CN111507432A - Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium - Google Patents
Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111507432A CN111507432A CN202010615969.5A CN202010615969A CN111507432A CN 111507432 A CN111507432 A CN 111507432A CN 202010615969 A CN202010615969 A CN 202010615969A CN 111507432 A CN111507432 A CN 111507432A
- Authority
- CN
- China
- Prior art keywords
- detected
- weight
- weighing
- picture
- measured
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Accounting & Taxation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Neurology (AREA)
- Finance (AREA)
- Multimedia (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of weighing, and particularly relates to an intelligent agricultural insurance claim settlement weighing method, an intelligent agricultural insurance claim settlement weighing system, electronic equipment and a storage medium. The technical scheme is as follows: the intelligent weighing method for the agricultural insurance claims comprises the following steps: collecting data, cleaning the data, determining the actual size profile of the object to be measured and calculating the weight of the object to be measured. The intelligent agricultural insurance claim settlement weighing system comprises a data acquisition module, a data cleaning module, an actual size contour determining module and a weight calculating module. The invention provides an intelligent agricultural insurance claim settlement weighing method, system, electronic equipment and storage medium for realizing non-contact weighing by utilizing photographing identification.
Description
Technical Field
The invention belongs to the technical field of weighing, and particularly relates to an intelligent agricultural insurance claim settlement weighing method, an intelligent agricultural insurance claim settlement weighing system, electronic equipment and a storage medium.
Background
The live pig breeding insurance covered by the insurance company needs to pay to farmers after the live pigs die. In the traditional mode, after the prospecting personnel arrive at the site, the weight of the dead pig needs to be measured, and the paid amount is checked according to the weight. The weighing adopts two modes, namely, the dead pigs are carried to be weighed to determine the weight, and the weight is estimated according to personal experience. The dead pig is carried in a weighing mode, time and labor are wasted, the dead pig is not sanitary, the estimation mode is strong in subjectivity, and the weight error is large.
The invention patent with the patent application number of CN201711282613.9 discloses an intelligent weighing system, which comprises a processing module, a data distribution module and a plurality of data acquisition modules; the data acquisition module is used for acquiring identification information of a pig to be detected, measuring the quality of the pig to be detected according to a preset period, acquiring a plurality of quality data, calculating the variance between the adjacent quality data after the quality data are sequenced according to time sequence, acquiring a plurality of variance data, and binding the identification information of the pig to be detected and the quality data and then outputting the binding information when the variance data is smaller than a preset threshold value; the data distribution module is connected with the data acquisition module and is used for uploading the obtained quality data and the identification information of the pig to be detected to the processing module; and the processing module stores the received quality data and the identification information of the pig to be detected in a database.
The weighing system described above is essentially aimed at determining whether the measurement is accurate, and only outputting mass data if the variance of two adjacent masses in time sequence is less than a predetermined threshold. When the pig is weighed, the pig still needs to be weighed by the weighing scale. Weighing multiple live pigs easily causes weighing errors, and weighing dead pigs by using a scale has the defects of time and labor waste, insanitation and the like.
Disclosure of Invention
In order to solve the above problems in the prior art, the present invention is directed to an intelligent method, system, electronic device and storage medium for claims settlement in agriculture, which realizes non-contact weighing by photographing and recognition.
The technical scheme adopted by the invention is as follows:
the intelligent weighing method for the agricultural insurance claims comprises the following steps:
s1: collecting data: collecting a common picture of an object to be detected and a reference object;
s2: data cleaning: marking the outline of the reference object and the outline of the object to be detected in the photos, reserving the photos containing the complete outline of the object to be detected, and enabling the object to be detected to face the same direction;
s3: determining the actual size profile of the object to be detected: determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object;
s4: calculating the weight of the object to be measured: in the machine learning stage, comparing the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
As a preferred embodiment of the present invention, in step S4, in the machine learning stage, the specific steps of comparing the actual size profile of the object to be tested in the picture with the weight determined by weighing the object to be tested and obtaining a weight calculation value of the object to be tested according to the deep neural network algorithm include:
s41: representing the picture containing the actual size outline of the object to be detected by using parameters, inputting each parameter into each unit of the neuron function for transformation, and outputting a weight calculation value of the object to be detected;
s42: obtaining loss by the weight calculation value of the object to be measured and the actual weighing weight value through a loss function;
s43: solving the derivative of the loss to the parameter of each neuron through a chain rule; updating parameters using back propagation; obtaining a corrected weight calculation value of the object to be detected;
as a preferred scheme of the invention, after the machine learning is finished, the specific steps of calculating the weight of the object to be measured comprise:
s41: the actual size contour of the object to be measured is represented by parameters, and each parameter is input into each unit of the neuron function to be transformed, and a weight calculation value of the object to be measured is output.
As a preferred embodiment of the present invention, in step S41, when the photo is represented by parameters, the photo is divided into three channels of RGB, and one value in each channel is a number between 0 and 255, i.e., h × w × 3 matrix parameters of the photo are obtained.
As a preferred embodiment of the present invention, step S42 specifically includes: and (3) making a difference value between the weight calculation value of the object to be measured and the actual weighing weight value, then optimizing the difference value, and updating the connection weight of each neuron.
The intelligent agricultural insurance claim settlement weighing system comprises a data acquisition module, a data cleaning module, an actual size contour determining module and a weight calculating module;
the data acquisition module is used for acquiring a common picture of the object to be measured and the reference object and acquiring the accurate weight of the object to be measured determined by the weighing;
the data cleaning module is used for marking the outline of the reference object and the outline of the object to be detected in the picture, reserving the picture containing the complete outline of the object to be detected and enabling the object to be detected to face the same direction;
the actual size contour determining module is used for determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object;
the weight calculation module is used for matching the actual size profile of the object to be measured in the picture with the weight determined by weighing the object to be measured in the machine learning stage, and obtaining a weight calculation value of the object to be measured according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
As a preferable aspect of the present invention, the weight calculation module includes a loss calculation module and a parameter update module;
the loss calculation module represents the picture by parameters, and then each parameter of the picture is input into each unit of the neuron function to be transformed, so that an output value is obtained; obtaining loss by the output value and the actual value through a loss function;
the parameter updating module calculates the derivative of the loss to the parameter of each neuron through a chain rule; parameters are updated using back propagation.
As a preferred scheme of the invention, the system also comprises a newly-built case module, a historical case query module and a temporary case modification module, wherein the newly-built case module is used for inputting the contents including the information of the insured person, the survey information, the harmless treatment information and the survey conclusion information and calculating the loss amount.
An electronic device comprising at least one processor, and at least one memory communicatively coupled to the processor; wherein the memory stores program instructions executable by the processor, and the processor calls the program instructions to perform any of the methods described above.
A non-transitory computer readable storage medium storing computer instructions that cause the computer to perform the method of any of the above.
The invention has the beneficial effects that:
the invention obtains the actual size outline of the object to be measured after comparing with the reference object. And correcting the calculated value of the weight of the object to be measured corresponding to the actual size contour by a deep neural network algorithm, so that the deep neural network has the function of automatically identifying the weight of the object to be measured after repeated learning. After the deep neural network is used for learning for a plurality of times, the photo containing the reference object and the object to be measured is input, the weight value of the object to be measured can be directly output, and the problems of troublesome operation and insanitation when the object to be measured is manually weighed are solved.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a flow chart of a method for calculating the weight of a test object during a machine learning phase;
fig. 3 is a schematic diagram of the architecture of the system of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
Example 1:
as shown in fig. 1, the intelligent method for weighing the agricultural insurance claims comprises the following steps:
s1: collecting data: the common photograph of the test object and the reference object is collected, and the test object is a dead pig in the embodiment. When in shooting, the reference object is placed beside the dead pig for shooting, and the whole outline of the dead pig with the weight to be measured needs to be shot completely.
S2: data cleaning: and marking the outline of the reference object and the outline of the object to be detected in the picture by using 'marking fairy' software, and processing or cleaning the picture which does not meet the standard. The last selected photo should satisfy: the placing position and the orientation of the object to be measured meet the preset standard requirements, and the picture contains the complete outline of the object to be measured. Such as: the test object has its head facing left and its foot facing down, and only one complete test object appears in one picture. During specific operation, PS software can be used for rotating the object to be detected to the position, and redundant objects to be detected in one picture are wiped off.
Due to the limitation of the condition of taking pictures on site, under the condition of meeting the requirements, the photos contain other sundries such as ears of other pigs, pigsties and the like. And after data cleaning, picking out a picture meeting the requirement, and facilitating subsequent judgment of the complete outline of the object to be detected, determination of the actual size outline of the object to be detected and calculation of the weight of the object to be detected.
S3: determining the actual size profile of the object to be detected: and determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object. The actual size of the reference object is fixed and unchanged, after the picture is taken, the proportion of the outline of the reference object to the outline of the object to be measured can be determined, and the model can obtain the actual size of the object to be measured through conversion.
S4: calculating the weight of the object to be measured: and in the machine learning stage, matching the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm. The step of weighing the object to be measured can be performed in advance, and the result of weighing the object to be measured can be input together when the picture is input.
And after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
The deep neural network is used as an algorithm of machine learning, and the basic principle is as follows: the goal of the deep feed-forward network is to approximate some function f. Deep feed-forward networks are typically compounded using a number of different functions, and how the functions are compounded is described by a directed acyclic graph. The simplest case is: directed acyclic graphs are chain structures. Each layer can also be thought of as being made up of many parallel units, each unit representing a vector to scalar function: the input to each cell comes from a number of cells in the previous layer, and the cells compute the output of the cell according to their activation function. Each cell thus resembles a neuron.
Therefore, in the machine learning stage, the specific steps of comparing the actual size profile of the object to be measured in the picture with the weight determined by weighing the object to be measured and obtaining the weight calculation value of the object to be measured according to the deep neural network algorithm comprise:
as shown in fig. 2, in step S4, the step of calculating the weight of the object to be measured includes:
s41: and dividing the outline of the object to be detected containing the label into three channels of RGB, wherein one value in each channel is a number between 0 and 255, and obtaining the matrix parameter of h x w x 3 of one photo. And inputting each parameter of the profile of the object to be measured into each unit of the neuron function for transformation to obtain a weight calculation value of the object to be measured. In the matrix parameters, the value of the area marked with the outline of the object to be measured is different from the value of the area not marked with the outline of the object to be measured.
S42: and obtaining the loss by the weight calculation value of the object to be measured and the actual weighing weight value through a loss function. Each number in the picture is input into each unit of the neuron, and then a calculated value of the weight of the object to be measured is output through a series of transformation, but the value output at the beginning is inaccurate, so a series of optimization methods are needed to correct a series of transformation processes among the neuron, and the final calculated value of the weight of the object to be measured is correct.
The correction method is to make a difference between the correct result and the result output by the neural network, then optimize the difference and update the connection weight of each neuron.
S43: once the loss is found, the gradient of the loss to each parameter can be calculated using a back propagation algorithm and then the parameters are updated. The specific process is as follows: solving the derivative of the loss to the parameter of each neuron through a chain rule; updating parameters using back propagation; and obtaining the corrected weight calculation value of the object to be detected.
After the machine learning is finished, the specific steps of calculating the weight of the object to be measured comprise:
s41: the actual size contour of the object to be measured is represented by parameters, and each parameter is input into each unit of the neuron function to be transformed, and a weight calculation value of the object to be measured is output. Therefore, the calculated value of the weight of the object to be measured corresponding to the actual size contour is corrected by the deep neural network algorithm, so that the deep neural network has the function of automatically identifying the weight of the object to be measured after repeated learning. After the deep neural network is used for learning for a plurality of times, the photo containing the reference object and the object to be measured is input, the weight value of the object to be measured can be directly output, and the problems of troublesome operation and insanitation when the object to be measured is manually weighed are solved.
Example 2:
as shown in FIG. 3, the intelligent agricultural risk claim settlement weighing system comprises a newly-built case module, a historical case query module, a temporary storage case modification module, a data acquisition module, a data cleaning module, an actual size and contour determination module and a weight calculation module.
The new case module is used for inputting the contents including information of the insured person, survey information, harmless treatment information and survey conclusion information and calculating the loss amount. The data acquisition module is used for acquiring common photos of the object to be measured and the reference object and acquiring the accurate weight of the object to be measured determined by the weighing. The data cleaning module is used for marking the outline of the reference object and the outline of the object to be detected in the picture, and processing or cleaning the picture which does not meet the standard, so that the placing position and the orientation of the object to be detected meet the standard requirements, and the picture contains the complete outline of the object to be detected. The actual size contour determining module is used for determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object. The weight calculation module is used for matching the actual size profile of the object to be measured in the picture with the weight determined by weighing the object to be measured in the machine learning stage, and obtaining a weight calculation value of the object to be measured according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
Still further, the weight calculation module includes a loss calculation module and a parameter update module. The loss calculation module represents the picture by parameters, and then each parameter of the picture is input into each unit of the neuron function to be transformed, so that an output value is obtained; and obtaining the loss by the loss function according to the output value and the actual value. The parameter updating module calculates the derivative of the loss to the parameter of each neuron through a chain rule; parameters are updated using back propagation.
The working process is as follows:
processing of input photos: the common photograph of the test object and the reference object is collected, and the test object is a dead pig in the embodiment. When in shooting, the reference object is placed beside the dead pig for shooting, and the whole outline of the dead pig with the weight to be measured needs to be shot completely. And marking the outline of the reference object and the outline of the object to be detected in the picture by using 'marking fairy' software, and processing or cleaning the picture which does not meet the standard. The last selected photo should satisfy: the placing position and the orientation of the object to be measured meet the preset standard requirements, and the picture contains the complete outline of the object to be measured.
Newly building a case: logging in an agricultural insurance claim settlement APP, entering a new case page, and then inputting information insured person information, survey information, harmless treatment information, survey conclusion information and the like according to system prompts. Inputting data-cleaned photograph containing reference substance and object to be tested
Calculating the weight of the object to be measured: and determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object. The actual size of the reference object is fixed and unchanged, after the picture is taken, the proportion of the outline of the reference object to the outline of the object to be measured can be determined, and the model can obtain the actual size of the object to be measured through conversion. And in the machine learning stage, matching the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm. The step of weighing the object to be measured can be performed in advance, and the result of weighing the object to be measured can be input together when the picture is input. And after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
In the process of calculating the weight of the object to be measured, the calculated value of the weight of the object to be measured corresponding to the actual size contour is corrected by the deep neural network algorithm, so that the deep neural network has the function of automatically identifying the weight of the object to be measured after repeated learning. After the deep neural network is used for learning for a plurality of times, the photo containing the reference object and the object to be measured is input, the weight value of the object to be measured can be directly output, and the problems of troublesome operation and insanitation when the object to be measured is manually weighed are solved.
Completing case survey: and after the system identifies the picture and gives the measured weight, recording a survey conclusion, and completing case survey.
Example 3:
an electronic device comprising at least one processor, and at least one memory communicatively coupled to the processor; wherein the memory stores program instructions executable by the processor, and the processor calls the program instructions to perform any of the methods described above.
The electronic device includes a processor (processor), a memory (memory), and a bus; wherein the processor and the memory are in communication with each other via the bus. The processor is used for calling the program instructions in the memory so as to execute the method provided by the above method embodiments. Examples include: collecting a common picture of an object to be detected and a reference object; marking the outline of the reference object and the outline of the object to be detected in the picture, and processing or cleaning the picture which does not meet the standard; determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object; in the machine learning stage, comparing the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
In addition, the logic instructions in the memory may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or make a contribution to the prior art, or may be implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Example 4:
a non-transitory computer readable storage medium storing computer instructions that cause the computer to perform the method of any of the above. The non-transitory computer readable storage medium of the present invention may store computer instructions, and the computer instructions cause a computer to execute the method of the present invention, facilitating automatic control. Examples include: collecting a common picture of an object to be detected and a reference object; marking the outline of the reference object and the outline of the object to be detected in the picture, and processing or cleaning the picture which does not meet the standard; determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object; in the machine learning stage, comparing the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
The invention is not limited to the above alternative embodiments, and any other various forms of products can be obtained by anyone in the light of the present invention, but any changes in shape or structure thereof, which fall within the scope of the present invention as defined in the claims, fall within the scope of the present invention.
Claims (10)
1. The intelligent method for weighing the agricultural insurance claims is characterized by comprising the following steps:
s1: collecting data: collecting a common picture of an object to be detected and a reference object;
s2: data cleaning: marking the outline of the reference object and the outline of the object to be detected in the photos, reserving the photos containing the complete outline of the object to be detected, and enabling the object to be detected to face the same direction;
s3: determining the actual size profile of the object to be detected: determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object;
s4: calculating the weight of the object to be measured: in the machine learning stage, comparing the actual size profile of the object to be detected in the picture with the weight determined by weighing the object to be detected, and obtaining a weight calculation value of the object to be detected according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
2. The intelligent method for claims weighing at risk according to claim 1, wherein in step S4, in the machine learning stage, the specific steps of comparing the actual size profile of the object in the picture with the weight determined by the object weighing, and obtaining the calculated value of the weight of the object according to the deep neural network algorithm include:
s41: representing the picture containing the actual size outline of the object to be detected by using parameters, inputting each parameter into each unit of the neuron function for transformation, and outputting a weight calculation value of the object to be detected;
s42: obtaining loss by the weight calculation value of the object to be measured and the actual weighing weight value through a loss function;
s43: solving the derivative of the loss to the parameter of each neuron through a chain rule; updating parameters using back propagation; and obtaining the corrected weight calculation value of the object to be detected.
3. The intelligent method for weighing claims in agriculture and insurance industries according to claim 1, wherein the specific steps of calculating the weight of the object to be measured after the machine learning is completed include:
s41: the actual size contour of the object to be measured is represented by parameters, and each parameter is input into each unit of the neuron function to be transformed, and a weight calculation value of the object to be measured is output.
4. The intelligent method for weighing claims in agriculture according to claim 2 or 3, wherein in step S41, when the photo is represented by parameters, the photo is divided into three channels RGB, and one value in each channel is a number between 0 and 255, i.e. h x w x 3 matrix parameters of the photo.
5. The intelligent method for weighing the claims in the agricultural insurance claim, according to claim 2, wherein the step S42 is specifically: and (3) making a difference value between the weight calculation value of the object to be measured and the actual weighing weight value, then optimizing the difference value, and updating the connection weight of each neuron.
6. The intelligent agricultural insurance claim settlement weighing system is characterized by comprising a data acquisition module, a data cleaning module, an actual size and contour determining module and a weight calculating module;
the data acquisition module is used for acquiring a common picture of the object to be measured and the reference object and acquiring the accurate weight of the object to be measured determined by the weighing;
the data cleaning module is used for marking the outline of the reference object and the outline of the object to be detected in the picture, reserving the picture containing the complete outline of the object to be detected and enabling the object to be detected to face the same direction;
the actual size contour determining module is used for determining the actual size contour of the object to be detected according to the size ratio of the reference object to the object to be detected in the marked picture and the actual size of the reference object;
the weight calculation module is used for matching the actual size profile of the object to be measured in the picture with the weight determined by weighing the object to be measured in the machine learning stage, and obtaining a weight calculation value of the object to be measured according to a deep neural network algorithm; and after the machine learning is finished, directly obtaining a calculated value of the weight of the object to be detected by a deep neural network algorithm according to the actual size profile of the object to be detected in the picture.
7. The intelligent claims settlement weighing system of claim 6, wherein the weight calculation module comprises a loss calculation module and a parameter update module;
the loss calculation module represents the picture by parameters, and then each parameter of the picture is input into each unit of the neuron function to be transformed, so that an output value is obtained; obtaining loss by the output value and the actual value through a loss function;
the parameter updating module calculates the derivative of the loss to the parameter of each neuron through a chain rule; parameters are updated using back propagation.
8. The intelligent agricultural insurance claim settlement weighing system of claim 6, further comprising a newly-built case module, a historical case query module and a temporary case modification module, wherein the newly-built case module is used for inputting contents including insured person information, survey information, harmless treatment information and survey conclusion information and calculating loss amount.
9. An electronic device, characterized in that: comprises at least one processor and at least one memory communicatively coupled to the processor; wherein the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method according to any one of claims 1 to 5.
10. A non-transitory computer-readable storage medium characterized in that: the non-transitory computer readable storage medium stores computer instructions that cause the computer to perform the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010615969.5A CN111507432A (en) | 2020-07-01 | 2020-07-01 | Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010615969.5A CN111507432A (en) | 2020-07-01 | 2020-07-01 | Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111507432A true CN111507432A (en) | 2020-08-07 |
Family
ID=71875374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010615969.5A Pending CN111507432A (en) | 2020-07-01 | 2020-07-01 | Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111507432A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112330677A (en) * | 2021-01-05 | 2021-02-05 | 四川智迅车联科技有限公司 | High-precision weighing method and system based on image, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764210A (en) * | 2018-06-12 | 2018-11-06 | 焦点科技股份有限公司 | A kind of method and system that the pig based on object of reference identifies again |
| CN108830293A (en) * | 2018-05-08 | 2018-11-16 | 北京佳格天地科技有限公司 | The recognition methods of the weight of animals and device |
| CN109165645A (en) * | 2018-08-01 | 2019-01-08 | 腾讯科技（深圳）有限公司 | A kind of image processing method, device and relevant device |
| CN109459119A (en) * | 2018-10-17 | 2019-03-12 | 北京京东金融科技控股有限公司 | A kind of body weight measurement, equipment and computer readable storage medium |
| CN109632058A (en) * | 2018-12-13 | 2019-04-16 | 北京小龙潜行科技有限公司 | A kind of intelligent group rearing method for measuring weight, device, electronic equipment and storage medium of raising pigs |
| CN109636826A (en) * | 2018-11-13 | 2019-04-16 | 平安科技（深圳）有限公司 | Live pig weight method for measurement, server and computer readable storage medium |
| CN110672189A (en) * | 2019-09-27 | 2020-01-10 | 北京海益同展信息科技有限公司 | Weight estimation method, device, system and storage medium |
-
2020
- 2020-07-01 CN CN202010615969.5A patent/CN111507432A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830293A (en) * | 2018-05-08 | 2018-11-16 | 北京佳格天地科技有限公司 | The recognition methods of the weight of animals and device |
| CN108764210A (en) * | 2018-06-12 | 2018-11-06 | 焦点科技股份有限公司 | A kind of method and system that the pig based on object of reference identifies again |
| CN109165645A (en) * | 2018-08-01 | 2019-01-08 | 腾讯科技（深圳）有限公司 | A kind of image processing method, device and relevant device |
| CN109459119A (en) * | 2018-10-17 | 2019-03-12 | 北京京东金融科技控股有限公司 | A kind of body weight measurement, equipment and computer readable storage medium |
| CN109636826A (en) * | 2018-11-13 | 2019-04-16 | 平安科技（深圳）有限公司 | Live pig weight method for measurement, server and computer readable storage medium |
| CN109632058A (en) * | 2018-12-13 | 2019-04-16 | 北京小龙潜行科技有限公司 | A kind of intelligent group rearing method for measuring weight, device, electronic equipment and storage medium of raising pigs |
| CN110672189A (en) * | 2019-09-27 | 2020-01-10 | 北京海益同展信息科技有限公司 | Weight estimation method, device, system and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 李卓: "基于立体视觉技术的生猪体重估测研究", 《中国博士学位论文全文数据库 农业科技辑》 * |
| 杨艳等: "基于计算机视觉技术估算种猪体重的应用研究", 《农业工程学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112330677A (en) * | 2021-01-05 | 2021-02-05 | 四川智迅车联科技有限公司 | High-precision weighing method and system based on image, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019101786A4 (en) | Intelligent pig group rearing weighing method and apparatus, electronic device and storage medium | |
| CN109785337B (en) | In-column mammal counting method based on example segmentation algorithm | |
| CN110443969B (en) | Fire detection method and device, electronic equipment and storage medium | |
| CN109284771B (en) | Tomato growth model determination method and device | |
| CN104344877B (en) | The field enumeration Weighing method of a kind of chicken and system | |
| CN110991222B (en) | Object state monitoring and sow oestrus monitoring method, device and system | |
| US20120114185A1 (en) | System and method for inventorying vegetal substance | |
| CN110232445B (en) | Cultural relic authenticity identification method based on knowledge distillation | |
| CN111178206A (en) | Building embedded part detection method and system based on improved YOLO | |
| CN111881958B (en) | License plate classification recognition method, device, equipment and storage medium | |
| CN112184791A (en) | Yak weight prediction method based on CNN-LSTM neural network | |
| CN115205695B (en) | Method and system for determining planting strategy according to planting data | |
| KR102180077B1 (en) | Livestock weight estimation system using livestock weight estimation application based on 3D images and Livestock weight estimation method using the same | |
| CN115294185B (en) | Pig weight estimation method and related equipment | |
| CN111507432A (en) | Intelligent weighing method and system for agricultural insurance claims, electronic equipment and storage medium | |
| CN116882079A (en) | Water pump characteristic curve self-adaptive calibration and prediction method | |
| CN109523509B (en) | Method and device for detecting heading stage of wheat and electronic equipment | |
| CN115131670B (en) | Intelligent auditing method, system, device and equipment for rice pictures | |
| KR20210069987A (en) | Crop growth diagnosis system and method using image | |
| CN114283344A (en) | Automatic real-time monitoring method and system for forest ecological hydrological process | |
| CN113762745A (en) | Live pig body shape assessment method and device based on machine vision | |
| CN112489008A (en) | Reinforcing steel bar identification system, reinforcing steel bar identification method and reinforcing steel bar counting and acceptance system | |
| KR102643254B1 (en) | Apparatus and method for calculating shipping information of livestock populations, and shipping information calculating system including the apparatus | |
| CN116524374B (en) | Satellite image real-time processing and distributing method and system | |
| CN112801118B (en) | Pork pig marketing benefit evaluation system and method based on artificial intelligence and big data |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200807 |