WO2023065622A1 - 一种基于分布式节点的水质报告的生成方法及电子设备 - Google Patents
一种基于分布式节点的水质报告的生成方法及电子设备 Download PDFInfo
- Publication number
- WO2023065622A1 WO2023065622A1 PCT/CN2022/087428 CN2022087428W WO2023065622A1 WO 2023065622 A1 WO2023065622 A1 WO 2023065622A1 CN 2022087428 W CN2022087428 W CN 2022087428W WO 2023065622 A1 WO2023065622 A1 WO 2023065622A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- water quality
- water
- real
- confidence
- quality index
- Prior art date
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 605
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 33
- 238000004590 computer program Methods 0.000 claims description 16
- 230000008676 import Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 13
- 239000007787 solid Substances 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000004141 dimensional analysis Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 238000009372 pisciculture Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 claims description 4
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 3
- 229910052760 oxygen Inorganic materials 0.000 claims description 3
- 239000001301 oxygen Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 10
- 238000001514 detection method Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 241000282414 Homo sapiens Species 0.000 description 7
- 230000002265 prevention Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 239000003344 environmental pollutant Substances 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 231100000719 pollutant Toxicity 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000003911 water pollution Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 241000282412 Homo Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000002542 deteriorative effect Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 238000012372 quality testing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 239000012459 cleaning agent Substances 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 238000002425 crystallisation Methods 0.000 description 1
- 230000008025 crystallization Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 239000013502 plastic waste Substances 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000005067 remediation Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Definitions
- the application belongs to the technical field of data processing, and in particular relates to a method for generating a distributed node-based water quality report and electronic equipment.
- the embodiment of the present application provides a distributed node-based water quality report generation method and electronic equipment to solve the existing water quality report generation technology, which requires a large number of experts with professional knowledge to complete, manpower High cost and low efficiency of report generation.
- the first aspect of the embodiment of the present application provides a method for generating a distributed node-based water quality report, including:
- each of the real-time water condition videos is configured with a corresponding confidence weight
- the water quality index information includes at least one water quality index and a weighted weight associated with the water quality index;
- a water quality report for the target watershed is generated based on the water quality grade.
- the second aspect of the embodiment of the present application provides a device for generating a water quality report based on distributed nodes, including:
- the real-time water condition video receiving unit is used to receive the real-time water condition video fed back by each distributed node deployed in the target watershed; each of the real-time water condition videos is configured with a corresponding confidence weight;
- a water quality index information acquisition unit configured to acquire water quality index information associated with the target watershed; the water quality index information includes at least one water quality index and a weighted weight associated with the water quality index;
- the water quality index parameter determination unit is used to import the real-time water condition video into the water quality index analysis algorithm associated with the water quality index information, and output the information corresponding to each of the water quality indexes in the target watershed based on the distributed nodes.
- water quality index parameters ;
- a water quality level calculation unit configured to calculate the water quality level of the target watershed according to the confidence weight, the weighted weight, and the water quality index parameters of all the real-time water condition videos;
- a water quality report generating unit configured to generate a water quality report for the target watershed based on the water quality grade.
- a third aspect of the embodiments of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, when the processor executes the computer program
- the various steps of the first aspect are realized.
- a fourth aspect of the embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, each step of the first aspect is implemented.
- the water quality of the target watershed can be detected as a whole through real-time water condition videos, and the target water quality can be determined before water quality detection.
- the water quality index information corresponding to the watershed and analyze the real-time water condition video through the water quality index analysis algorithm associated with the water quality index information, and determine the water quality index parameters corresponding to the real-time water condition video in multiple water quality index dimensions.
- the confidence weight corresponding to the video, the weighted weight associated with the water quality index, and the water quality index parameters can be calculated to obtain the water quality level corresponding to the target watershed, and a water quality report can be generated based on the water quality level, realizing the purpose of automatically generating a water quality report.
- the embodiment of the present application does not rely on experts to perform water quality detection on water body samples, but can conduct real-time water condition videos through the water quality index analysis algorithm associated with the water quality index information of the target river basin.
- Fig. 1 is the implementation flowchart of a method for generating a water quality report based on distributed nodes provided by the first embodiment of the present application;
- Fig. 2 is a specific implementation flowchart of a method for generating a water quality report based on distributed nodes provided by the second embodiment of the present application;
- Fig. 3 is a schematic diagram of multi-angle acquisition of real-time water condition video provided by an embodiment of the present application
- Fig. 4 is a schematic structural diagram of a confidence recognition network provided by an embodiment of the present application.
- Fig. 5 is a specific implementation flowchart of a distributed node-based water quality report generation method S201 provided in the third embodiment of the present application;
- FIG. 6 is a flow chart of a specific implementation of a distributed node-based water quality report generation method S104 provided in the fourth embodiment of the present application;
- FIG. 7 is a flow chart of a specific implementation of a distributed node-based water quality report generation method S102 provided in the fifth embodiment of the present application;
- Fig. 8 is a specific implementation flowchart of a distributed node-based water quality report generation method S105 provided by the fifth embodiment of the present application;
- Fig. 9 is a specific implementation flowchart of a distributed node-based water quality report generation method S101 provided in the fifth embodiment of the present application.
- Fig. 10 is a structural block diagram of a device for generating a water quality report based on distributed nodes provided by an embodiment of the present application;
- Fig. 11 is a schematic diagram of an electronic device provided by an embodiment of the present application.
- water quality monitoring as an important part of water pollution prevention and control, plays a very important role in water pollution early warning, pollutant monitoring and treatment assessment and prevention.
- Water quality assessment sets weights for each water quality index through various means to realize the quantification of water resource quality, so as to achieve the purpose of water quality supervision.
- the main existing means is to determine the weight of parameters according to their environmental importance and the guideline value suggested by experts. But often the weights of the same parameter vary greatly among different methods, which indicates that assigning appropriate weight values is difficult. On the whole, it requires a lot of professional knowledge to complete the trade-off of water quality indicators, with high labor costs and low penetration rate.
- the water quality of the target watershed can be detected as a whole through real-time water condition videos, and the target water quality can be determined before water quality detection.
- the water quality index information corresponding to the watershed and analyze the real-time water condition video through the water quality index analysis algorithm associated with the water quality index information, and determine the water quality index parameters corresponding to the real-time water condition video in multiple water quality index dimensions.
- the confidence weight corresponding to the video, the weighted weight associated with the water quality index, and the water quality index parameters can be calculated to obtain the water quality level corresponding to the target watershed, and a water quality report can be generated based on the water quality level, realizing the purpose of automatically generating a water quality report and solving the existing problems.
- the generation technology of the water quality report requires a large number of experts with professional knowledge to complete, the labor cost is high, and the report generation efficiency is low.
- Fig. 1 shows the implementation flowchart of the method for generating water quality reports based on distributed nodes provided by the first embodiment of the present application, which is described in detail as follows:
- each distributed node deployed in the target watershed is received; each said real-time water condition video is configured with a corresponding confidence weight.
- the target watershed may be a river, a stream, a river, a lake, or an area with a certain volume of water such as an estuary. Since the water body of the above-mentioned type of watershed is flowing, the water quality is in a state of dynamic change. Therefore, it is necessary to perform water quality testing on the water body of the above-mentioned watershed regularly or in real time to determine the water quality of the watershed.
- electronic equipment can establish communication connections with multiple distributed nodes, and obtain real-time water condition videos fed back by different distributed nodes, so as to be able to Multiple monitoring points (that is, where distributed nodes are placed) monitor water quality to obtain a more comprehensive water quality assessment.
- Each distributed node can obtain the real-time water condition video of the corresponding location, and feed back the collected real-time water condition video to the electronic device through the communication connection with the electronic device.
- the distributed nodes are equipped with a camera module, and the camera module may be located in the target watershed to obtain real-time water condition video of the target watershed.
- the distributed node can be configured with a video optimization algorithm. Before the distributed node sends the real-time water condition video to the electronic device, the original video can be optimized by the above-mentioned video optimization algorithm, and the optimized video (i.e., the real-time water condition video) can be optimized. status video) to electronic devices.
- the distributed node can determine the environmental compensation coefficient according to the acquisition time and the average pixel value of the original video, and adjust the pixel value of each pixel in the original video based on the above environmental compensation coefficient, and the pixel value of each video image frame in the original video. Contrast, so as to generate the corresponding real-time water condition video.
- the distributed nodes can adjust the working mode of the camera module according to the collection time. For example, when collecting during the day, the camera module can be set to the full-color working mode; if collecting at night, the camera module can be set to Set to night work mode.
- the electronic device may send a video feedback instruction to the distributed nodes when a water quality analysis report needs to be generated.
- the distributed nodes can send the collected real-time water quality video to the electronic equipment.
- the distributed node can set the effective feedback time, and the distributed node can collect the video obtained between a certain moment before receiving the video feedback instruction and the moment of receiving the video feedback instruction, as a real-time water condition video feedback to the An electronic device; wherein, the difference between the above-mentioned certain moment and the moment when the video feedback instruction is received is the above-mentioned effective feedback time.
- the distributed node can establish a long connection with the electronic device, and the distributed node can send the video (that is, the real-time water condition video) collected in real time to the electronic device through the long connection, and the electronic device
- the distributed node can send the video (that is, the real-time water condition video) collected in real time to the electronic device through the long connection
- the electronic device Corresponding video databases can be configured for different distributed nodes, and the received real-time water condition videos are stored in the associated video databases.
- the water quality index information associated with the target watershed is acquired; the water quality index information includes at least one water quality index and a weight associated with the water quality index.
- different target watersheds have different focus points.
- different water quality index information may be associated with different target watersheds.
- the types and numbers of water quality indicators contained in different water quality indicator information are different, and the weighting weights associated with different water quality indicators may also be different.
- its associated water quality index information may include: water temperature, pH, and suspended solid density, and the corresponding weights are: 0.3, 0.5, and 0.2; for the second target watershed, its associated water quality
- the index information may include: pH, turbidity, and suspended solid density, and the corresponding weights are 0.4, 0.2, and 0.4, respectively.
- the electronic device may store the correspondence between the target watershed and the water quality index information, and the electronic device may query the above correspondence according to the watershed identifier of the target watershed to determine the water quality index information corresponding to the target watershed.
- the electronic device determines the water quality index information of the target watershed, it can obtain the water quality index analysis algorithm associated with the water quality execution information, and analyze the above-mentioned real-time water condition video through the water quality index algorithm to determine the corresponding water quality index dimensions. Water quality indicator parameters.
- the electronic device can identify the water quality indicators included in the water quality indicator information, and obtain the analysis algorithm associated with different water quality indicators, and the analysis algorithm associated with the water quality indicators is used to output the water quality indicators corresponding to the corresponding water quality indicator dimensions parameter.
- the water quality index information of a certain target watershed includes the following three water quality indexes, namely: temperature, pH and suspended solid density, then the electronic device can obtain three types of data when analyzing the real-time water condition video
- the water quality index analysis algorithms are the first analysis algorithm corresponding to the temperature dimension, the second analysis algorithm corresponding to the pH value, and the third analysis algorithm corresponding to the suspended solid density, and then determine the different real-time water conditions through the above three types of analysis algorithms.
- the water quality index parameters corresponding to the dimensions are the first analysis algorithm corresponding to the temperature dimension, the second analysis algorithm corresponding to the pH value, and the third analysis algorithm corresponding to the suspended solid density, and then determine the different real-time water conditions through the above three types of analysis algorithms.
- the water quality grade of the target watershed is calculated according to the confidence weights, the weighted weights, and the water quality index parameters of all the real-time water condition videos.
- the electronic device calculates and obtains the water quality index parameters, it needs to make a comparison based on the confidence weight and the weighted weight pair.
- the water quality index parameters are converted and superimposed based on all converted water quality index parameters corresponding to all real-time water condition videos, so that the water quality grade of the target watershed can be calculated.
- the water quality level is used to determine the cleanliness of the water body in the target watershed; if the numerical value of the water quality level is larger, it means that the water quality of the target watershed is higher; otherwise, if the numerical value of the water quality level is higher The smaller the value, the lower the degree of crystallization of the water body in the target watershed.
- the method of calculating the above water quality level may be as follows: the electronic device may store the conversion network consisting of the water quality level, and the electronic device may calculate the confidence weight corresponding to each real-time water condition video and based on the real-time water condition video The determined multiple water quality index parameters and corresponding weighted weights are all imported into the above conversion network, and then the above water quality grades can be generated.
- a water quality report of the target watershed is generated based on the water quality level.
- the electronic device can store the corresponding report template, and after determining the water quality level corresponding to the target watershed, the water quality level and the water quality index parameters of each water quality index can be imported into the above report template, thereby generating information about the water quality level.
- after S105 it may also include: if it is detected that the water quality level of the target watershed is lower than the preset level threshold, corresponding water quality early warning information may be generated to notify the user of the above abnormal situation. deal with.
- each distributed node can be configured with a corresponding water quality adjustment module, for example, corresponding cleaning agents, acid-base neutralizers, etc. can be put into the target watershed to perform corresponding abnormal water quality processing.
- the electronic device detects that the water quality level of the target watershed is lower than the level threshold, it can determine the corresponding water quality anomaly processing according to the abnormal water quality index parameters, and send the water quality anomaly processing corresponding to the distributed node.
- An adjustment instruction after receiving the adjustment instruction, the distributed node can execute the corresponding water quality anomaly processing operation through the water quality adjustment module, so as to be able to handle the water quality anomaly in the target watershed.
- the method for generating a water quality report based on distributed nodes deploys corresponding distributed nodes at multiple locations in the target watershed and collects corresponding real-time water condition videos, which can pass real-time
- the water condition video detects the water quality of the target river basin as a whole.
- the water quality index information corresponding to the target river basin is determined, and the real-time water condition video is analyzed through the water quality index analysis algorithm associated with the water quality index information to determine the real-time
- the water quality index parameters corresponding to the water condition video in multiple water quality index dimensions can be calculated according to the confidence weight corresponding to the real-time water condition video, the weighted weight associated with the water quality index and the water quality index parameters, and the water quality level corresponding to the target watershed can be calculated, and based on The water quality grade generates a water quality report, realizing the purpose of automatically generating a water quality report.
- the embodiment of the present application does not rely on experts to perform water quality detection on water body samples, but can conduct real-time water condition videos through the water quality index analysis algorithm associated with the water quality index information of the target river basin. Analyze and determine the corresponding water quality index parameters, and determine the corresponding confidence weight according to the deployment location of each real-time water condition video, which can improve the contribution of the real-time water condition video at a location with a higher confidence level when calculating the water quality level, To further improve the accuracy of water quality grades, according to different target river basins and different focus points, different weighting weights can be configured for different water quality indicators of the target river basin, which can match the water quality grade with the detection content of the target river basin, and improve the Accuracy and flexibility of water quality report generation.
- Fig. 2 shows a specific implementation flowchart of a method for generating a water quality report based on distributed nodes provided by the second embodiment of the present application.
- a method for generating a water quality report based on distributed nodes provided by this embodiment is before receiving the real-time water condition video fed back by each distributed node deployed in the target watershed , also includes: S201 ⁇ S204, the specific details are as follows:
- each of the distributed nodes includes a plurality of shooting angles, and each of the shooting angles corresponds to one of the real-time water condition videos;
- a distributed node can be configured with multiple shooting angles, and a camera module can be configured at a position corresponding to a different shooting angle, and the real-time water condition video corresponding to the shooting angle can be obtained through the camera module.
- FIG. 3 shows a schematic diagram of multi-angle acquisition of real-time water condition video provided by an embodiment of the present application.
- a distributed node includes 3 shooting angles, which are respectively used to obtain real-time water condition videos of the three areas of the water surface, the water and the bottom of the water. Due to different pollutants, their buoyancy will be different. For example, some plastic waste will float on the water surface, while other solid pollutants will exist in the water, and some heavier pollutants will settle at the bottom. Based on this, through Obtaining real-time water condition videos corresponding to different shooting angles can provide an overall understanding of the water quality conditions corresponding to the locations corresponding to the distributed nodes.
- the electronic device may also configure corresponding confidence weights for different shooting angles. For example, if some parts of the target watershed have strong wind and waves on the water surface while the water is relatively calm, the real-time water condition video collected on the water surface will be difficult to identify and perform water quality testing. The video can more accurately determine the water quality of the target watershed. Based on this, the confidence weight of the real-time water condition video from the shooting angle in the water can be greater than the real-time water condition video from the shooting angle of the water surface. Therefore, in addition to configuring different confidence weights according to different locations, the electronic device can also adaptively adjust the confidence weights according to different shooting angles, thereby greatly improving the accuracy of water quality detection in the target watershed.
- the electronic device may generate a network configured with confidence weights through training and learning, so as to output confidence weights corresponding to different distributed nodes. Based on this, the electronic device may configure corresponding initial confidence levels for different shooting angles of a certain distributed node.
- the initial confidence can be manually configured based on the user, and can also be automatically identified according to the location corresponding to the distributed node.
- the training water condition video obtained from the shooting angle is recognized by a preset 3D analysis network, and the 3D video visual analysis is performed to generate 3D characteristic data corresponding to the training water condition video.
- the above-mentioned network for determining confidence weights includes at least two parts, namely a 3D analysis network for performing 3D analysis on videos, and a fully connected network for determining confidence weights based on feature data. Both of the above networks contain adjustable learning parameters.
- the electronic device can import the training water condition video obtained at the corresponding shooting angle into the preset three-dimensional analysis network, perform three-dimensional video visual analysis on the training water condition video, and identify the information contained in the training water condition video.
- the three-dimensional feature data is specifically the data obtained after encoding the training water condition video.
- the three-dimensional feature data is imported into a preset fully connected network, and the confidence level to be verified corresponding to the shooting angle is calculated.
- the 3D feature data can be imported into the fully connected network, and the 3D feature data can be analyzed through the fully connected network to output the 3D feature data in a certain The confidence level to be verified corresponding to the training water condition video obtained from a shooting angle.
- the confidence level to be verified corresponding to the training water condition video obtained from a shooting angle.
- a corresponding normalized network can also be configured, and the normalized network is specifically a softmax function, which can perform logical regression processing on the data output by the above fully connected network , and the normalized value is used as the above-mentioned confidence level to be verified.
- Loss is the loss value
- InitialDegree i is the initial confidence level of the i-th shooting angle
- VerifyDegree i is the unverified confidence level of the i-th shooting angle
- Num 1 is the total number of the shooting angles.
- the electronic device can import the initial confidence level corresponding to the training water condition video and the confidence level to be verified output by the two networks into the preset loss calculation function, so as to determine the loss value corresponding to the above two networks, If the loss value is larger, it means that the degree of distortion is greater; on the contrary, if the value of the loss is smaller, it means that the degree of distortion is smaller.
- the three-dimensional analytical network and the fully connected network are trained based on the loss value to generate a confidence recognition network; the confidence recognition network is based on the three-dimensional analytical network and the fully connected network A network constructed when the corresponding loss value is less than a preset loss threshold; the confidence identification network is used to determine the confidence weight of each of the real-time water condition videos.
- the electronic device can train and learn the three-dimensional analysis network and the fully connected network according to the above loss value, and adjust the parameters in the above two networks until the loss values corresponding to the confidence levels to be verified output by the above two networks converge. , it means that the above two networks have been trained, and the two networks after training are used as the confidence recognition network, so as to determine the confidence weight corresponding to each subsequent real-time water condition video through the confidence recognition network.
- FIG. 4 shows a schematic structural diagram of a confidence recognition network provided by an embodiment of the present application.
- the confidence recognition network specifically includes at least three parts, namely, the three-dimensional analysis network for encoding the real-time water condition video, the fully connected network for analyzing the encoded data, and the numerical analysis for the fully connected network. Normalized softmax function.
- the three-dimensional analysis network and the fully connected network are trained and learned, so that the confidence level that can automatically determine the confidence level weight can be generated.
- the high-degree identification network improves the automation of water quality report generation and the accuracy of identification, without the need for manual configuration of confidence levels, reducing labor costs.
- FIG. 5 shows a specific implementation flowchart of a distributed node-based water quality report generation method S201 provided in the third embodiment of the present application.
- S201 in a method for generating a water quality report based on distributed nodes provided by this embodiment includes: S2011-S2014, which are detailed as follows:
- the deployment location of the distributed node is obtained, and an inflection point angle of the deployment location in the target watershed is queried to obtain a location confidence factor.
- the electronic device can determine the initial confidence level of the training water condition video corresponding to different shooting angles by means of automatic identification, wherein the factors affecting the initial confidence level include at least three aspects, which are: The location factor related to the curvature of the watershed, the velocity factor related to the water flow velocity in the watershed, and the shooting confidence factor related to the obstacle distance in the target watershed.
- the electronic device can obtain factors corresponding to the above three aspects respectively.
- the location factor can be determined by the deployment location of the distributed nodes, each distributed node can be configured with a corresponding location sensor, or the electronic device can mark the deployment location of each distributed node on a preset map, and the distributed After the node is deployed, the general situation is fixed. Therefore, the deployment location corresponding to each distributed node can be determined by querying the pre-generated map. According to the deployment location and the watershed trend of the target watershed, the corresponding location can be determined. The curvature degree of the watershed, which can be represented by the inflection point angle, and the corresponding position confidence factor is generated based on the inflection point angle.
- a gear for measuring the water flow velocity can be placed, and the distributed node can determine the corresponding rotation speed of the motor through the value fed back by the motor associated with the gear, and The water flow velocity corresponding to the position is determined based on the rotation speed.
- the electronic device can By selecting the reference water condition image corresponding to any frame in the training water condition video, the obstacle recognition is performed on the reference water condition image, and the distance value between the obstacle and the camera module is determined, and the distance value can be determined according to the focal length of the camera module And determine the image height of each obstacle in the reference water condition image, and determine the corresponding shooting confidence factor according to the distance value obtained from the above recognition.
- the position confidence factor, the shooting confidence factor and the water flow velocity are imported into a preset confidence conversion model, and the initial confidence corresponding to the shooting angle is calculated;
- the confidence conversion model is specifically :
- InitialDegree i is the initial confidence degree corresponding to the i-th shooting angle; Location is the position confidence factor; ⁇ Base is the preset compensation angle; Speed is the water flow velocity; Dist is the shooting confidence degree; BaseDist is the preset reference distance; ⁇ and ⁇ are the preset coefficients.
- the electronic device may import the factors in the above three aspects into the confidence conversion model, so as to calculate the initial confidence corresponding to the training water condition video obtained from the shooting angle.
- the initial confidence level corresponding to the training water condition video can be automatically determined by obtaining various factors, so that the purpose of automatically setting the confidence level label for the training water condition video can be achieved, and the generation of water quality reports can be further improved. degree of automation.
- FIG. 6 shows a specific implementation flowchart of a distributed node-based water quality report generation method S104 provided by the fourth embodiment of the present application.
- S104 in a method for generating a water quality report based on distributed nodes provided by this embodiment specifically includes S1041 to S1043, which are described in detail as follows:
- the weighted parameters corresponding to the water quality index parameters are calculated according to the weighted weights associated with the water quality index parameters.
- a weighting operation is performed on each of the weighting parameters, and a water quality index factor corresponding to the real-time water condition video is calculated.
- the water quality grade of the target watershed is calculated based on the water quality index factors of all the real-time water condition videos of all the distributed nodes; wherein, the water quality grade is specifically:
- WQI is the water quality grade
- DOC j is the confidence weight corresponding to the jth real-time water condition video
- Weogjt ji is the described weighting parameter of the i-th water quality index factor in the jth real-time water condition video
- Inder ji is the i-th water quality factor in the jth real-time water condition video
- quota is the total number of the water quality indicators
- Num 2 is the total number of the real-time water condition video.
- the electronic device after the electronic device calculates each water quality index parameter, it can adjust the corresponding weighted weight to the water quality index parameter, so as to calculate and obtain the weighted water quality index parameter, that is, the above weighted parameter, and then Then according to the confidence weight of the real-time water condition video corresponding to the water quality index parameter, the weighted parameters are adjusted, and all adjusted weighted parameters corresponding to the real-time water condition video are superimposed, thereby calculating the real-time water condition video Corresponding water quality index parameters.
- a target watershed contains multiple different distributed nodes
- different distributed nodes can also have multiple different shooting angles, and different shooting angles can correspond to a real-time water condition video, so all distributed nodes can be placed in each
- the water quality index factors corresponding to the real-time water condition video obtained from the shooting angle are superimposed to calculate the water quality grade.
- the water quality index factors determined by each real-time water condition video are adjusted through weighted weights and confidence weights, so as to calculate the water quality grade, which can improve the accuracy of the water quality grade.
- Fig. 7 shows a specific implementation flowchart of a distributed node-based water quality report generation method S102 provided by the fifth embodiment of the present application.
- S102 in a method for generating a water quality report based on distributed nodes provided by this embodiment specifically includes S1021 to S1023 , which are described in detail as follows:
- a watershed event corresponding to the target watershed is acquired, and a watershed type of the target watershed is determined based on the watershed event.
- the electronic device can determine the watershed events of the target watershed, and the watershed events are specifically the events that humans engage in in the target watershed, such as boat navigation, breeding, fishing, and berthing of boats, etc., and the electronic device can Watershed events that occur, determine the watershed type corresponding to the target watershed, and determine the specific role of the target watershed in human activities. Different watershed types have different focus on water quality, so different water quality index information can be configured .
- the first index information is used as the water quality index information; the water quality index of the first index information includes turbidity, pH, buoyancy coefficient, and suspended solid density.
- the second index information is used as the water quality index information; the water quality index of the second index information includes turbidity, temperature, pH, suspended solid density, and dissolved oxygen concentration.
- the above-mentioned watershed type may include shipping and shipping type, that is, the target watershed is a canal; the above-mentioned watershed type may also include fish farming type, that is, the target watershed is specifically used for fish farming activities.
- the electronic device can configure corresponding water quality index information according to the focus of attention of different river basin types.
- the matching degree between the water quality report and the target watershed can be improved.
- FIG. 8 shows a specific implementation flowchart of a distributed node-based water quality report generation method S105 provided in the sixth embodiment of the present application.
- S105 in a method for generating a water quality report based on distributed nodes provided by this embodiment specifically includes S1051-S1054, which are detailed as follows:
- a report template associated with the water quality level is acquired; the report template includes a plurality of report items associated with the water quality level.
- the electronic device After the electronic device determines the water quality level corresponding to the target watershed, it can obtain a report template corresponding to the water quality level, wherein, based on the numerical value of the water quality level, it can be divided into the following types:
- different report templates and corresponding report items can be corresponding to different water quality levels. For example, when the water quality is bad, it may need to include the corresponding remediation items, and when the water quality is moderate or good, it can include the corresponding Projects to optimize water quality. Therefore, different water quality grades correspond to different report items, and corresponding report templates are also different.
- the electronic device can determine the corresponding descriptive segment according to the water quality level corresponding to the target watershed, and in order to improve the readability of the water quality report, the corresponding key video segment can be intercepted from the real-time water condition video and added to the above report template , so as to generate the corresponding water quality report.
- different water quality grades are associated with the corresponding report templates, and the corresponding description segments are determined according to the numerical value of the water quality grades, and key video segments are imported into the report templates, which can improve the quality of the water quality templates. Accuracy while further improving readability.
- FIG. 9 shows a specific implementation flowchart of a distributed node-based water quality report generation method S101 provided by the seventh embodiment of the present application.
- S101 in a method for generating a distributed node-based water quality report provided by this embodiment specifically includes S1011-S1013, which are detailed as follows:
- the electronic device in addition to being able to monitor the water quality of the target watershed in real time, that is, to determine the water quality level in real time, the electronic device can also trigger the generation process of the water quality report through an event trigger, wherein the detection of the event trigger is through the distributed
- the distributed node detects that there is a corresponding detection trigger event in the target watershed, it can generate a trigger instruction and send the trigger instruction to the electronic device.
- the distributed node can generate a corresponding trigger instruction, and add the above trigger event to the above trigger instruction, so that electronic The device determines whether to start the process of generating the water quality report according to the event type. If the electronic device detects that the event type is within the preset water quality change event, that is, it is determined that the activities performed by humans on the target watershed will affect the water quality of the target watershed, video feedback instructions can be sent to each distributed node so that each distributed video Send real-time video of water conditions to electronic devices to perform the water quality report generation process.
- the timeliness of water quality change identification can be improved by using distributed nodes to detect whether there is an event of water quality change.
- FIG. 10 shows a structural block diagram of an apparatus for generating a water quality report based on distributed nodes provided by an embodiment of the present application.
- the units included in the electronic device are used to execute the steps in the embodiment corresponding to FIG. 1 .
- FIG. 1 and related descriptions in the embodiment corresponding to FIG. 1 For ease of description, only the parts related to this embodiment are shown.
- the generation device of the distributed node-based water quality report includes:
- the real-time water condition video receiving unit 11 is used to receive the real-time water condition video fed back by each distributed node deployed in the target watershed; each of the real-time water condition videos is configured with a corresponding confidence weight;
- a water quality index information acquisition unit 12 configured to acquire water quality index information associated with the target watershed; the water quality index information includes at least one water quality index and a weighted weight associated with the water quality index;
- the water quality index parameter determination unit 13 is used to import the real-time water condition video into the water quality index analysis algorithm associated with the water quality index information, and output the information about the target watershed in each of the water quality indexes determined based on the distributed nodes. Corresponding water quality index parameters;
- a water quality level calculation unit 14 configured to calculate the water quality level of the target basin according to the confidence weights, the weighted weights, and the water quality index parameters of all the real-time water condition videos;
- a water quality report generation unit 15 configured to generate a water quality report for the target watershed based on the water quality level.
- each of the distributed nodes includes a plurality of shooting angles, and each of the shooting angles corresponds to one of the real-time water condition videos;
- the generating device of the water quality report based on distributed nodes also includes:
- An initial confidence determination unit configured to determine the initial confidence of the training water condition video corresponding to each of the shooting angles
- a three-dimensional feature data generating unit used to identify the training water condition video obtained from the shooting angle through a preset three-dimensional analysis network, perform three-dimensional video visual analysis, and generate three-dimensional feature data corresponding to the training water condition video;
- a confidence degree determination unit to be verified configured to import the three-dimensional feature data into a preset fully connected network, and calculate the confidence degree to be verified corresponding to the shooting angle;
- a loss value calculation unit configured to import the initial confidence degree and the confidence degree to be verified into a preset algorithm loss calculation function, and calculate the loss value corresponding to the three-dimensional analytical network and the fully connected network; the loss calculation The function is specifically:
- Loss is the loss value
- InitialDegree i is the initial degree of confidence of the i-th shooting angle
- VerifyDegree i is the confidence to be verified of the i-th shooting angle
- Num 1 is the total number of the shooting angles
- a training learning unit configured to train the three-dimensional analysis network and the fully connected network based on the loss value to generate a confidence recognition network; based on the loss value, perform the training on the three-dimensional analysis network and the fully connected network Training to generate a confidence recognition network;
- the confidence recognition network is a network constructed when the loss value corresponding to the three-dimensional analysis network and the fully connected network is less than a preset loss threshold; the confidence recognition The network is used to determine the confidence weight of each of the real-time water condition videos; the confidence identification network is used to determine the confidence weight of each of the real-time water condition videos.
- the initial confidence determination unit includes:
- a location confidence factor determining unit configured to obtain the deployment location of the distributed node, and query the inflection point angle of the deployment location in the target watershed to obtain a location confidence factor
- the water flow velocity determination unit is used to identify the rotation speed of the gear placed at the corresponding position of the shooting angle, and determine the water flow velocity associated with the shooting angle;
- a photographing confidence factor determining unit configured to identify obstacles contained in the reference water condition image based on any reference water condition image in the training water condition video collected at the shooting angle, and to correspond to all obstacles based on the shooting angle
- the distance value between the camera modules determines the shooting confidence factor
- An initial confidence calculation unit configured to import the position confidence factor, the shooting confidence factor, and the water flow velocity into a preset confidence conversion model, and calculate the initial confidence corresponding to the shooting angle; the confidence
- the conversion model is specifically:
- the confidence conversion model is specifically:
- InitialDegree i is the initial degree of confidence corresponding to the i-th shooting angle; Location is the position confidence factor; ⁇ Base is the preset compensation angle; Speed is the water velocity; Dist is the shooting degree of confidence; BaseDist is the preset reference distance; ⁇ and ⁇ are the preset coefficients.
- the water quality level calculation unit 13 includes:
- a weighted parameter calculation unit configured to calculate the weighted parameter corresponding to the water quality index parameter according to the weighted weight associated with the water quality index parameter
- the water quality index factor calculation unit is used to perform a weighted operation on each of the weighted parameters according to the confidence weight of the real-time water condition video, and calculate the water quality indicator factor corresponding to the real-time water condition video;
- WQI is the water quality grade
- DOC j is the confidence weight corresponding to the jth real-time water condition video
- Weight ji is the described weighting parameter of the i-th water quality index factor in the jth real-time water condition video
- Inder ji is the i-th water quality factor in the jth real-time water condition video
- quota is the total number of the water quality indicators
- Num 2 is the total number of the real-time water condition video.
- the water quality index information acquisition unit 12 includes:
- a watershed type determining unit configured to acquire a watershed event corresponding to the target watershed, and determine the watershed type of the target watershed based on the watershed event;
- the first type of response unit is used to use the first index information as the water quality index information if the watershed type is the shipping type;
- the water quality index of the first index information includes turbidity, pH, buoyancy coefficient and suspended solids density;
- the second type of response unit is used to use the second index information as the water quality index information if the watershed type is fish farming type; the water quality index of the second index information includes turbidity, temperature, pH, suspended solids density and dissolved oxygen concentration.
- the water quality report generating unit 15 includes:
- a report template determining unit configured to obtain a report template associated with the water quality level; the report template includes a plurality of report items associated with the water quality level;
- a descriptive segment determining unit configured to determine the descriptive segment of each of the report items based on the numerical value of the water quality grade
- a key video segment intercepting unit configured to intercept key video segments from the real-time water condition video of the distributed nodes
- An information importing unit configured to import the description segment and the key video segment into the report template to generate the water quality report.
- the real-time water condition video receiving unit 11 includes:
- a distributed trigger unit configured to parse the event type corresponding to the trigger instruction if a trigger instruction fed back by any distributed node is received;
- a video feedback command sending unit configured to send a video feedback command to each of the distributed nodes if the event type is within a preset water quality change event
- the real-time water condition video feedback unit is configured to receive the real-time water condition video sent by each of the distributed nodes based on the video feedback instruction.
- the electronic device does not rely on experts to perform water quality detection on water samples, but can analyze the real-time water condition video through the water quality index analysis algorithm associated with the water quality index information of the target watershed, and determine the water quality related to it.
- the corresponding water quality index parameters and according to the deployment position of each real-time water condition video, determine the corresponding confidence weight, which can improve the contribution of the real-time water condition video in the location with higher confidence when calculating the water quality level, and further improve the water quality level.
- different weighting weights can be configured for different water quality indicators of the target watershed, which can match the water quality level with the detection content of the target watershed, improving the accuracy of water quality report generation flexibility and flexibility.
- Fig. 11 is a schematic diagram of an electronic device provided by another embodiment of the present application.
- the electronic device 11 of this embodiment includes: a processor 110, a memory 111, and a computer program 112 stored in the memory 111 and operable on the processor 110, such as a distributed node-based A program for generating water quality reports.
- the processor 110 executes the computer program 112 it implements the steps in the above embodiments of the method for generating water quality reports based on distributed nodes, such as S101 to S105 shown in FIG. 1 .
- the processor 110 executes the computer program 112
- the functions of the units in the above-mentioned device embodiments are implemented, for example, the functions of the modules 11 to 15 shown in FIG. 10 .
- the computer program 112 may be divided into one or more units, and the one or more units are stored in the memory 111 and executed by the processor 110 to complete the present application.
- the one or more units may be a series of computer program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program 112 in the electronic device 11 .
- the electronic device may include, but not limited to, a processor 110 and a memory 111 .
- FIG. 11 is only an example of the electronic device 11, and does not constitute a limitation to the electronic device 11. It may include more or less components than shown in the figure, or combine certain components, or different components. , for example, the electronic device may also include an input and output device, a network access device, a bus, and the like.
- the so-called processor 110 may be a central processing unit (Central Processing Unit, CPU), and may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
- a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
- the storage 111 may be an internal storage unit of the electronic device 11 , such as a hard disk or memory of the electronic device 11 .
- the memory 111 can also be an external storage device of the electronic device 11, such as a plug-in hard disk equipped on the electronic device 11, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, flash memory card (Flash Card), etc. Further, the memory 111 may also include both an internal storage unit of the electronic device 11 and an external storage device.
- the memory 111 is used to store the computer program and other programs and data required by the electronic device.
- the memory 111 can also be used to temporarily store data that has been output or will be output.
- each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
- the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Automatic Analysis And Handling Materials Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本申请适用于数据处理技术领域，提供了一种基于分布式节点的水质报告的生成方法及电子设备，包括：接收部署于目标流域的各个分布式节点反馈的实时水况视频；获取所述目标流域关联的水质指标信息；将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；基于所述水质等级生成所述目标流域的水质报告。采用本申请能够提高置信度较高的位置的实时水况视频在计算水质等级时的贡献，进一步提高水质等级的准确性。
Description
本申请要求于2021年10月20日在中国专利局提交的、申请号为202111219181.3的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本申请属于数据处理技术领域，尤其涉及一种基于分布式节点的水质报告的生成方法及电子设备。
近年来，随着社会的飞速发展，人类对于自然资源的使用需求不断增加，其中与人类息息相关的水资源需求量尤为突出。但由于人类活动的介入，水资源的质量一直在恶化，影响和损害群众健康，不利于经济社会持续发展。因此，如何防止水资源污染以及对水资源的防治的工作迫在眉睫。其中水质监控作为水污染防治工作的重要环节，在水污染预警，污染物监测和治理评定预防都发挥着十分重要的作用。
现有的水质报告的生成技术，一般通过对水资源进行采样后，通过专家人工对采集得到的水体样本进行水质评估，并生成对应的水质报告。由此可见，现有的水质报告的生成方法，需要大量拥有专业知识的专家人员才能够完成，人力成本较高，报告生成效率较低。
有鉴于此，本申请实施例提供了一种基于分布式节点的水质报告的生成方法及电子设备，以解决现有的水质报告的生成技术，需要大量拥有专业知识的专家人员才能够完成，人力成本较高，报告生成效率较低的问题。
本申请实施例的第一方面提供了一种基于分布式节点的水质报告的生成方法，包括：
接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重；
获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重；
将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；
根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；
基于所述水质等级生成所述目标流域的水质报告。
本申请实施例的第二方面提供了一种基于分布式节点的水质报告的生成装置，包括：
实时水况视频接收单元，用于接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重；
水质指标信息获取单元，用于获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重；
水质指标参数确定单元，用于将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；
水质等级计算单元，用于根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；
水质报告生成单元，用于基于所述水质等级生成所述目标流域的水质报告。
本申请实施例的第三方面提供了一种电子设备，包括存储器、处理器以及存储在所述 存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现第一方面的各个步骤。
本申请实施例的第四方面提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现第一方面的各个步骤。
实施本申请实施例提供的一种基于分布式节点的水质报告的生成方法及电子设备具有以下有益效果：
本申请实施例通过在目标流域的多个位置部署对应的分布式节点，并采集对应的实时水况视频，能够通过实时水况视频对目标流域整体进行水质检测，在进行水质检测之前，确定目标流域对应的水质指标信息，并通过与该水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定该实时水况视频在多个水质指标维度对应的水质指标参数，根据实时水况视频对应的置信度权重、水质指标关联的加权权重以及水质指标参数，可以计算得到该目标流域对应的水质等级，并基于水质等级生成水质报告，实现了自动生成水质报告的目的。与现有的水质报告的生成技术相比，本申请实施例不依赖专家人员对水体样本进行水质检测，而是可以通过与目标流域的水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定与之对应的水质指标参数，并且根据各个实时水况视频的部署位置，确定对应的置信度权重，能够提高置信度较高的位置的实时水况视频在计算水质等级时的贡献，进一步提高水质等级的准确性，根据目标流域的不同，关注的重点的不同，可以为该目标流域的不同水质指标配置不同的加权权重，能够实现水质等级与目标流域的检测内容相匹配，提高了水质报告生成的准确性以及灵活性。
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本申请第一实施例提供的一种基于分布式节点的水质报告的生成方法的实现流程图；
图2是本申请第二实施例提供的一种基于分布式节点的水质报告的生成方法具体实现流程图；
图3是本申请一实施例提供的多角度采集实时水况视频的示意图；
图4是本申请一实施例提供的置信度识别网络的结构示意图；
图5是本申请第三实施例提供的一种基于分布式节点的水质报告的生成方法S201具体实现流程图；
图6是本申请第四实施例提供的一种基于分布式节点的水质报告的生成方法S104具体实现流程图；
图7是本申请第五实施例提供的一种基于分布式节点的水质报告的生成方法S102具体实现流程图；
图8是本申请第五实施例提供的一种基于分布式节点的水质报告的生成方法S105具体实现流程图；
图9是本申请第五实施例提供的一种基于分布式节点的水质报告的生成方法S101具体实现流程图；
图10是本申请一实施例提供的一种基于分布式节点的水质报告的生成装置的结构框图；
图11是本申请一实施例提供的一种电子设备的示意图。
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
近年来，随着社会的飞速发展，人类对于自然资源的使用需求不断增加，其中与人类息息相关的水资源需求量尤为突出。但由于人类活动的介入，水资源的质量一直在恶化，影响和损害群众健康，不利于经济社会持续发展。因此，如何防止水资源污染以及对水资源的防治的工作迫在眉睫。其中水质监控作为水污染防治工作的重要环节，在水污染预警，污染物监测和治理评定预防都发挥着十分重要的作用。水质评估通过各种手段为每一项水质指标设定权值，实现对水资源质量的量化，从而达到水质监管的目的。现有的主要手段是根据参数的环境重要性以及专家建议的指导值来确定其权重。但是往往同一参数的权重在不同方法之间差别很大，这表明分配适当的权重值很困难。整体上来说，需要大量的专业知识才能完成对水质指标的权衡，人力成本高，普及率低。
本申请实施例通过在目标流域的多个位置部署对应的分布式节点，并采集对应的实时水况视频，能够通过实时水况视频对目标流域整体进行水质检测，在进行水质检测之前，确定目标流域对应的水质指标信息，并通过与该水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定该实时水况视频在多个水质指标维度对应的水质指标参数，根据实时水况视频对应的置信度权重、水质指标关联的加权权重以及水质指标参数，可以计算得到该目标流域对应的水质等级，并基于水质等级生成水质报告，实现了自动生成水质报告的目的，解决了现有的水质报告的生成技术，需要大量拥有专业知识的专家人员才能够完成，人力成本较高，报告生成效率较低的问题。
在本申请实施例中，流程的执行主体为电子设备，该电子设备包括但不限于：服务器、计算机、智能手机、笔记本电脑以及平板电脑等能够执行水质报告的生成流程的设备。图1示出了本申请第一实施例提供的基于分布式节点的水质报告的生成方法的实现流程图，详述如下：
在S101中，接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重。
在本实施例中，目标流域可以为一河流、溪流、江河、湖泊或出海口等具有一定体量水体的区域。上述类型的流域由于水体是流动的，因此水质是处于一个动态变化的状态，因此需要定时或实时对上述流域的水体进行水质检测，以确定该流域的水质情况。
在本实施例中，由于水的流动性，为了提高水质检测的准确性，电子设备可以与多个分布式节点建立通信连接，并获取不同的分布式节点反馈的实时水况视频，从而能够在多个监测点(即分布式节点放置的位置)进行水质监控，以获取更为全面的水质评估。每个分布式节点可以获取对应位置的实时水况视频，并通过与电子设备之间的通信连接，将采集得到的实时水况视频反馈给电子设备。
在本实施例中，分布式节点配置有摄像模块，可以摄像模块具体位于目标流域内，可以获取目标流域的实时水况视频。可选地，分布式节点可以配置有视频优化算法，分布式节点在向电子设备发送实时水况视频之前，可以通过上述视频优化算法对原始视频进行优化，并将优化后的视频(即实时水况视频)发送给电子设备。示例性地，分布式节点可以根据采集时刻以及采集原始视频的平均像素均值，确定环境补偿系数，基于上述环境补偿系数调整原始视频中各个像素点的像素值，以及原始视频中各个视频图像帧的对比度，从而生成对应的实时水况视频。
在一种可能的实现方式中，分布式节点可以根据采集时间调整摄像模块的工作模式，例如在白天采集，则可以将摄像模块设置为全彩工作模式；若在夜晚采集，则可以将摄像模块设置为夜间工作模式。
在一种可能的实现方式中，电子设备可以在需要生成水质分析报告时，向分布式节点发送一个视频反馈指令。分布式节点在接收到该视频反馈指令后，可以将采集得到的实时 水质视频发送给电子设备。其中，分布式节点可以设置有效反馈时间，分布式节点可以在将距离接收到视频反馈指令之前的某一时刻至接收到视频反馈指令的时刻之间采集得到的视频，作为实时水况视频反馈给电子设备；其中，上述某一时刻与接收到视频反馈指令的时刻之间的差值为上述有效反馈时间。
在一种可能的实现方式中，分布式节点可以与电子设备之间建立长连接，分布式节点可以将实时采集到的视频(即实时水况视频)通过上述长连接发送给电子设备，电子设备可以为不同的分布式节点配置对应的视频数据库，将接收得到的实时水况视频存储于关联的视频数据库内。
在S102中，获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重。
在本实施例中，不同的目标流域所关注的重点不同，为了根据实际情况对不同的流域进行水质检测以及生成对应的水质报告，可以为不同的目标流域关联不同的水质指标信息。其中，不同的水质指标信息内包含的水质指标的类型以及个数不同，不同的水质指标关联的加权权重也可以存在差异。
示例性地，对于第一目标流域，其关联的水质指标信息可以包括：水温、酸碱度以及悬浮固体密度，对应的加权权重分别为：0.3、0.5以及0.2；对于第二目标流域，其关联的水质指标信息可以包括：酸碱度、浊度以及悬浮固体密度，对应的加权权重分别为0.4、0.2以及0.4。由此可见，不同的目标流域的关注点不同，所包含的水质指标可以不同，即便存在相同的水质指标，对应的加权权重也可以不同。
在本实施例中，电子设备可以存储有目标流域与水质指标信息之间的对应关系，电子设备可以根据目标流域的流域标识，查询上述对应关系以确定该目标流域对应的水质指标信息。
在S103中，将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数。
在本实施例中，由于水质指标信息的不同，对应的水质指标解析算法也存在差异。电子设备在确定了目标流域的水质指标信息后，可以获取与该水质执行信息关联的水质指标解析算法，并通过该水质指标算法对上述实时水况视频进行解析，以确定不同水质指标维度对应的水质指标参数。
在一种可能的实现方式中，电子设备可以识别水质指标信息中包含的水质指标，并获取不同水质指标关联的解析算法，水质指标关联的解析算法用于输出对应的水质指标维度对应的水质指标参数。示例性地，某一目标流域的水质指标信息包含有以下三个水质指标，分别为：温度、酸碱度以及悬浮固体密度，则电子设备在对实时水况视频进行解析时，可以获取三种类型的水质指标解析算法，分别为温度维度对应的第一解析算法、酸碱度对应的第二解析算法以及悬浮固体密度对应的第三解析算法，继而通过上述三种类型的解析算法分别确定实时水况视频不同维度对应的水质指标参数。
在S104中，根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级。
在本实施例中，由于不同的实时水况视频对应不同的置信度权重，不同的水质指标对应不同的加权权重，电子设备在计算得到水质指标参数后，需要对基于置信度权重以及加权权重对水质指标参数进行转换，并基于所有实时水况视频对应的所有转换后的水质指标参数进行叠加，从而能够计算的目标流域的水质等级。
在本实施例中，水质等级用于确定目标流域的水体的洁净程度；若该水质等级的数值越大，则表示该目标流域的水体的洁净程度越高；反之，若该水质等级的数值越小，则表示该目标流域的水体的结晶程度越低。
在一种可能的实现方式中，计算上述水质等级的方式可以为：电子设备可以存储由水 质等级的转换网络，电子设备将每个实时水况视频对应的置信度权重以及基于该实时水况视频确定的多个水质指标参数和对应的加权权重均导入到上述转换网络内，则可以生成上述水质等级。
在S105中，基于所述水质等级生成所述目标流域的水质报告。
在本实施例中，电子设备可以存储由对应的报告模板，在确定了目标流域对应的水质等级后，可以将水质等级以及各个水质指标的水质指标参数导入到上述报告模板内，从而生成关于该目标流域的水质报告。
在一种可能的实现方式中，在S105之后，还可以包括：若检测到目标流域的水质等级低于预设的等级阈值，则可以生成对应的水质预警信息，以通知用户对上述异常情况进行处理。
在一种可能的实现方式中，每个分布式节点可以配置有对应的水质调节模块，例如可以向目标流域投放对应的清洁剂、酸碱中和剂等执行相应的水质异常处理。在该情况下，电子设备在检测到目标流域的水质等级低于等级阈值时，则可以根据异常的水质指标参数，确定与之对应的水质异常处理，并向分布式节点发送水质异常处理对应的调节指令，分布式节点在接收到该调节指令后，可以通过水质调节模块执行对应的水质异常处理操作，从而能够对目标流域的水质异常情况进行处理。
以上可以看出，本申请实施例提供的一种基于分布式节点的水质报告的生成方法通过在目标流域的多个位置部署对应的分布式节点，并采集对应的实时水况视频，能够通过实时水况视频对目标流域整体进行水质检测，在进行水质检测之前，确定目标流域对应的水质指标信息，并通过与该水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定该实时水况视频在多个水质指标维度对应的水质指标参数，根据实时水况视频对应的置信度权重、水质指标关联的加权权重以及水质指标参数，可以计算得到该目标流域对应的水质等级，并基于水质等级生成水质报告，实现了自动生成水质报告的目的。与现有的水质报告的生成技术相比，本申请实施例不依赖专家人员对水体样本进行水质检测，而是可以通过与目标流域的水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定与之对应的水质指标参数，并且根据各个实时水况视频的部署位置，确定对应的置信度权重，能够提高置信度较高的位置的实时水况视频在计算水质等级时的贡献，进一步提高水质等级的准确性，根据目标流域的不同，关注的重点的不同，可以为该目标流域的不同水质指标配置不同的加权权重，能够实现水质等级与目标流域的检测内容相匹配，提高了水质报告生成的准确性以及灵活性。
图2示出了本申请第二实施例提供的一种基于分布式节点的水质报告的生成方法的具体实现流程图。参见图2，相对于图1所述实施例，本实施例提供的一种基于分布式节点的水质报告的生成方法在所述接收部署于目标流域的各个分布式节点反馈的实时水况视频之前，还包括：S201～S204，具体详述如下：
进一步地，每个所述分布式节点包含有多个拍摄角度，每个所述拍摄角度对应一个所述实时水况视频；
在所述接收部署于目标流域的各个分布式节点反馈的实时水况视频之前，还包括：
在S201中，确定各个所述拍摄角度对应的训练水况视频的初始置信度。
在本实施例中，一个分布式节点内可以配置有多个拍摄角度，在不同的拍摄角度对应的位置，可以配置一个摄像模块，通过该摄像模块获取该拍摄角度下对应的实时水况视频。示例性地，图3示出了本申请一实施例提供的多角度采集实时水况视频的示意图。参见图3所示，一个分布式节点中包含有3个拍摄角度，分别用于获取水面、水中以及水底这三个区域的实时水况视频。由于污染物的不同，其浮力会存在差异，例如部分塑料垃圾会漂浮于水面上，而其他固体污染物则会存在于水中，而部分较重的污染物则会沉降于水底，基于此，通过获取不同拍摄角度对应的实时水况视频，能够对分布式节点对应的位置对应的水质情况有一个整体的了解。
在本实施例中，电子设备除了可以为不同的分布式节点配置对应的置信度权重外，还可以为不同的拍摄角度配置对应的置信度权重。例如，若目标流域的部分位置，水面的风浪较大，而水中较为平静，则水面采集得到的实时水况视频则较难进行识别并进行水质检测，而由于水中较为平静，水中的实时水况视频则较能够准确确定目标流域的水质情况，基于此，水中的拍摄角度的实时水况视频的置信度权重，可以大于水面的拍摄角度的实时水况视频。因此，电子设备除了可以根据位置不同，配置不同的置信度权重外，还可以根据拍摄角度的不同，对置信度权重进行适应性调整，从而大大提高了目标流域的水质检测的准确性。
在本实施例中，电子设备可以通过训练学习的方式，生成配置置信度权重的网络，以便输出不同分布式节点对应的置信度权重。基于此，电子设备可以为某一分布式节点的不同拍摄角度配置对应的初始置信度。其中，该初始置信度可以基于用户进行手动配置，还可以根据该分布式节点对应的位置进行自动识别。
在S202中，通过预设的三维解析网络识别所述拍摄角度获取的训练水况视频进行三维视频视觉解析，生成所述训练水况视频对应的三维特征数据。
在本实施例中，上述确定置信度权重的网络包含至少两部分，分别为用于对视频进行三维解析的三维解析网路，以及基于特征数据确定置信度权重的全连接网络。上述两个网络均包含可调整的学习参量。基于此，电子设备可以为将在对应拍摄角度获取得到的训练水况视频导入到预设的三维解析网络中，对训练水况视频进行三维视频视觉解析，识别得到该训练水况视频中包含的拍摄对象，并基于所有识别得到的拍摄对象以及视频背景特征，生成与之对应的三维特征数据，该三维特征数据具体为对训练水况视频进行编码后得到的数据。
在S203中，将所述三维特征数据导入到预设的全连接网络，计算所述拍摄角度对应的待验证置信度。
在本实施例中，电子设备在生成了训练水况视频对应的三维特征数据后，可以将该三维特征数据导入到全连接网络，能够通过全连接网络对三维特征数据进行解析，以输出在某一拍摄角度获取得到的训练水况视频对应的待验证置信度。其中，由于上述全连接网络以及三维解析网络是未进行学习训练得到的，因此输出的待验证置信度与实际的置信度会存在一定的偏差，因此，需要根据待验证置信度对上述两个网络进行训练学习。
在一种可能的实现方式中，在全连接网络之后，还可以配置有对应的归一化网络，该归一化网络具体为一softmax函数，可以对上述全连接网络输出的数据进行逻辑归回处理，并归一化后的数值作为上述待验证置信度。
在S204中，将所述初始置信度以及待验证置信度导入到预设的算法损失计算函数内，计算所述三维解析网络以及所述全连接网络对应的损失值；所述损失计算函数具体为：
其中，Loss为所述损失值，InitialDegree
i为第i个拍摄角度的所述初始置信度；VerifyDegree
i为第i个拍摄角度的所述待验证置信度；Num
1为所述拍摄角度的总数。
在本实施例中，电子设备可以将训练水况视频对应的初始置信度以及通过两个网络输出的待验证置信度导入到预设的损失计算函数，从而确定上述两个网络对应的损失值，若该损失值越大，则表示失真度越大；反之，若该损失值越小，则表示失真度越小。
在S204中，基于所述损失值对所述三维解析网络以及所述全连接网络进行训练，生成置信度识别网络；所述置信度识别网路是在所述三维解析网络以及所述全连接网络对应的所述损失值小于预设的损失阈值时构建的网络；所述置信度识别网络用于确定各个所述实时水况视频的所述置信度权重。
在本实施例中，电子设备可以根据上述损失值对三维解析网络以及全连接网络进行训练学习，调整上述两个网络中的参数，直到上述两个网络输出的待验证置信度对应的损失值收敛，则表示对于上述两个网络已训练完毕，并将训练完成后的两个网络作为置信度识别网络，以通过该置信度识别网络确定后续的各个实时水况视频对应的置信度权重。
示例性地，图4示出了本申请一实施例提供的置信度识别网络的结构示意图。参见图4所示，该置信度识别网络具体包含至少三部分，分别为对实时水况视频进行编码的三维解析网络，以及对编码数据进行解析的全连接网络，以及对于全连接网络的数值进行归一化处理的softmax函数。
在本申请实施例中，通过确定多个训练水况视频对应的初始置信度，并基于该初始置信度对三维解析网络以及全连接网络进行训练学习，从而能够生成可以自动确定置信度权重的置信度识别网络，提高了水质报告生成的自动化程度以及识别的准确性，无需人工手动配置置信度，降低了人力成本。
图5示出了本申请第三实施例提供的一种基于分布式节点的水质报告的生成方法S201的具体实现流程图。参见图5，相对于图2所述实施例，本实施例提供的一种基于分布式节点的水质报告的生成方法中S201包括：S2011～S2014，具体详述如下：
在S2011中，获取所述分布式节点的部署位置，并查询所述部署位置在所述目标流域中的拐点角度，得到位置置信因子。
在S2012中，识别在拍摄角度对应位置放置的齿轮的转动速度，确定所述拍摄角度关联的水流流速。
在S2013中，基于所述拍摄角度采集的训练水况视频中的任一基准水况图像，识别所述基准水况图像内包含的障碍物，并基于所有障碍物与拍摄角度对应的摄像模块之间的距离值，确定拍摄置信因子。
在本实施例中，电子设备可以通过自动识别的方式，确定不同拍摄角度对应的训练水况视频的初始置信度，其中，影响上述初始置信度的因子至少包含三个方面，分别为与位置中流域弯曲程度相关的位置因子、流域中水流流速相关的流速因子，以及与该目标流域中障碍物距离相关的拍摄置信因子。
基于此，电子设备可以分别获取上述三个方面对应的因子。其中，对于位置因子可以通过分布式节点的部署位置确定，每个分布式节点可以配置有对应的位置传感器，或者电子设备可以在预设的地图上标记出各个分布式节点的部署位置，分布式节点在部署完成后，一般情况是固定不变的，因此可以通过查询预先生成的地图，确定各个分布式节点对应的部署位置，根据该部署位置以及目标流域的流域走势，可以确定该位置对应的流域的弯曲程度，该弯曲程度可以通过拐点角度来表示，并基于该拐点角度生成与之对应的位置置信度因子。
对于水流流速，在每个分布式节点对应的拍摄角度下，可以放置有一个用于测量水流流速的齿轮，分布式节点可以通过齿轮关联的马达反馈的数值，确定该马达对应的转动速度，并基于该转动速度确定该位置对应的水流流速。
对于与障碍物相关的拍摄置信因子，由于拍摄过程中障碍物与摄像模块之间的距离越近，则遮挡画面的程度越大，无法准确表征水质相关的内容，在该情况下，电子设备可以通过训练水况视频中选取任一帧对应的基准水况图像，对该基准水况图像进行障碍物识别，并确定障碍物与摄像模块之间的距离值，该距离值可以根据摄像模块的焦距以及各个障碍物在基准水况图像中的像高确定，并根据上述识别得到的距离值，确定对应的拍摄置信因子。
在S2014中，将所述位置置信因子、所述拍摄置信因子以及所述水流流速导入到预设的置信度转换模型，计算所述拍摄角度对应的初始置信度；所述置信度转换模型具体为：
其中，InitialDegree
i为第i个拍摄角度对应的所述初始置信度；Location为所述位置置信因子；ω
Base为预设的补偿角度；Speed为所述水流流速；Dist为所述拍摄置信度；BaseDist为预设的基准距离；α和β为预设的系数。
在本实施例中，电子设备可以将上述三个方面的因子导入到置信度转换模型内，从而计算得到该拍摄角度获取得到的训练水况视频对应的初始置信度。
在本申请实施例中，通过获取多个方面的因子以自动确定该训练水况视频对应的初始置信度，能够实现自动对训练水况视频进行置信度标签设置的目的，进一步提高了水质报告生成的自动化程度。
图6示出了本申请第四实施例提供的一种基于分布式节点的水质报告的生成方法S104的具体实现流程图。参见图6，与图1的实施例相比，本实施例提供的一种基于分布式节点的水质报告的生成方法中S104具体包括S1041～S1043，具体详述如下：
在S1041中，根据所述水质指标参数关联的加权权重，计算所述水质指标参数对应的加权参数。
在S1042中，根据所述实时水况视频的置信度权重，对各个所述加权参数进行加权运算，计算所述实时水况视频对应的水质指标因子。
在S1043中，基于所有所述分布式节点的所有所述实时水况视频的水质指标因子，计算所述目标流域的水质等级；其中，所述水质等级具体为：
其中，WQI为所述水质等级；DOC
j为第j个实时水况视频对应的置信度权重；Weogjt
ji为第j个实时水况视频中第i个水质指标因子的所述加权参数；Inder
ji为第j个实时水况视频中第i个水质因子；quota为所述水质指标的总数；Num
2为所述实时水况视频的总数。
在本实施例中，电子设备在计算得到各个水质指标参数后，可以将与其对应的加权权 重对水质指标参数进行加权调整，从而计算能得到加权后的水质指标参数，即上述的加权参数，然后再根据确定该水质指标参数对应的实时水况视频的置信度权重，对加权参数进行调整，并将该实时水况视频对应的所有调整后的加权参数进行叠加，从而计算得到该实时水况视频对应的水质指标参数。由于一个目标流域包含有多个不同的分布式节点，不同的分布式节点还可以存在多个不同的拍摄角度，不同的拍摄角度可以对应一个实时水况视频，因此可以将所有分布式节点在各个拍摄角度获取得到的实时水况视频对应的水质指标因子进行叠加，从而计算得到水质等级。
在本申请实施例中，通过加权权重以及置信度权重对各个实时水况视频确定的水质指标因子进行调整，从而计算得到水质等级，能够提高水质等级的准确性。
图7示出了本申请第五实施例提供的一种基于分布式节点的水质报告的生成方法S102的具体实现流程图。参见图7，与图1的实施例相比，本实施例提供的一种基于分布式节点的水质报告的生成方法中S102具体包括S1021～S1023，具体详述如下：
在S1021中，获取所述目标流域对应的流域事件，基于所述流域事件确定所述目标流域的流域类型。
在本实施例中，电子设备可以确定目标流域的流域事件，该流域事件具体为人类在目标流域从事的事件，例如船只航行、养殖、捕捞、船只的停泊等等，电子设备可以根据该目标流域发生的流域事件，确定该目标流域所对应的流域类型，确定该目标流域具体在人类活动中所起到的作用，不同的流域类型对于水质所关注的重点不同，因此可以配置不同的水质指标信息。
在S1022中，若所述流域类型为船航货运类型，则将第一指标信息作为所述水质指标信息；所述第一指标信息的水质指标包含浊度、酸碱度、浮力系数以及悬浮固体密度。
在S1023中，若所述流域类型为渔业养殖类型，则将第二指标信息作为所述水质指标信息；所述第二指标信息的水质指标包含浊度、温度、酸碱度、悬浮固体密度以及溶解氧浓度。
在本实施例中，上述流域类型可以包含船航货运类型，即该目标流域为一运河；上述流域类型还可以包含渔业养殖类型，即该目标流域具体用于渔业养殖活动。基于此，电子设备可以根据不同的流域类型的关注侧重点，配置与之对应的水质指标信息。
在本申请实施例中，通过确定目标流域的流域类型，配置对应的水质指标信息，能够提高水质报告与目标流域之间的匹配度。
图8示出了本申请第六实施例提供的一种基于分布式节点的水质报告的生成方法S105的具体实现流程图。参见图8，与图1-图7的实施例相比，本实施例提供的一种基于分布式节点的水质报告的生成方法中S105具体包括S1051～S1054，具体详述如下：
在S1051中，获取与所述水质等级关联的报告模板；所述报告模板包含与所述水质等级关联的多个报告项目。
在S1052中，基于所述水质等级的数值大小，确定各个所述报告项目的描述语段。
在S1053中，从所述分布式节点的所述实时水况视频中截取关键视频段。
在S1054中，将所述描述语段以及所述关键视频段导入所述报告模板内，生成所述水质报告。
在本实施例中，电子设备在确定了目标流域对应的水质等级后，可以获取与水质等级对应的报告模板，其中，基于水质等级的数值大小，可以划分为以下各种类型：
1)优(WQI＝90-100)
2)良(WQI＝70-89)
3)中(WQI＝50-69)
4)坏(WQI＝25-49)
5)极差(WQI＝0-24)
其中，根据不同的水质等级可以对应不同的报告模板，以及对应的报告项目，例如水 质为坏时，则可以需要包含对应的整治方式的项目，而在水质为中以及良时，可以包含对应的优化水质的项目。因此，不同的水质等级对应的报告项目不同，则对应的报告模板也不同。电子设备可以根据目标流域对应的水质等级，确定与之对应的描述语段，并且为了提高水质报告的可读性，可以从实时水况视频中截取对应的关键视频段，添加到上述报告模板内，从而生成对应的水质报告。
在本申请实施例中，为不同的水质等级关联与之对应的报告模板，并根据水质等级的数值大小，确定对应的描述语段，并将关键视频段导入到报告模板内，能够提高水质模板准确性的同时，进一步提高可读性。
图9示出了本申请第七实施例提供的一种基于分布式节点的水质报告的生成方法S101的具体实现流程图。参见图9，与图1-图7的实施例相比，本实施例提供的一种基于分布式节点的水质报告的生成方法中S101具体包括S1011～S1013，具体详述如下：
在S1011中，若接收到任一分布式节点反馈的触发指令，则解析所述触发指令对应的事件类型。
在S1012中，若所述事件类型在预设的水质变化事件内，则向各个所述分布式节点发送视频反馈指令。
在S1013中，接收各个所述分布式节点基于所述视频反馈指令发送的所述实时水况视频。
在本实施例中，电子设备除了能够实时监控目标流域的水质情况外，即实时确定水质等级外，还可以通过事件触发的方式触发水质报告的生成流程，其中，该事件触发的检测是通过分布式节点完成的，分布式节点在检测到目标流域存在对应的检测触发事件时，可以生成一个触发指令，并将触发指令发生给电子设备。例如，检测到有大量船只在目标流域上航行，或者检测到目标流域存在大面积的捕捞行为时，分布式节点可以生成对应的触发指令，并将上述触发事件添加到上述触发指令内，以便电子设备根据事件类型确定是否需要启动水质报告的生成流程。若电子设备检测到事件类型在预设的水质变化事件内，即确定人类在目标流域上执行的活动会影响目标流域的水质，则可以向各个分布式节点发送视频反馈指令，以便各个分布式视频向电子设备发送实时水况视频，以执行水质报告的生成流程。
在本申请实施例中，通过分布式节点来检测是否存在水质变化的事件，能够提高水质变化识别的及时性。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
图10示出了本申请一实施例提供的一种基于分布式节点的水质报告的生成装置的结构框图，该电子设备包括的各单元用于执行图1对应的实施例中的各步骤。具体请参阅图1与图1所对应的实施例中的相关描述。为了便于说明，仅示出了与本实施例相关的部分。
参见图10，所述基于分布式节点的水质报告的生成装置包括：
实时水况视频接收单元11，用于接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重；
水质指标信息获取单元12，用于获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重；
水质指标参数确定单元13，用于将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；
水质等级计算单元14，用于根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；
水质报告生成单元15，用于基于所述水质等级生成所述目标流域的水质报告。
可选地，每个所述分布式节点包含有多个拍摄角度，每个所述拍摄角度对应一个所述 实时水况视频；
所述基于分布式节点的水质报告的生成装置还包括：
初始置信度确定单元，用于确定各个所述拍摄角度对应的训练水况视频的初始置信度；
三维特征数据生成单元，用于通过预设的三维解析网络识别所述拍摄角度获取的训练水况视频进行三维视频视觉解析，生成所述训练水况视频对应的三维特征数据；
待验证置信度确定单元，用于将所述三维特征数据导入到预设的全连接网络，计算所述拍摄角度对应的待验证置信度；
损失值计算单元，用于将所述初始置信度以及待验证置信度导入到预设的算法损失计算函数内，计算所述三维解析网络以及所述全连接网络对应的损失值；所述损失计算函数具体为：
其中，Loss为所述损失值，InitialDegree
i为第i个拍摄角度的所述初始置信度；VerifyDegree
i为第i个拍摄角度的所述待验证置信度；Num
1为所述拍摄角度的总数；
训练学习单元，用于基于所述损失值对所述三维解析网络以及所述全连接网络进行训练，生成置信度识别网络；基于所述损失值对所述三维解析网络以及所述全连接网络进行训练，生成置信度识别网络；所述置信度识别网路是在所述三维解析网络以及所述全连接网络对应的所述损失值小于预设的损失阈值时构建的网络；所述置信度识别网络用于确定各个所述实时水况视频的所述置信度权重；所述置信度识别网络用于确定各个所述实时水况视频的所述置信度权重。
可选地，所述初始置信度确定单元包括：
位置置信因子确定单元，用于获取所述分布式节点的部署位置，并查询所述部署位置在所述目标流域中的拐点角度，得到位置置信因子；
水流流速确定单元，用于识别在拍摄角度对应位置放置的齿轮的转动速度，确定所述拍摄角度关联的水流流速；
拍摄置信因子确定单元，用于基于所述拍摄角度采集的训练水况视频中的任一基准水况图像，识别所述基准水况图像内包含的障碍物，并基于所有障碍物与拍摄角度对应的摄像模块之间的距离值，确定拍摄置信因子；
初始置信度计算单元，用于将所述位置置信因子、所述拍摄置信因子以及所述水流流速导入到预设的置信度转换模型，计算所述拍摄角度对应的初始置信度；所述置信度转换模型具体为：
所述置信度转换模型具体为：
其中，InitialDegree
i为第i个拍摄角度对应的所述初始置信度；Location 为所述位置置信因子；ω
Base为预设的补偿角度；Speed为所述水流流速；Dist为所述拍摄置信度；BaseDist为预设的基准距离；α和β为预设的系数。
可选地，所述水质等级计算单元13包括：
加权参数计算单元，用于根据所述水质指标参数关联的加权权重，计算所述水质指标参数对应的加权参数；
水质指标因子计算单元，用于根据所述实时水况视频的置信度权重，对各个所述加权参数进行加权运算，计算所述实时水况视频对应的水质指标因子；
加权叠加单元，用于基于所有所述分布式节点的所有所述实时水况视频的水质指标因子，计算所述目标流域的水质等级；其中，所述水质等级具体为：
其中，WQI为所述水质等级；DOC
j为第j个实时水况视频对应的置信度权重；Weight
ji为第j个实时水况视频中第i个水质指标因子的所述加权参数；Inder
ji为第j个实时水况视频中第i个水质因子；quota为所述水质指标的总数；Num
2为所述实时水况视频的总数。
可选地，所述水质指标信息获取单元12包括：
流域类型确定单元，用于获取所述目标流域对应的流域事件，基于所述流域事件确定所述目标流域的流域类型；
第一类型响应单元，用于若所述流域类型为船航货运类型，则将第一指标信息作为所述水质指标信息；所述第一指标信息的水质指标包含浊度、酸碱度、浮力系数以及悬浮固体密度；
第二类型响应单元，用于若所述流域类型为渔业养殖类型，则将第二指标信息作为所述水质指标信息；所述第二指标信息的水质指标包含浊度、温度、酸碱度、悬浮固体密度以及溶解氧浓度。
可选地，所述水质报告生成单元15包括：
报告模板确定单元，用于获取与所述水质等级关联的报告模板；所述报告模板包含与所述水质等级关联的多个报告项目；
描述语段确定单元，用于基于所述水质等级的数值大小，确定各个所述报告项目的描述语段；
关键视频段截取单元，用于从所述分布式节点的所述实时水况视频中截取关键视频段；
信息导入单元，用于将所述描述语段以及所述关键视频段导入所述报告模板内，生成所述水质报告。
可选地，所述实时水况视频接收单元11包括：
分布式触发单元，用于若接收到任一分布式节点反馈的触发指令，则解析所述触发指令对应的事件类型；
视频反馈指令发送单元，用于若所述事件类型在预设的水质变化事件内，则向各个所述分布式节点发送视频反馈指令；
实时水况视频反馈单元，用于接收各个所述分布式节点基于所述视频反馈指令发送的所述实时水况视频。
因此，本申请实施例提供的电子设备同样不依赖专家人员对水体样本进行水质检测，而是可以通过与目标流域的水质指标信息关联的水质指标解析算法对实时水况视频进行解析，确定与之对应的水质指标参数，并且根据各个实时水况视频的部署位置，确定对应的置信度权重，能够提高置信度较高的位置的实时水况视频在计算水质等级时的贡献，进一步提高水质等级的准确性，根据目标流域的不同，关注的重点的不同，可以为该目标流域的不同水质指标配置不同的加权权重，能够实现水质等级与目标流域的检测内容相匹配，提高了水质报告生成的准确性以及灵活性。
图11是本申请另一实施例提供的一种电子设备的示意图。如图11所示，该实施例的电子设备11包括：处理器110、存储器111以及存储在所述存储器111中并可在所述处理器110上运行的计算机程序112，例如基于分布式节点的水质报告的生成程序。所述处理器110执行所述计算机程序112时实现上述各个基于分布式节点的水质报告的生成方法实施例中的步骤，例如图1所示的S101至S105。或者，所述处理器110执行所述计算机程序112时实现上述各装置实施例中各单元的功能，例如图10所示模块11至15功能。
示例性的，所述计算机程序112可以被分割成一个或多个单元，所述一个或者多个单元被存储在所述存储器111中，并由所述处理器110执行，以完成本申请。所述一个或多个单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述所述计算机程序112在所述电子设备11中的执行过程。
所述电子设备可包括，但不仅限于，处理器110、存储器111。本领域技术人员可以理解，图11仅仅是电子设备11的示例，并不构成对电子设备11的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述电子设备还可以包括输入输出设备、网络接入设备、总线等。
所称处理器110可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器111可以是所述电子设备11的内部存储单元，例如电子设备11的硬盘或内存。所述存储器111也可以是所述电子设备11的外部存储设备，例如所述电子设备11上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，所述存储器111还可以既包括所述电子设备11的内部存储单元也包括外部存储设备。所述存储器111用于存储所述计算机程序以及所述电子设备所需的其他程序和数据。所述存储器111还可以用于暂时地存储已经输出或者将要输出的数据。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施 例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。
Claims (10)
- 一种基于分布式节点的水质报告的生成方法，其特征在于，包括：接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重；获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重；将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；基于所述水质等级生成所述目标流域的水质报告。
- 根据权利要求1所述的生成方法，其特征在于，每个所述分布式节点包含有多个拍摄角度，每个所述拍摄角度对应一个所述实时水况视频；在所述接收部署于目标流域的各个分布式节点反馈的实时水况视频之前，还包括：确定各个所述拍摄角度对应的训练水况视频的初始置信度；通过预设的三维解析网络识别所述拍摄角度获取的训练水况视频进行三维视频视觉解析，生成所述训练水况视频对应的三维特征数据；将所述三维特征数据导入到预设的全连接网络，计算所述拍摄角度对应的待验证置信度；将所述初始置信度以及待验证置信度导入到预设的算法损失计算函数内，计算所述三维解析网络以及所述全连接网络对应的损失值；所述损失计算函数具体为：
其中，Loss为所述损失值，InitialDegree i为第i个拍摄角度的所述初始置信度；VerifyDegree i为第i个拍摄角度的所述待验证置信度；Num 1为所述拍摄角度的总数；基于所述损失值对所述三维解析网络以及所述全连接网络进行训练，生成置信度识别网络；所述置信度识别网路是在所述三维解析网络以及所述全连接网络对应的所述损失值小于预设的损失阈值时构建的网络；所述置信度识别网络用于确定各个所述实时水况视频的所述置信度权重。 - 根据权利要求2所述的生成方法，其特征在于，所述确定各个所述拍摄角度对应的训练水况视频的初始置信度，包括：获取所述分布式节点的部署位置，并查询所述部署位置在所述目标流域中的拐点角度，得到位置置信因子；识别在拍摄角度对应位置放置的齿轮的转动速度，确定所述拍摄角度关联的水流流速；基于所述拍摄角度采集的训练水况视频中的任一基准水况图像，识别所述基准水况图像内包含的障碍物，并基于所有障碍物与拍摄角度对应的摄像模块之间的距离值，确定拍摄置信因子；将所述位置置信因子、所述拍摄置信因子以及所述水流流速导入到预设的置信度转换模型，计算所述拍摄角度对应的初始置信度；所述置信度转换模型具体为：
其中，InitialDegree i为第i个拍摄角度对应的所述初始置信度；Location为所述位置置信因子；ω Base为预设的补偿角度；Speed为所述水流流速；Dist为所述拍摄置信度；BaseDist为预设的基准距离；α和β为预设的系数。 - 根据权利要求1所述的生成方法，其特征在于，所述根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级，包括：根据所述水质指标参数关联的加权权重，计算所述水质指标参数对应的加权参数；根据所述实时水况视频的置信度权重，对各个所述加权参数进行加权运算，计算所述实时水况视频对应的水质指标因子；基于所有所述分布式节点的所有所述实时水况视频的水质指标因子，计算所述目标流域的水质等级；其中，所述水质等级具体为：
其中，WQI为所述水质等级；DOC j为第j个实时水况视频对应的置信度权重；Weight ji为第j个实时水况视频中第i个水质指标因子的所述加权参数；Index ji为第j个实时水况视频中第i个水质因子；quota为所述水质指标的总数；Num 2为所述实时水况视频的总数。 - 根据权利要求1所述的生成方法，其特征在于，所述获取所述目标流域关联的水质指标信息，包括：获取所述目标流域对应的流域事件，基于所述流域事件确定所述目标流域的流域类型；若所述流域类型为船航货运类型，则将第一指标信息作为所述水质指标信息；所述第一指标信息的水质指标包含浊度、酸碱度、浮力系数以及悬浮固体密度；若所述流域类型为渔业养殖类型，则将第二指标信息作为所述水质指标信息；所述第二指标信息的水质指标包含浊度、温度、酸碱度、悬浮固体密度以及溶解氧浓度。
- 根据权利要求1-5任一项所述的生成方法，其特征在于，所述基于所述水质等级生成所述目标流域的水质报告，包括：获取与所述水质等级关联的报告模板；所述报告模板包含与所述水质等级关联的多个报告项目；基于所述水质等级的数值大小，确定各个所述报告项目的描述语段；从所述分布式节点的所述实时水况视频中截取关键视频段；将所述描述语段以及所述关键视频段导入所述报告模板内，生成所述水质报告。
- 根据权利要求1-5任一项所述的生成方法，其特征在于，所述接收部署于目标流域的各个分布式节点反馈的实时水况视频，包括：若接收到任一分布式节点反馈的触发指令，则解析所述触发指令对应的事件类型；若所述事件类型在预设的水质变化事件内，则向各个所述分布式节点发送视频反馈指令；接收各个所述分布式节点基于所述视频反馈指令发送的所述实时水况视频。
- 一种基于分布式节点的水质报告的生成的控制装置，其特征在于，包括：实时水况视频接收单元，用于接收部署于目标流域的各个分布式节点反馈的实时水况视频；每个所述实时水况视频配置有对应的置信度权重；水质指标信息获取单元，用于获取所述目标流域关联的水质指标信息；所述水质指标信息包含有至少一个水质指标以及所述水质指标关联的加权权重；水质指标参数确定单元，用于将所述实时水况视频导入与所述水质指标信息关联的水质指标解析算法，输出基于所述分布式节点确定的关于所述目标流域在各个所述水质指标对应的水质指标参数；水质等级计算单元，用于根据所有所述实时水况视频的所述置信度权重、所述加权权重以及所述水质指标参数，计算所述目标流域的水质等级；水质报告生成单元，用于基于所述水质等级生成所述目标流域的水质报告。
- 一种电子设备，其特征在于，所述电子设备包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时如权利要求1至7任一项所述方法的步骤。
- 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至7任一项所述方法的步骤。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| ZA2023/05525A ZA202305525B (en) | 2021-10-20 | 2023-05-22 | Method and device for generating water quality report based on distributed nodes, electronic device, and storage medium |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111219181.3 | 2021-10-20 | ||
| CN202111219181.3A CN113673970B (zh) | 2021-10-20 | 2021-10-20 | 一种基于分布式节点的水质报告的生成方法及电子设备 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023065622A1 true WO2023065622A1 (zh) | 2023-04-27 |
Family
ID=78550647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/087428 WO2023065622A1 (zh) | 2021-10-20 | 2022-04-18 | 一种基于分布式节点的水质报告的生成方法及电子设备 |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN113673970B (zh) |
| WO (1) | WO2023065622A1 (zh) |
| ZA (1) | ZA202305525B (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117630309A (zh) * | 2023-10-17 | 2024-03-01 | 水利部珠江水利委员会水文局 | 一种河道水质智能监测方法及系统 |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673970B (zh) * | 2021-10-20 | 2022-02-22 | 长视科技股份有限公司 | 一种基于分布式节点的水质报告的生成方法及电子设备 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110887792A (zh) * | 2019-12-17 | 2020-03-17 | 深圳慧格科技服务咨询有限公司 | 一种基于无人机高光谱影像的河流水质快速监测系统 |
| CN112101790A (zh) * | 2020-09-16 | 2020-12-18 | 清华大学合肥公共安全研究院 | 一种水体监测视频及数据联动预警方法及系统 |
| US20200408094A1 (en) * | 2018-09-30 | 2020-12-31 | China University Of Mining And Technology | Mine exploitation based on stoping, separation and filling control |
| CN113673970A (zh) * | 2021-10-20 | 2021-11-19 | 长视科技股份有限公司 | 一种基于分布式节点的水质报告的生成方法及电子设备 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6195393B1 (en) * | 1998-07-06 | 2001-02-27 | General Instrument Corporation | HDTV video frame synchronizer that provides clean digital video without variable delay |
| CN106404785A (zh) * | 2013-03-18 | 2017-02-15 | 丁雪强 | 通过水面影像获得水质情况的水情监测系统 |
| CN107065984B (zh) * | 2017-02-14 | 2019-04-09 | 江苏财经职业技术学院 | 基于物联网的水产养殖水质监测和预警系统 |
| CN112388655B (zh) * | 2020-12-04 | 2021-06-04 | 齐鲁工业大学 | 一种基于触觉振动信号与视觉图像融合的抓取物识别方法 |
-
2021
- 2021-10-20 CN CN202111219181.3A patent/CN113673970B/zh active Active
-
2022
- 2022-04-18 WO PCT/CN2022/087428 patent/WO2023065622A1/zh unknown
-
2023
- 2023-05-22 ZA ZA2023/05525A patent/ZA202305525B/en unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200408094A1 (en) * | 2018-09-30 | 2020-12-31 | China University Of Mining And Technology | Mine exploitation based on stoping, separation and filling control |
| CN110887792A (zh) * | 2019-12-17 | 2020-03-17 | 深圳慧格科技服务咨询有限公司 | 一种基于无人机高光谱影像的河流水质快速监测系统 |
| CN112101790A (zh) * | 2020-09-16 | 2020-12-18 | 清华大学合肥公共安全研究院 | 一种水体监测视频及数据联动预警方法及系统 |
| CN113673970A (zh) * | 2021-10-20 | 2021-11-19 | 长视科技股份有限公司 | 一种基于分布式节点的水质报告的生成方法及电子设备 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117630309A (zh) * | 2023-10-17 | 2024-03-01 | 水利部珠江水利委员会水文局 | 一种河道水质智能监测方法及系统 |
| CN117630309B (zh) * | 2023-10-17 | 2024-05-28 | 水利部珠江水利委员会水文局 | 一种河道水质智能监测方法及系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113673970B (zh) | 2022-02-22 |
| ZA202305525B (en) | 2023-12-20 |
| CN113673970A (zh) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023065622A1 (zh) | 一种基于分布式节点的水质报告的生成方法及电子设备 | |
| WO2021087985A1 (zh) | 模型训练方法、装置、存储介质及电子设备 | |
| US20210182611A1 (en) | Training data acquisition method and device, server and storage medium | |
| Yu et al. | Segmentation and measurement scheme for fish morphological features based on Mask R-CNN | |
| CN111222526B (zh) | 一种渔船实时捕捞行为识别方法、装置、设备及存储介质 | |
| CN110738235B (zh) | 肺结核判定方法、装置、计算机设备及存储介质 | |
| CN114926726B (zh) | 基于多任务网络的无人船感知方法及相关设备 | |
| WO2022078002A1 (zh) | 一种图像处理方法、装置、设备及可读存储介质 | |
| CN110751643A (zh) | 一种水质异常检测方法、装置及设备 | |
| CN108900622B (zh) | 基于物联网的数据融合方法、装置与计算机可读存储介质 | |
| WO2021052020A1 (zh) | 振动检测系统 | |
| WO2021169464A1 (zh) | 养殖场风险预测方法、装置、设备及存储介质 | |
| CN109614924A (zh) | 一种基于深度学习算法的水面垃圾检测方法 | |
| CN113409167A (zh) | 一种水质异常分析方法及装置 | |
| CN114998679A (zh) | 深度学习模型的在线训练方法、装置、设备及存储介质 | |
| CN113484867A (zh) | 一种基于成像声呐封闭空间下鱼群密度探测方法 | |
| CN113589230B (zh) | 一种基于联合优化网络的目标声源定位方法及系统 | |
| CN108414016A (zh) | 一种基于大数据技术的污水管网监测系统 | |
| CN111178381A (zh) | 一种禽蛋估重、图像处理方法及装置 | |
| CN116466058B (zh) | 水质检测数据处理方法、水质评估系统、设备及介质 | |
| CN116311089B (zh) | 一种基于图像处理的污水水质数据智能分析方法和系统 | |
| CN112434717B (zh) | 一种模型训练方法及装置 | |
| CN114399669B (zh) | 目标检测方法和装置 | |
| CN115843733A (zh) | 一种基于机器视觉的河蟹养殖电子食台装置及工作方法 | |
| CN113376160A (zh) | 一种扫地机器人针对动物排泄物的识别及处理方法和装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22882247 Country of ref document: EP Kind code of ref document: A1 |