WO2021027156A1 - Procédé et appareil d'enquête sur la récolte utilisant une vidéo, et dispositif d'ordinateur - Google Patents

Procédé et appareil d'enquête sur la récolte utilisant une vidéo, et dispositif d'ordinateur Download PDF

Info

Publication number
WO2021027156A1
WO2021027156A1 PCT/CN2019/118245 CN2019118245W WO2021027156A1 WO 2021027156 A1 WO2021027156 A1 WO 2021027156A1 CN 2019118245 W CN2019118245 W CN 2019118245W WO 2021027156 A1 WO2021027156 A1 WO 2021027156A1
Authority
WO
WIPO (PCT)
Prior art keywords
assisting
video
requesting
picture
terminal
Prior art date
Application number
PCT/CN2019/118245
Other languages
English (en)
Chinese (zh)
Inventor
詹友能
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2021027156A1 publication Critical patent/WO2021027156A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/90335—Query processing
    • G06F16/90344—Query processing by using string matching techniques
    • 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
    • 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
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
    • 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
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture

Definitions

  • This application relates to the field of image recognition technology, and in particular to a video-based crop survey method, device and computer equipment.
  • Insurance survey refers to the comprehensive analysis of the insured target through scientific and systematic professional inspection, testing and survey methods, and then the scientific and systematic valuation of damages.
  • crops as the subject of insurance have been promoted to a certain extent. If the insured has suffered damage to the crops after insuring the crops, the surveyor of the insurance company is required to investigate the damage on the spot.
  • the surveyors have high requirements for professional knowledge in the agricultural field, which makes it impossible to accurately assess the risk situation by themselves.
  • consulting professionals are involved in the damage assessment, it is impossible to establish contact with professionals in real time, resulting in low processing efficiency.
  • the embodiments of this application provide a video-based crop survey method, device, computer equipment, and storage medium, which are designed to solve the problem that in the prior art, when surveying and determining the damage of the insured object of crops, the damage is generally determined manually due to lack of professional knowledge. The result of loss determination is inaccurate and the problem of inefficiency is handled.
  • an embodiment of the present application provides a video-based crop survey method, which includes:
  • the video information between the assisting end and the requesting end is obtained and saved; among them, the video information between the assisting end and the requesting end includes the corresponding information obtained by the assisting end Assisting end video data and requesting end video data correspondingly obtained by the requesting end; the requesting end video data includes crop video information and requester audio data; the assisting end video data includes assisting end audio data;
  • an embodiment of the present application provides a video-based crop survey device, which includes:
  • a location obtaining unit configured to obtain location information of the requesting end if a survey assistance request instruction issued by the requesting end is detected
  • the assisting terminal set acquiring unit is used to acquire the assisting terminal whose distance from the location information is within a preset distance threshold to form the assisting terminal set;
  • the video acquisition unit is used to obtain and save the video information between the assisting terminal and the requesting terminal if a successful video connection instruction is detected between the requesting terminal and the assisting terminal set; among them, the video information between the assisting terminal and the requesting terminal Including the assisting end video data correspondingly obtained by the assisting end and the requesting end video data correspondingly obtained by the requesting end; the requesting end video data includes crop video information and requester audio data; the assisting end video data includes the assisting end audio data;
  • An audio recognition unit configured to perform audio extraction on the video information to obtain an audio extraction result, and obtain the text of the audio extraction result through a voice recognition model to obtain the recognition result;
  • the result sending unit is configured to send the recognition result to the requesting end.
  • an embodiment of the present application provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and running on the processor, and the processor executes the computer
  • the program implements the video-based crop survey method described in the first aspect.
  • the embodiments of the present application also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the above-mentioned The video-based crop survey method described in one aspect.
  • FIG. 1 is a schematic diagram of an application scenario of a video-based crop survey method provided by an embodiment of the application
  • FIG. 2 is a schematic flowchart of a video-based crop survey method provided by an embodiment of the application
  • FIG. 3 is a schematic diagram of another process of a video-based crop survey method provided by an embodiment of the application.
  • FIG. 4 is a schematic block diagram of a video-based crop survey device provided by an embodiment of the application.
  • FIG. 5 is another schematic block diagram of a video-based crop survey device provided by an embodiment of the application.
  • Fig. 6 is a schematic block diagram of a computer device provided by an embodiment of the application.
  • Figure 1 is a schematic diagram of an application scenario of a video-based crop surveying method provided by an embodiment of this application
  • Figure 2 is a schematic flowchart of a video-based crop surveying method provided by an embodiment of the application, which is video-based
  • the crop survey method of is applied to a server, and the method is executed by application software installed in the server.
  • the method includes steps S110 to S150.
  • Crop insurance is based on various artificially planted crops, including food crop insurance (mainly covering cereals, wheat, potato and legumes), cash crop insurance (mainly covering cotton, hemp, oil, sugar cane) Such as sugar, tobacco and medicinal crops), horticultural crop insurance (mainly covering fruits, vegetables and flowers and other crops).
  • Crop insurance can cover only one risk, or it can cover mixed liability and all risk liability. In the subject of insurance, it can be the insurance of the harvest value of crops (that is, 5-7 of the average annual harvest value of the past three years is used as the insured amount.
  • the insurance company will compensate the difference for the insufficient amount of insurance) It can also be crop production cost insurance (taking the production cost of planting crops as the subject of insurance, and the insurance company is responsible for compensating the actual cost loss of the crop after the disaster within the scope of the planting cost).
  • the following rules can be set:
  • the insurer is not responsible for compensation if the planted rice suffers from natural disasters within the scope of insurance liability, but the loss rate is below 30%.
  • the actual loss rate is between 30% (inclusive) and 70% to be paid proportionally, and 70% (including 70%) or more are paid in full.
  • the compensation shall be calculated according to the ratio of the insured area to the actual planting area.
  • the surveyor of the insurance company receives a claim from the insured, he needs to go to the scene to investigate the damage. If the investigator is unable to accurately assess the risk situation by himself during the on-site investigation, he can invite other salesmen in the company or experts outside the company to assist in the investigation process such as risk assessment and other investigation operations through video conference.
  • One is the requesting terminal which is an intelligent terminal used by the surveyor.
  • the surveyor arrives at the inspection site to survey the target crop, he can click the survey assistance virtual button on the requesting terminal to trigger the sending of a survey assistance request instruction to the server.
  • the server feedbacks the assisting terminal set to select one of the assisting terminals to establish a connection, online video can be carried out with the assisting terminal to assist in crop survey and damage assessment.
  • the second is the server, which is used to receive the survey assistance request instruction from the requesting end, feed back a set of eligible assistants according to the positioning information of the requesting end, and obtain and save the video information after the connection between the requesting end and the assisting end is established.
  • the saved video information can also be extracted from text information and sent to the requesting end after the recognition result is obtained, so that the requesting end can look back at the communication record after completing the contact with the assisting end.
  • the third is the assistance terminal, which is an intelligent terminal used by professionals in the field of crop claims.
  • the assisting terminal When the assisting terminal establishes a connection with the requesting terminal for online video, it can assist the surveyor to determine the damage on-site and improve the efficiency of viewing the damage.
  • the server detects the survey assistance request instruction sent by the requesting end, it obtains the location information of the requesting end (The positioning information is latitude and longitude information).
  • the server when the server receives the survey assistance request instruction and positioning information from the requesting end, in order to recommend other salespersons or experts in the library who have a better understanding of this area, it can obtain the distance from the location information A collection of assisting ends composed of assisting ends within a preset distance threshold (for example, setting the distance threshold to 30KM). Since the server can quickly query the set of assisting terminals that meet the conditions, the efficiency of obtaining the basic data required before the requesting terminal and the assisting terminal quickly establish a video connection is improved.
  • a preset distance threshold for example, setting the distance threshold to 30KM
  • step S120 the method further includes:
  • the target label is used to form the target label set
  • the target tag set whose similarity with the requested tag exceeds the similarity threshold among the tags corresponding to each assisting end in the assisting end set can be calculated, and the assisting end corresponding to the target tag set is used as the updated assisting end set.
  • acquiring the similarity between the tag corresponding to each assisting end of the set of assisting ends and the requested tag includes:
  • the string edit distance between the two tags can be calculated as the similarity between the two tags.
  • the string edit distance is the minimum number of times required to edit a single character (such as modification, insertion, deletion) when changing from one string to another. For example, to modify the string "kitten" to the string “sitting" only three single-character editing operations, such as sitten (k ⁇ s), sittin (e ⁇ i), sittin (_ ⁇ g), so "kitten The editing distance between "" and “sitting” is 3.
  • the video information between the assisting end and the requesting end is obtained and saved; wherein the video information between the assisting end and the requesting end includes the assisting end correspondence
  • the requesting end video data includes crop video information and requester audio data;
  • the assisting end video data includes assisting end audio data.
  • a successful video connection instruction of an assisting end in the requesting end and the assisting end set is detected, it means that the requesting end has established a video connection with the selected cooperative end.
  • the server is used to obtain the requesting end's video information and Video information on the assisting side.
  • the video information of the assistant terminal is saved for subsequent processing of the video information through the background of the server, such as voice recognition.
  • the server When the server saves the video information between the assisting end and the requesting end, it generates a serial number with the user ID corresponding to the requesting end (such as the surveyor’s job number), the user ID corresponding to the assisting end, and the current system time, and uses the serial number Create a new folder for the file name in the storage area of the server, and save the video information between the assisting end and the requesting end in the new folder.
  • a serial number with the user ID corresponding to the requesting end (such as the surveyor’s job number), the user ID corresponding to the assisting end, and the current system time
  • the user ID corresponding to the requesting end, the user ID corresponding to the assisting end and the current system time generation serial number can be saved in the corresponding folder to save the video data corresponding to the requesting end, and the assisting end The corresponding video data.
  • the server will view the corresponding feedback data according to the requesting terminal's data view request, thus realizing the effective preservation of historical data.
  • S140 Perform audio extraction on the video information to obtain an audio extraction result, and obtain a text of the audio extraction result through a voice recognition model to obtain a recognition result.
  • the audio extraction result can be obtained by removing the video channel information in the video information.
  • the audio extraction result is recognized by the voice recognition model to obtain the recognition result.
  • the recognition result is extracted to ensure that after the requesting end and the assisting end interrupt communication, in order to facilitate the requesting end to obtain the detailed text information of the previous video communication process to review the survey plan, at this time, the video information of the assisting end can be audio extracted through the server , Obtain the audio extraction result, obtain the text of the audio extraction result through the speech recognition model, and obtain the recognition result.
  • step S140 includes:
  • the audio extraction result is recognized through the N-gram model to obtain the recognition result.
  • the whole sentence is obtained by recognition, for example, "The loss rate of XX farm Y field is more than 30%".
  • the N-gram model can effectively recognize the speech to be recognized, and obtain the sentence with the largest recognition probability as the recognition result.
  • step S140 it also includes:
  • the training set corpus is received, and the training set corpus is input to the initial N-gram model for training to obtain an N-gram model; wherein, the N-gram model is an N-gram model.
  • the training set corpus is a general corpus, and the vocabulary in the general corpus is not biased towards a specific field, but vocabulary in each field is involved.
  • the N-gram model for speech recognition can be obtained by inputting the training set corpus to the initial N-gram model for training.
  • the method further includes:
  • the server after the server saves the video information, in addition to extracting the text of the audio extraction result in the video information, it can also perform image recognition on the video information to determine the crops in the video information. Category, using the crop category as the attribute tag of the recognition result.
  • the step of acquiring the crop category existing in the video information through image recognition specifically includes:
  • Pearson similarity calculation is performed on the picture feature vector corresponding to each picture in the picture set and the feature vector of each picture in the pre-built picture library to obtain the Pearson similarity of the picture feature vector corresponding to each picture in the picture set
  • the feature vector whose degree is greater than the preset similarity threshold is used as the feature vector of the retrieval result;
  • the retrieval result picture corresponding to the retrieval result feature vector in the picture library and the crop category label corresponding to the retrieval result picture are acquired, and the crop category label corresponding to the retrieval result picture is taken as the crop category existing in the video information.
  • the starting time point such as the 15th second
  • the acquisition duration such as 15 seconds
  • the target video segment is acquired from the video information according to the start time point and the acquisition duration. For example, at this time, a 15-second-long video is acquired from the 15th second from the video information corresponding to the requesting end as the target video segment.
  • multiple frames of pictures in the target video segment are obtained through video splitting to form a target picture set.
  • obtaining the picture feature vector of each target picture first obtain the pixel matrix corresponding to each target picture, and then divide each The pixel matrix corresponding to the target image is used as the input of the input layer in the convolutional neural network model to obtain multiple feature maps, and then the feature maps are input to the pooling layer to obtain the one-dimensional vector corresponding to the maximum value corresponding to each feature map, and finally The one-dimensional vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer to obtain a picture feature vector corresponding to each target picture.
  • the feature templates stored in the image library store the feature vectors of a large number of people pictures that have been collected, with these mass feature templates as a data basis, they can be used to determine the crop category corresponding to the target image, thereby achieving image recognition .
  • the server when the server has completed the text extraction of the audio extraction result, it sends the recognition result to the requesting end, and the salesperson corresponding to the requesting end can obtain the requirements for claim settlement according to the survey plan in the recognition result. Obtain important parameters to realize on-site survey.
  • This method realizes the inviting professional online video to assist in the damage assessment in the process of damage assessment, and improves the accuracy and efficiency of the damage assessment result.
  • the embodiments of the present application also provide a video-based crop surveying device, which is used to execute any embodiment of the aforementioned video-based crop surveying method.
  • FIG. 4 is a schematic block diagram of a video-based crop survey device provided by an embodiment of the present application.
  • the video-based crop survey device 100 can be configured in a server.
  • the video-based crop surveying device 100 includes a positioning acquisition unit 110, an assistance terminal collection acquisition unit 120, a video acquisition unit 130, an audio recognition unit 140, and a result sending unit 150.
  • the location obtaining unit 110 is configured to obtain location information of the requesting end if a survey assistance request instruction issued by the requesting end is detected.
  • the insurer is not responsible for compensation if the planted rice suffers from natural disasters within the scope of insurance liability, but the loss rate is below 30%.
  • the actual loss rate is between 30% (inclusive) and 70% to be paid proportionally, and 70% (including 70%) or more are paid in full.
  • the compensation shall be calculated according to the ratio of the insured area to the actual planting area.
  • the surveyor of the insurance company receives a claim from the insured, he needs to go to the scene to investigate the damage. If the investigator is unable to accurately assess the risk situation by himself during the on-site investigation, he can invite other salesmen in the company or experts outside the company to assist in the investigation process such as risk assessment and other investigation operations through video conference.
  • the server detects the survey assistance request instruction sent by the requesting end, it obtains the location information of the requesting end (The positioning information is latitude and longitude information).
  • the assisting terminal set acquiring unit 120 is configured to acquire the assisting terminals whose distance from the located information is within a preset distance threshold to form the assisting terminal set.
  • the server when the server receives the survey assistance request instruction and positioning information from the requesting end, in order to recommend other salespersons or experts in the library who have a better understanding of this area, it can obtain the distance from the location information A collection of assisting ends composed of assisting ends within a preset distance threshold (for example, setting the distance threshold to 30KM). Since the server can quickly query the set of assisting terminals that meet the conditions, the efficiency of obtaining the basic data required before the requesting terminal and the assisting terminal can quickly establish a video connection is improved.
  • a preset distance threshold for example, setting the distance threshold to 30KM
  • the video-based crop survey device 100 further includes:
  • a similarity obtaining unit configured to obtain a request label corresponding to the survey assistance request, and obtain the similarity between the label corresponding to each assisting end of the set of assisting ends and the request label;
  • a target tag set obtaining unit configured to form a target tag set with the target tags if there are target tags whose similarity with the requested tag exceeds a preset similarity threshold among the tags corresponding to each assisting end of the assisting end set;
  • the set update unit is used to obtain the assisting terminal corresponding to the target tag set to obtain the updated assisting terminal set.
  • the target tag set whose similarity with the requested tag exceeds the similarity threshold among the tags corresponding to each assisting end in the assisting end set can be calculated, and the assisting end corresponding to the target tag set is used as the updated assisting end set.
  • the similarity acquisition unit is further configured to:
  • the string edit distance between the two tags can be calculated as the similarity between the two tags.
  • the string edit distance is the minimum number of times required to edit a single character (such as modification, insertion, deletion) when changing from one string to another. For example, to modify the string "kitten" to the string “sitting" only three single-character editing operations, such as sitten (k ⁇ s), sittin (e ⁇ i), sittin (_ ⁇ g), so "kitten The editing distance between "" and “sitting” is 3.
  • the video acquisition unit 130 is configured to obtain and save the video information between the assisting terminal and the requesting terminal if a successful video connection instruction is detected between the requesting terminal and the assisting terminal set; among them, the video between the assisting terminal and the requesting terminal
  • the information includes the assisting end video data corresponding to the assisting end and the requesting end video data corresponding to the requesting end; the requesting end video data includes crop video information and requester audio data; the assisting end video data includes the assisting end audio data .
  • a successful video connection instruction of an assisting end in the requesting end and the assisting end set is detected, it means that the requesting end has established a video connection with the selected cooperative end.
  • the server is used to obtain the requesting end's video information and Video information on the assisting side.
  • the video information of the assistant terminal is saved for subsequent processing of the video information through the background of the server, such as voice recognition.
  • the server When the server saves the video information between the assisting end and the requesting end, it generates a serial number with the user ID corresponding to the requesting end (such as the surveyor’s job number), the user ID corresponding to the assisting end, and the current system time, and uses the serial number Create a new folder for the file name in the storage area of the server, and save the video information between the assisting end and the requesting end in the new folder.
  • a serial number with the user ID corresponding to the requesting end (such as the surveyor’s job number), the user ID corresponding to the assisting end, and the current system time
  • the user ID corresponding to the requesting end, the user ID corresponding to the assisting end and the current system time generation serial number can be saved in the corresponding folder to save the video data corresponding to the requesting end, and the assisting end The corresponding video data.
  • the server will view the corresponding feedback data according to the requesting terminal's data view request, thus realizing the effective storage of historical data.
  • the audio recognition unit 140 is configured to perform audio extraction on the video information to obtain an audio extraction result, and obtain the text of the audio extraction result through a voice recognition model to obtain the recognition result.
  • the audio extraction result can be obtained by removing the video channel information in the video information.
  • the audio extraction result is recognized by the voice recognition model to obtain the recognition result.
  • the recognition result is extracted to ensure that after the requesting end and the assisting end interrupt communication, in order to facilitate the requesting end to obtain the detailed text information of the previous video communication process to review the survey plan, at this time, the video information of the assisting end can be audio extracted through the server , Obtain the audio extraction result, obtain the text of the audio extraction result through the speech recognition model, and obtain the recognition result.
  • the audio recognition unit 140 is further used to:
  • the audio extraction result is recognized through the N-gram model to obtain the recognition result.
  • the whole sentence is obtained by recognition, for example, "The loss rate of XX farm Y field is more than 30%".
  • the N-gram model can effectively recognize the speech to be recognized, and obtain the sentence with the largest recognition probability as the recognition result.
  • video-based crop survey device 100 further includes:
  • the model training unit is configured to receive a training set corpus, and input the training set corpus to the initial N-gram model for training to obtain an N-gram model; wherein the N-gram model is an N-gram model.
  • the training set corpus is a general corpus, and the vocabulary in the general corpus is not biased towards a specific field, but vocabulary in each field is involved.
  • the N-gram model for speech recognition can be obtained by inputting the training set corpus to the initial N-gram model for training.
  • the video-based crop survey device 100 further includes:
  • the crop category identifying unit 141 is configured to obtain the crop category existing in the video information through image recognition, and use the crop category as the attribute tag of the recognition result.
  • the server after the server saves the video information, in addition to extracting the text of the audio extraction result in the video information, it can also perform image recognition on the video information to determine the crops in the video information. Category, using the crop category as the attribute tag of the recognition result.
  • the crop category identification unit 141 includes:
  • a target video segment acquiring unit configured to acquire a target video segment in the video information according to a preset starting time point and acquisition duration
  • a video splitting unit configured to obtain multiple frames of pictures in the target video segment through video splitting to form a target picture set
  • the feature vector obtaining unit is configured to perform feature extraction on each picture in the target picture set through a convolutional neural network model to obtain a picture feature vector corresponding to each picture in the target picture set;
  • the retrieval result feature vector acquiring unit is used to calculate the Pearson similarity between the feature vector of each picture corresponding to each picture in the picture set and the feature vector of each picture in the pre-built picture library, and obtain each picture in the picture set.
  • the feature vector whose Pearson similarity of the corresponding image feature vector is greater than the preset similarity threshold is used as the feature vector of the retrieval result;
  • the crop category label obtaining unit is configured to obtain the retrieval result picture corresponding to the retrieval result feature vector in the picture library and the crop category label corresponding to the retrieval result picture, and use the crop category label corresponding to the retrieval result picture as the video information
  • the starting time point such as the 15th second
  • the acquisition duration such as 15 seconds
  • the target video segment is acquired from the video information according to the start time point and the acquisition duration. For example, at this time, a 15-second-long video is acquired from the 15th second from the video information corresponding to the requesting end as the target video segment.
  • multiple frames of pictures in the target video segment are obtained through video splitting to form a target picture set.
  • obtaining the picture feature vector of each target picture first obtain the pixel matrix corresponding to each target picture, and then divide each The pixel matrix corresponding to the target image is used as the input of the input layer in the convolutional neural network model to obtain multiple feature maps, and then the feature maps are input to the pooling layer to obtain the one-dimensional vector corresponding to the maximum value corresponding to each feature map, and finally The one-dimensional vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer to obtain a picture feature vector corresponding to each target picture.
  • the feature templates stored in the image library store the feature vectors of a large number of people pictures that have been collected, with these mass feature templates as a data basis, they can be used to determine the crop category corresponding to the target image, thereby achieving image recognition .
  • the result sending unit 150 is configured to send the recognition result to the requesting end.
  • the server when the server has completed the text extraction of the audio extraction result, it sends the recognition result to the requesting end, and the salesperson corresponding to the requesting end can obtain the requirements for claim settlement according to the survey plan in the recognition result. Obtain important parameters to realize on-site survey.
  • the device realizes the inviting professional online video to assist in the damage assessment during the damage investigation and assessment process, and improves the accuracy of the investigation and assessment results and the efficiency of the investigation and assessment.
  • the above-mentioned video-based crop surveying device can be implemented in the form of a computer program, and the computer program can be run on a computer device as shown in FIG. 6.
  • FIG. 6 is a schematic block diagram of a computer device according to an embodiment of the present application.
  • the computer device 500 is a server, and the server may be an independent server or a server cluster composed of multiple servers.
  • the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.
  • the non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032.
  • the processor 502 can execute a video-based crop survey method.
  • the processor 502 is used to provide calculation and control capabilities, and support the operation of the entire computer device 500.
  • the internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503.
  • the processor 502 can execute the video-based crop survey method.
  • the network interface 505 is used for network communication, such as providing data information transmission.
  • the structure shown in FIG. 6 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied.
  • the specific computer device 500 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.
  • the processor 502 is configured to run a computer program 5032 stored in a memory to implement the video-based crop survey method in the embodiment of the present application.
  • the embodiment of the computer device shown in FIG. 6 does not constitute a limitation on the specific configuration of the computer device.
  • the computer device may include more or less components than those shown in the figure. Or combine certain components, or different component arrangements.
  • the computer device may only include a memory and a processor. In such embodiments, the structures and functions of the memory and the processor are the same as those of the embodiment shown in FIG. 6, which will not be repeated here.
  • the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.
  • a computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • the computer-readable storage medium stores a computer program, where the computer program is executed by a processor to implement the video-based crop survey method in the embodiments of the present application.
  • the storage medium is a physical, non-transitory storage medium, such as a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk, or an optical disk that can store program codes. medium.
  • a physical, non-transitory storage medium such as a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk, or an optical disk that can store program codes. medium.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Multimedia (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Library & Information Science (AREA)
  • General Business, Economics & Management (AREA)
  • Technology Law (AREA)
  • Agronomy & Crop Science (AREA)
  • Animal Husbandry (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Mining & Mineral Resources (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Signal Processing (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Image Analysis (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention porte sur un procédé et sur un appareil d'enquête sur la récolte utilisant une vidéo, sur un dispositif d'ordinateur et sur un support d'informations. Le procédé consiste : lors de la détection d'une instruction de demande d'aide à l'enquête émise par une extrémité demandeuse, à acquérir des informations de positionnement de l'extrémité demandeuse ; à acquérir des extrémités d'assistance ayant une distance par rapport aux informations de positionnement à l'intérieur d'un seuil de distance prédéfini et à former un groupe d'extrémités d'assistance ; lors de la détection d'une instruction indiquant une connexion vidéo réussie entre l'extrémité demandeuse et une extrémité d'assistance du groupe d'extrémités d'assistance, à acquérir des informations vidéo transmises depuis l'extrémité demandeuse à l'extrémité d'assistance et à sauvegarder les informations vidéo ; à effectuer une extraction audio sur les informations vidéo de sorte à obtenir un résultat d'extraction audio et à acquérir un texte du résultat d'extraction audio au moyen d'un modèle de reconnaissance vocale de sorte à obtenir un résultat de reconnaissance ; et à envoyer le résultat de la reconnaissance à l'extrémité demandeuse.
PCT/CN2019/118245 2019-08-15 2019-11-14 Procédé et appareil d'enquête sur la récolte utilisant une vidéo, et dispositif d'ordinateur WO2021027156A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910752906.1 2019-08-15
CN201910752906.1A CN110688513B (zh) 2019-08-15 2019-08-15 基于视频的农作物查勘方法、装置及计算机设备

Publications (1)

Publication Number Publication Date
WO2021027156A1 true WO2021027156A1 (fr) 2021-02-18

Family

ID=69108260

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/118245 WO2021027156A1 (fr) 2019-08-15 2019-11-14 Procédé et appareil d'enquête sur la récolte utilisant une vidéo, et dispositif d'ordinateur

Country Status (2)

Country Link
CN (1) CN110688513B (fr)
WO (1) WO2021027156A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115311103A (zh) * 2022-09-29 2022-11-08 中国太平洋财产保险股份有限公司四川分公司 一种风险评估系统和方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111698446B (zh) * 2020-05-26 2021-09-21 上海智勘科技有限公司 在实时视频中同时进行文本信息传输的方法
CN113689430B (zh) * 2021-10-26 2022-02-15 紫东信息科技(苏州)有限公司 肠镜检查状态监测的图像处理方法与装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104866956A (zh) * 2015-05-12 2015-08-26 集时通（福建）信息科技有限公司 应用于养殖业和种植业的现场勘险信息获取辅助管理系统
US20160055592A1 (en) * 2014-08-25 2016-02-25 David P. Groeneveld System, Method and Product for Automated Crop Insurance Loss Adjusting for Prevented Planting Conditions
CN105898214A (zh) * 2016-03-24 2016-08-24 北京农业信息技术研究中心 现场查勘装置、现场查勘确认的方法及服务器
CN106303426A (zh) * 2016-08-17 2017-01-04 苏州华兴源创电子科技有限公司 一种车辆远程定损方法及系统
CN107169018A (zh) * 2017-04-06 2017-09-15 河南云保遥感科技有限公司 一种农业保险查勘、定损系统及其实现方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8688483B2 (en) * 2013-05-17 2014-04-01 Watts And Associates, Inc. Systems, computer-implemented methods, and computer medium to determine premiums and indemnities for supplemental crop insurance
KR20170084790A (ko) * 2016-01-13 2017-07-21 주식회사 테크인모션 음성 인식 기반의 건강 관리 앱을 실행하는 이동통신 단말기 및 그 동작 방법
CN107194323B (zh) * 2017-04-28 2020-07-03 阿里巴巴集团控股有限公司 车辆定损图像获取方法、装置、服务器和终端设备
CN107833141A (zh) * 2017-11-27 2018-03-23 无锡同春新能源科技有限公司 基于北斗导航的无人机测定山药田水灾的保险理赔系统
CN108053325A (zh) * 2017-11-30 2018-05-18 河南云保遥感科技有限公司 一种基于农作物遥感技术的农业保险定损方法
CN107909492A (zh) * 2017-11-30 2018-04-13 河南云保遥感科技有限公司 一种利用机器学习技术的农业保险查勘定损方法
CN108769624B (zh) * 2018-07-09 2020-09-22 北京精友世纪软件技术有限公司 一种智能车险移动视频查勘系统
CN109859055A (zh) * 2018-12-13 2019-06-07 中国平安财产保险股份有限公司 保费计算方法、装置、计算机设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160055592A1 (en) * 2014-08-25 2016-02-25 David P. Groeneveld System, Method and Product for Automated Crop Insurance Loss Adjusting for Prevented Planting Conditions
CN104866956A (zh) * 2015-05-12 2015-08-26 集时通（福建）信息科技有限公司 应用于养殖业和种植业的现场勘险信息获取辅助管理系统
CN105898214A (zh) * 2016-03-24 2016-08-24 北京农业信息技术研究中心 现场查勘装置、现场查勘确认的方法及服务器
CN106303426A (zh) * 2016-08-17 2017-01-04 苏州华兴源创电子科技有限公司 一种车辆远程定损方法及系统
CN107169018A (zh) * 2017-04-06 2017-09-15 河南云保遥感科技有限公司 一种农业保险查勘、定损系统及其实现方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHEN, JIANWANG: "Uses of New Technology in Plant Risk Assessment", CHINA INSURANCE, no. 01, 15 January 2018 (2018-01-15), pages 50 - 54, XP055780717 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115311103A (zh) * 2022-09-29 2022-11-08 中国太平洋财产保险股份有限公司四川分公司 一种风险评估系统和方法
CN115311103B (zh) * 2022-09-29 2022-12-30 中国太平洋财产保险股份有限公司四川分公司 一种风险评估系统和方法

Also Published As

Publication number Publication date
CN110688513A (zh) 2020-01-14
CN110688513B (zh) 2023-08-18

Similar Documents

Publication Publication Date Title
US10958748B2 (en) Resource push method and apparatus
CN110363210B (zh) 一种图像语义分割模型的训练方法和服务器
WO2021027156A1 (fr) Procédé et appareil d'enquête sur la récolte utilisant une vidéo, et dispositif d'ordinateur
WO2019119505A1 (fr) Procédé et dispositif de reconnaissance faciale, dispositif informatique et support d'enregistrement
CN106873799B (zh) 输入方法和装置
CN108287857B (zh) 表情图片推荐方法及装置
US11544307B2 (en) Personnel selecting device, personnel selecting system, personnel selecting method, and recording medium
CN110765882B (zh) 一种视频标签确定方法、装置、服务器及存储介质
CN108427941B (zh) 用于生成人脸检测模型的方法、人脸检测方法和装置
US20160314109A1 (en) Recommending form fragments
CN109872026A (zh) 评测结果生成方法、装置、设备及计算机可读存储介质
CN113362314B (zh) 医学图像识别方法、识别模型训练方法及装置
CN113255713A (zh) 用于跨对象变化的数字图像选择的机器学习
WO2023065989A1 (fr) Procédé et système de diagnostic de maladie de plante et d'insecte nuisible, et support de stockage lisible
CN115331150A (zh) 图像识别方法、装置、电子设备、存储介质
JP2022185143A (ja) テキスト検出方法、テキスト認識方法及び装置
US10296924B2 (en) Document performance indicators based on referral context
CN109993026A (zh) 亲属识别网络模型的训练方法及装置
CN116259078B (zh) 一种杀虫剂推荐方法、装置、设备及存储介质
CN115620886B (zh) 一种数据审核方法和装置
CN114373046B (zh) 辅助机器人运行的方法、装置及存储介质
CN112470216A (zh) 语音应用平台
CN111209407B (zh) 数据处理方法、装置、电子设备和计算机可读存储介质
WO2021036305A1 (fr) Procédé de traitement de données, appareil, dispositif et support de stockage
CN113869099A (zh) 一种图像处理方法、装置、电子设备和存储介质

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: 19941096

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19941096

Country of ref document: EP

Kind code of ref document: A1