing unit is used for establishing variable self-encoders for three new data sets after data standardization is completed; the corresponding point distance acquisition unit is used for acquiring the distance between each individual point in the new data set and the optimal individual, and the distance between each individual point and the point corresponding to the worst individual; the distance weight acquisition unit is used for acquiring the distance weights of different species according to the distance between the point of each individual in the new data set and the point corresponding to the optimal individual and the point of each individual and the point corresponding to the worst individual; the projection distance acquisition unit is used for acquiring the projection distance according to the distance between the point of each individual in the new data set and the point corresponding to the optimal individual and the point of each individual and the point corresponding to the worst individual; the comprehensive evaluation score acquisition unit is used for acquiring comprehensive evaluation scores according to the distance weights and the projection distances of different species.
As above, in the raw data set acquisition unit, an aquatic organism monitoring data set to be evaluated is obtainedDividing the aquatic organism monitoring data set into common species, contaminant resistant species and invasive species data sets according to different attributes of aquatic organism, and using +.>Representation of->For n evaluation of the water body->A data set of different aquatic organisms, and +.>。
As described above, the new data set obtaining unit adds the optimal individual and the worst individual in the three different data sets after the division, and the obtaining of the three new data sets includes adding the optimal individual and the worst individual in the common species, the contamination resistant species, and the invasive species data sets, respectively.
As above, in the normalization processing unit, three new data sets are respectively normalized for aquatic organisms according to the following formula:
，
where d represents the old dataset element, s is the new dataset element,monitoring data for aquatic organism identification i for individual j identified as k for dataset,/>For normalized data, +.>A monitoring data sequence for all individuals identified as k for the dataset with aquatic organisms identified as i,/>The structured standardized dataset is defined as +.>A, b, c are symbols of species, a is common, b is stain resistant and c is intrusion.
As described above, the variation self-encoder creation unit creates the variation self-encoder for three new data sets after the data normalization is completed, and obtains the point of each individual on the new two-dimensional space.
The application has the following beneficial effects:
the method can realize unified and reasonable objective evaluation on all the drainage basins through the established evaluation method; meanwhile, the evaluation method provided by the application takes the monitoring data of all biological groups into account, and considers the influence of different species, so that the evaluation result of the obtained aquatic organisms is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings may also be obtained according to these drawings for a person having ordinary skill in the art.
FIG. 1 is a flow chart of a water ecology assessment method based on a deep learning algorithm provided in accordance with an embodiment of the present application;
fig. 2 is a schematic diagram of an internal structure of a water ecology evaluation system based on a deep learning algorithm according to an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application, taken in conjunction with the accompanying drawings, clearly and completely describes the technical solutions of the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
Example 1
As shown in fig. 1, the water ecology evaluation method based on the deep learning algorithm provided in this embodiment specifically includes the following steps:
step S110: the aquatic organism monitoring data set is acquired, and three different data sets are acquired by dividing the aquatic organism data set.
Wherein an aquatic monitoring dataset is obtained which is to be evaluatedAnd the aquatic organism monitoring data set is according to the aquatic organismIs divided into common species, contaminant resistant species and invasive species data sets, respectivelyRepresentation of->For n evaluation of the water body->A data set composed of different aquatic organisms, and。
step S120: adding optimal individuals and worst individuals into the three different data sets after division to obtain three new data sets.
Wherein the optimal individual and the worst individual are added to the common species, contaminant-resistant species, and invasive species data sets (the common species, contaminant-resistant species, and invasive species data sets are collectively referred to as "old data sets"), respectively (the data sets after adding the optimal individual and the worst individual are collectively referred to as "new data sets").
In the common species data set, the monitoring number of one aquatic organism of the optimal individual is the maximum monitoring data of the same aquatic organism of other individuals; the monitoring number of certain aquatic organisms of the worst individuals is the minimum monitoring data of the same aquatic organisms of other individuals.
Adding optimal individuals and worst individuals in common speciation data set to form new common speciation data set。
The optimal individual aquatic organism monitoring number is the minimum monitoring data of the same aquatic organism of other individuals divided by the stain resistance value corresponding to each species according to the stain resistance value of the species in the stain resistance seed data set; the monitoring number of certain aquatic organisms of the worst individuals is the maximum monitoring data of the same aquatic organisms of other individuals multiplied by the stain resistance value corresponding to the species.
Adding optimal individuals and worst individuals into the soil-resistant seed data set to form a new soil-resistant seed data set。
In the invasive species data set, the monitoring number of a certain aquatic organism of the optimal individual is the minimum monitoring data of the same aquatic organism of other individuals; the monitoring number of the aquatic organisms of the worst individuals is the maximum monitoring data of the same aquatic organisms of other individuals.
Adding optimal individuals and worst individuals in invasive species dataset formation to form new invasive species dataset。
Step S130: data normalization processing is performed on the three new data sets.
Three new data sets were normalized separately for aquatic organisms according to the following equation:
，
where d represents the old dataset element, s is the new dataset element,the aquatic life of individual j identified as k in the old dataset is identified as i (since the species of the old three datasets are different for different regions, some species are common in region a and may be invasive in region B, thus when k is a, the common dataset, k is B is the stain resistant dataset, k is c is the invasive dataset), the->For normalized data, +.>A monitoring data sequence for all individuals identified as k for the dataset with aquatic organisms identified as i,/>The structured standardized dataset is defined as +.>。
The data normalization of the three new data sets is accomplished by the above formula.
Step S140: and establishing a variational self-encoder for three new data sets after data normalization is completed.
Wherein the variational self-encoder is built for each of the three new data sets. The variable self-encoder is a self-encoder in the prior art, and the specific structure is not described here.
Building and training different variational self-encoder models for three new data sets to obtain points of each individual on a new two-dimensional spaceThe method is specifically expressed as follows:
，
wherein the method comprises the steps ofFor the common species (or contaminant resistant species), the invasive species data set is identified as +.>Is self-contained in the encoder, +.>Individuals whose representative species is identified as h->Data of->Mapped to points in two dimensions.
And establishing three new variable self-encoders of the data sets through the formula.
Step S150: and in response to completion of the establishment of the variation self-encoder, obtaining the distances between the point of each individual in the new data set and the optimal individual, and between the point of each individual and the point corresponding to the worst individual.
Each new dataset has an optimum, worst individuals, definitionsThe mapping values obtained by the variational self-encoder are respectively (/ for the worst, optimal individuals in the new common species (or stain-resistant species, invasive species) dataset, respectively, identified as k>) And (/ -and)>）。
The Wasserstein distance between any individual in the new dataset and the optimal, worst individual is calculated as follows:
；
wherein the method comprises the steps of
；
Wherein the method comprises the steps of，/>The distance between the individual identified as i in the new dataset and the worst, optimal individual, respectively.
Step S160: and obtaining the distance weights of different species according to the distance between each individual point in the new data set and the optimal individual and the distance between each individual point and the point corresponding to the worst individual.
Before the distance weights of different species are obtained, a distance set between each sample except the optimal sample and the worst sample and the optimal sample and the worst sample is obtained according to the distance between the point of each individual and the optimal individual and the distance between the point of each individual and the point corresponding to the worst individual.
Wherein each sample except the optimal and the worst samples is combined with the optimal, and the distance set between the worst samples is defined as,
，
Wherein the method comprises the steps ofIs->Average value of>Is->A, b, c are the sign of the species, a is common, b is stain resistant c is intrusion;
weights for sequences identified as k for the datasetThe calculation formula is as follows:
。
step S170: and obtaining projection distances according to the distances between the point of each individual and the optimal individual and the distances between the point of each individual and the point corresponding to the worst individual.
Calculating, for each individual, a projection distance between it and the optimal individual; depending on the characteristics of the distance, for any individual i:
；
wherein the method comprises the steps ofThe projection distance of sample i from sample g, identified as k for the dataset.
Step S180: and obtaining comprehensive evaluation scores according to the distance weights and the projection distances of different species.
Based on the obtained weights and the corresponding projection distances, a final score is calculated for each individualI.e., the composite score of the individual sample numbered i.
。
Example two
As shown in fig. 2, the present application provides a water ecology evaluation system based on a deep learning algorithm, wherein the system specifically includes: an original dataset acquisition unit 210, a new dataset acquisition unit 220, a normalization processing unit 230, a variation self-encoder creation unit 240, a corresponding point distance acquisition unit 250, a distance weight acquisition unit 260, a projection distance acquisition unit 270, and a comprehensive evaluation score acquisition unit 280.
The raw dataset acquisition unit 210 is configured to acquire aquatic organism monitoring datasets, and divide the aquatic organism datasets to acquire three different datasets.
Wherein an aquatic monitoring dataset is obtained which is to be evaluatedDividing the aquatic organism monitoring data set into common species, contaminant-resistant species and invasive species data sets according to different attributes of aquatic organisms, respectivelyRepresentation of->For n evaluation of the water body->A data set composed of different aquatic organisms, and。
the new data set obtaining unit 220 is configured to add the optimal individual and the worst individual to the three different data sets after the division, so as to obtain three new data sets.
Wherein optimal individuals and worst individuals are added to the common species, contaminant resistant species, and invasive species data sets, respectively.
In the common species data set, the monitoring number of one aquatic organism of the optimal individual is the maximum monitoring data of the same aquatic organism of other individuals; the monitoring number of certain aquatic organisms of the worst individuals is the minimum monitoring data of the same aquatic organisms of other individuals.
Adding optimal individuals and worst individuals in common speciation data set to form new common speciation data set。
The optimal individual aquatic organism monitoring number is the minimum monitoring data of the same aquatic organism of other individuals divided by the stain resistance value corresponding to each species according to the stain resistance value of the species in the stain resistance seed data set; the monitoring number of certain aquatic organisms of the worst individuals is the maximum monitoring data of the same aquatic organisms of other individuals multiplied by the stain resistance value corresponding to the species.
Adding optimal individuals and worst individuals into the soil-resistant seed data set to form a new soil-resistant seed data set。
In the invasive species data set, the monitoring number of a certain aquatic organism of the optimal individual is the minimum monitoring data of the same aquatic organism of other individuals; the monitoring number of the aquatic organisms of the worst individuals is the maximum monitoring data of the same aquatic organisms of other individuals.
Adding optimal individuals and worst individuals in invasive species dataset formation to form new invasive species dataset。
The normalization processing unit 230 is configured to perform data normalization processing on three new data sets.
Three new data sets were normalized separately for aquatic organisms according to the following equation:
，
where d represents the old dataset element, s is the new dataset element,the aquatic organisms identified as k in the old dataset are identified as i (since the species of the old three datasets are different for different regions, some species are common in region a and may be invasive in region B, thus when k is a common dataset, k is B is a stain resistant dataset, k is c is an invasive dataset, a, B, c is the sign of the species, a is common, B is stain resistant c is invasive).
The data normalization of the three new data sets is accomplished by the above formula.
The variable self-encoder creation unit 240 is used to create a variable self-encoder for three new data sets after data normalization is completed.
The corresponding point distance acquiring unit 250 is configured to acquire a distance between a point of each individual and a point corresponding to an optimal individual, and a point of each individual and a point corresponding to a worst individual.
The distance weight obtaining unit 260 is configured to obtain distance weights of different species according to distances between the point of each individual and the optimal individual, and between the point of each individual and the point corresponding to the worst individual.
Before the distance weights of different species are obtained, a distance set between each sample except the optimal sample and the worst sample and the optimal sample and the worst sample is obtained according to the distance between the point of each individual and the optimal individual and the distance between the point of each individual and the point corresponding to the worst individual.
Wherein each sample except the optimal and the worst samples is combined with the optimal, and the distance set between the worst samples is defined as,
，
Wherein the method comprises the steps ofIs->Average value of>Is->A, b, c are the sign of the species, a is common, b is stain resistant c is intrusion;
the weight calculation formula for the sequence identified as k for the dataset is:
，
the projection distance acquiring unit 270 is configured to acquire a projection distance according to a distance between a point of each individual and a point corresponding to an optimal individual, and a point of each individual and a point corresponding to a worst individual.
Calculating, for each individual, a projection distance between it and the optimal individual; depending on the characteristics of the distance, for any individual i:
，
wherein the method comprises the steps ofThe projection distance of sample i to sample g identified as k for the dataset,
the comprehensive evaluation score acquisition unit 280 is configured to acquire a comprehensive evaluation score according to the distance weights and the projection distances of different species.
Based on the obtained weights and the corresponding projection distances, a final score for each individual, i.e. a comprehensive evaluation score for the sample individual numbered i, is calculated.
。
The application has the following beneficial effects:
the method can realize unified and reasonable objective evaluation on all the drainage basins through the established evaluation method; meanwhile, the evaluation method provided by the application takes the monitoring data of all biological groups into account, and considers the influence of different species, so that the evaluation result of the obtained aquatic organisms is more accurate.
Although the examples referred to in the present application are described for illustrative purposes only and not as limitations on the present application, variations, additions and/or deletions to the embodiments may be made without departing from the scope of the application.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. The water ecology evaluation method based on the deep learning algorithm is characterized by comprising the following steps of:
acquiring aquatic organism monitoring data sets, and dividing the aquatic organism data sets to acquire three different data sets;
adding optimal individuals and worst individuals into the three divided different data sets to obtain three new data sets;
performing data standardization processing on three new data sets;
establishing a variable self-encoder for three new data sets after data standardization is completed;
in response to completion of the establishment of the variation self-encoder, obtaining distances between the point of each individual in the new data set and the optimal individual, and between the point of each individual and the point corresponding to the worst individual;
obtaining distance weights of different species according to the distances between each individual point in the new data set and the optimal individual and the distances between each individual point and the point corresponding to the worst individual;
obtaining projection distances according to the distances between the point of each individual in the new data set and the optimal individual and the distances between the point of each individual and the point corresponding to the worst individual;
acquiring comprehensive evaluation scores according to the distance weights and the projection distances of different species;
obtaining aquatic monitoring data sets to be evaluatedDividing the aquatic organism monitoring data set into common species, contaminant resistant species and invasive species data sets according to different attributes of aquatic organism, and using +.>Representation of->For n evaluation of the water body->A data set of different aquatic organisms, and +.>A, b, c are symbols of species, a is common, b is stain resistant c is intrusion;
acquiring each of the new data setsThe distances between the individual point and the optimal individual, and between the individual point and the point corresponding to the worst individual include: each new dataset has an optimum, worst individuals, definitionsThe worst and best individuals in the new data set identified as k are respectively mapped values obtained by the variational self-encoder are respectively +.>A kind of electronic device；
Wasserstein distance between any individual in the new dataset and the optimal, worst individual:
；
；
；
wherein the method comprises the steps of
；
；
；
Wherein the method comprises the steps of，/>，/>The distance between the individual identified as i in the new dataset and the worst, optimal individual, respectively.
2. The method of claim 1, wherein adding optimal individuals and worst individuals to the three partitioned different data sets to obtain three new data sets comprises adding optimal individuals and worst individuals to the common species, contaminant-resistant species and invasive species data sets, respectively.
3. A deep learning algorithm-based water ecology assessment method according to claim 2 wherein three new data sets are data normalized for aquatic organisms respectively according to the following equation:
；
where d represents the old dataset element, s is the new dataset element,monitoring data for aquatic organism identification i for individual j identified as k for dataset,/>For normalized data, +.>A monitoring data sequence for all individuals identified as k for the dataset with aquatic organisms identified as i,/>The structured standardized dataset is defined as +.>A, b, c are symbols of species, a is common, b is stain resistant and c is intrusion.
4. A method of evaluating water ecology based on a deep learning algorithm as recited in claim 3 wherein a variational self-encoder is built for three new data sets after data normalization is completed to obtain points of each individual in a new two-dimensional space.
5. The water ecology evaluation system based on the deep learning algorithm is characterized by comprising an original data set acquisition unit, a new data set acquisition unit, a standardized processing unit, a variation self-encoder establishment unit, a corresponding point distance acquisition unit, a distance weight acquisition unit, a projection distance acquisition unit and a comprehensive evaluation score acquisition unit;
the original data set acquisition unit is used for acquiring aquatic organism monitoring data sets and dividing the aquatic organism data sets to acquire three different data sets;
the new data set acquisition unit is used for adding an optimal individual and a worst individual into the three different divided data sets to obtain three new data sets;
the standardized processing unit is used for carrying out data standardization processing on three new data sets;
the variable self-encoder establishing unit is used for establishing variable self-encoders for three new data sets after data standardization is completed;
the corresponding point distance acquisition unit is used for acquiring the distance between each individual point in the new data set and the optimal individual, and the distance between each individual point and the point corresponding to the worst individual;
the distance weight acquisition unit is used for acquiring the distance weights of different species according to the distance between the point of each individual in the new data set and the point corresponding to the optimal individual and the point of each individual and the point corresponding to the worst individual;
the projection distance acquisition unit is used for acquiring the projection distance according to the distance between the point of each individual in the new data set and the point corresponding to the optimal individual and the point of each individual and the point corresponding to the worst individual;
the comprehensive evaluation score acquisition unit is used for acquiring comprehensive evaluation scores according to the distance weights and the projection distances of different species;
in the original data set acquisition unit, an aquatic organism monitoring data set needing to be evaluated is obtainedDividing the aquatic organism monitoring data set into common species, contaminant resistant species and invasive species data sets according to different attributes of aquatic organism, and using +.>Representation of->For n evaluation of the water body->A data set of different aquatic organisms, and +.>A, b, c are symbols of species, a is common, b is stain resistant c is intrusion;
the step of obtaining the distance between each individual point and the optimal individual and the distance between each individual point and the point corresponding to the worst individual in the new data set comprises the following steps: each new dataset has an optimum, worst individuals, definitionsThe worst and best individuals in the new data set identified as k are respectively mapped values obtained by the variational self-encoder are respectively +.>A kind of electronic device；
Wasserstein distance between any individual in the new dataset and the optimal, worst individual:
；
；
；
wherein the method comprises the steps of
；
；
；
Wherein the method comprises the steps of，/>，/>The distance between the individual identified as i in the new dataset and the worst, optimal individual, respectively.
6. The water ecology evaluation system based on a deep learning algorithm of claim 5, wherein the new data set obtaining unit adds an optimal individual and a worst individual in the three different data sets after division, and the obtaining of three new data sets includes adding the optimal individual and the worst individual in the common species, the soil-resistant species, and the invasive species data sets, respectively.
7. A deep learning algorithm based water ecology assessment system according to claim 6 wherein in the normalization processing unit, three new data sets are data normalized for aquatic organisms respectively according to the following formula:
；
where d represents the old dataset element, s is the new dataset element,monitoring data for aquatic organism identification i for individual j identified as k for dataset,/>For normalized data, +.>A monitoring data sequence for all individuals identified as k for the dataset with aquatic organisms identified as i,/>The structured standardized dataset is defined as +.>A, b, c are symbols of species, a is common, b is stain resistant and c is intrusion.
8. The water ecology evaluation system based on a deep learning algorithm as recited in claim 7, wherein the variation self-encoder creation unit creates a variation self-encoder for three new data sets after data normalization is completed, and obtains a point of each individual on a new two-dimensional space.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311190762.8A CN116933042B (en) | 2023-09-15 | 2023-09-15 | Water ecology evaluation method and system based on deep learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311190762.8A CN116933042B (en) | 2023-09-15 | 2023-09-15 | Water ecology evaluation method and system based on deep learning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116933042A CN116933042A (en) | 2023-10-24 |
| CN116933042B true CN116933042B (en) | 2023-12-19 |
Family
ID=88377486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311190762.8A Active CN116933042B (en) | 2023-09-15 | 2023-09-15 | Water ecology evaluation method and system based on deep learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116933042B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009212A (en) * | 2019-03-28 | 2019-07-12 | 交通运输部水运科学研究所 | A kind of ecology navigation channel accordance evaluation index system framework |
| AU2020102177A4 (en) * | 2020-04-26 | 2020-10-15 | National Marine Environmental Monitoring Center | Method for Evaluating Ecological Vulnerability of Island |
| CN116482313A (en) * | 2023-04-20 | 2023-07-25 | 成都南易生态环境科技有限公司 | Water ecology monitoring and comprehensive evaluation method based on environment DNA technology |
-
2023
- 2023-09-15 CN CN202311190762.8A patent/CN116933042B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009212A (en) * | 2019-03-28 | 2019-07-12 | 交通运输部水运科学研究所 | A kind of ecology navigation channel accordance evaluation index system framework |
| AU2020102177A4 (en) * | 2020-04-26 | 2020-10-15 | National Marine Environmental Monitoring Center | Method for Evaluating Ecological Vulnerability of Island |
| CN116482313A (en) * | 2023-04-20 | 2023-07-25 | 成都南易生态环境科技有限公司 | Water ecology monitoring and comprehensive evaluation method based on environment DNA technology |
Non-Patent Citations (4)
| Title |
|---|
| Deep learning in diverse intelligent sensor based systems;Yanming Zhu等;《Sensors》;第23卷(第1期);全文 * |
| 基于大型底栖动物完整性指数B-IBI的东江流域水生态健康评价;付岚;江源;刘琦;侯兆疆;;生态环境学报(第08期);全文 * |
| 基于环境管理的河流健康评价体系的研究进展;黄艺;文航;蔡佳亮;;生态环境学报(第04期);全文 * |
| 淡水生态***监测评价指标体系初步研究;朱爱民;;人民长江(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116933042A (en) | 2023-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kruk et al. | Classification of Reynolds phytoplankton functional groups using individual traits and machine learning techniques | |
| Ovaskainen et al. | How to make more out of community data? A conceptual framework and its implementation as models and software | |
| Yeh et al. | Determinism of bacterial metacommunity dynamics in the southern East China Sea varies depending on hydrography | |
| Rabosky et al. | Density-dependent diversification in North American wood warblers | |
| Thackeray et al. | Long‐term change in the phenology of spring phytoplankton: species‐specific responses to nutrient enrichment and climatic change | |
| Spitale et al. | Environmental classification of springs of the Italian Alps and its consistency across multiple taxonomic groups | |
| Franca et al. | Predicting fish community properties within estuaries: influence of habitat type and other environmental features | |
| Jeanmougin et al. | Fine-scale urbanization affects Odonata species diversity in ponds of a megacity (Paris, France) | |
| CN111882138B (en) | Water quality prediction method, device, equipment and storage medium based on space-time fusion | |
| Du et al. | Water quality assessment with hierarchical cluster analysis based on Mahalanobis distance | |
| Reis-Santos et al. | Reconciling differences in natural tags to infer demographic and genetic connectivity in marine fish populations | |
| Yazdian et al. | Relationship between benthic macroinvertebrate bio-indices and physicochemical parameters of water: a tool for water resources managers | |
| Wu et al. | Modeling daily chlorophyll a dynamics in a German lowland river using artificial neural networks and multiple linear regression approaches | |
| Bourel et al. | Machine learning methods for imbalanced data set for prediction of faecal contamination in beach waters | |
| Litsios et al. | Host specialist clownfishes are environmental niche generalists | |
| CN112364238A (en) | Deep learning-based user interest point recommendation method and system | |
| Lawniczak-Malińska et al. | Impact of lake morphology and shallowing on the rate of overgrowth in hard-water eutrophic lakes | |
| Wagner et al. | Improved understanding and prediction of freshwater fish communities through the use of joint species distribution models | |
| Bevilacqua et al. | The use of taxonomic relationships among species in applied ecological research: Baseline, steps forward and future challenges | |
| Schunter et al. | A novel integrative approach elucidates fine-scale dispersal patchiness in marine populations | |
| Hersh et al. | Using identity calls to detect structure in acoustic datasets | |
| Bona et al. | Unfavourable habitat conditions can facilitate hybridisation between the endangered Betula humilis and its widespread relatives B. pendula and B. pubescens | |
| Bobbin et al. | Knowledge discovery for prediction and explanation of blue-green algal dynamics in lakes by evolutionary algorithms | |
| Ondračková et al. | Parasite infection reflects host genetic diversity among non-native populations of pumpkinseed sunfish in Europe | |
| Cao et al. | How to statistically disentangle the effects of environmental factors and human disturbances: a review |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |