Large-scale video algorithm data mining platform

Abstract

The invention discloses a large-scale video algorithm data mining platform, which comprises a video uploading module, the uploading module is bidirectionally controlled by means of a transceiving base station, and a video classification module for classifying and uploading videos is arranged at an induction end of the transceiving base station. The video classification module and the wireless transceiving base station are arranged in an induction manner, the video classification module is arranged in an induction manner through one end of the transceiving base station, the video classification module and the specified video storage module are arranged in a bidirectional manner, and the video classification module and the video storage module are arranged in a bidirectional manner. According to the invention, the data reckoning platform, the video uploading module and the transceiving base station are mutually combined to achieve access and reading of a part of data, and after uploading is completed, based on mutual cooperation between the video classification module and the video storage module, monitoring and analysis of a part of algorithm data are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of algorithm data arrangement, and particularly to a large-scale video algorithm data mining platform. Background Art

[0002] Algorithm data is a process that takes one or a group of values as input content and generates one or a group of values as output results. It contains a series of clear instructions for solving problems, and also includes the logical rules and methods behind each clear instruction;

[0003] An algorithm is a tool for implementing data structures, and a data structure is a carrier for implementing algorithms. Different algorithms have different execution efficiencies. That is to say, in the process of solving problems, not only the storage method of data will affect the efficiency, but also the quality of the algorithm will affect the efficiency.

[0004] After retrieval, it is found that a video algorithm development platform and its development method with the application number: CN200810119589.1, the content of which is: The present invention discloses a video algorithm development platform and its development method. The platform includes: a video source module for reading data from a data source and generating video stream data according to the data; a video processing module including one or more common image processing interfaces for performing encoding / decoding and compression / decompression processing on the video stream data through the common image processing interfaces; a video display module for displaying the video stream data after the encoding / decoding and compression / decompression processing. The development platform and development method of the present invention can realize the dynamic debugging of video algorithms, so that the occupancy rate of resources such as the CPU by video algorithms in different situations can be tested by setting different video playback frame rates, and the final effect of the video algorithm can be observed in real time.

[0005] In the state where traditional large-scale algorithm data platforms achieve data access and analysis, it is easy to make it difficult to analyze large-scale analysis and detection of data, and it is impossible to replace manual work to further play the role of data calculation, resulting in difficulties in coordinating and analyzing data in the later stage of data analysis and mining. In addition, in the traditional data analysis stage, it is relatively difficult to further mine and analyze data access, resulting in inconvenience in realizing algorithm mining and detection of data during later analysis, and thus making it difficult in the later stage of data analysis and mining. Summary of the Invention

[0006] An object of the present invention is to provide a large-scale video algorithm data mining platform to solve the problems raised in the above background art.

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: a large-scale video algorithm data mining platform, the data inference platform includes a video uploading module, wherein the uploading module is bidirectionally controlled by means of a transceiver base station, wherein the sensing end of the transceiver base station is provided with a video classification module for video classification uploading, and the video classification module is set with a wireless transceiver base station sensing device;

[0008] The video classification module is set by sensing at one end of the transceiver base station, wherein the video classification module is set bidirectionally with the designated video storage module, and the video classification module is set bidirectionally with the video storage module;

[0009] The video classification module and the video storage module are bidirectionally arranged, and further include a video data review module, an exposure module, a push module, an algorithm service module and a collection module;

[0010] The push module and the video storage module are bidirectionally set, the video exposure module and the algorithm service module are linked bidirectionally set, and the algorithm service module and the collection module are inductively set;

[0011] wherein one end of the video exposure module is bidirectionally configured via a designated operation module;

[0012] The algorithm-based service module also includes a programming module, a disk program cache module, a data writing module and an analysis module for video algorithm analysis.

[0013] As a preferred solution of the present invention: the analysis end located in the analysis module is set to sense the management module specified for storage management, and is set bidirectionally between the data writing module and the disk program cache module, wherein the disk program cache module is a redundant disk, and the disk program cache module and the analysis module are set to sense bidirectionally, and the analysis data of the analysis module is sorted by the management module.

[0014] As a preferred solution of the present invention: the management module also includes an access analysis for data, wherein the collection module has an extended presentation visualization mining module, and the visualization mining module includes a playback data display module, a video data compression playback module, a traffic timing module and a data mining module;

[0015] The playback data display module and the collection module are bidirectionally set, and are located in a bidirectional sensing setting between the flow timing module and the video data compression playback module, and the flow timing module and the collection module are bidirectionally set on the surface of the flow timing module, and a sensing setting is located at one end of the data mining module.

[0016] As a preferred embodiment of the present invention: One end of the data mining module is bidirectionally arranged with the management module, and the management module further includes a general control management module, which covers: a data presentation module, an editing module, a correction module, and a data editing module;

[0017] The data presentation module is bidirectionally arranged with the editing module, the correction module is bidirectionally arranged with the editing module, and the data editing module is bidirectionally arranged with the correction module;

[0018] The data presentation module further includes an imaging module, a data algorithm input module, an external speaker module, and a logic training module.

[0019] As a preferred embodiment of the present invention: The imaging module is bidirectionally arranged with the data presentation module, and the data algorithm input module is inductively arranged with the imaging module;

[0020] Among them, the push data of the external speaker module is inductively arranged with the data algorithm input module. The logic training module covers a servo server module, a servo graphics card module, and an integrated operation module based on hardware. The integrated operation module is bidirectionally arranged with the servo graphics card module and is bidirectionally controlled by the servo server module during this period.

[0021] As a preferred embodiment of the present invention: The algorithm data mining mechanism includes an acquisition module, a data snapshot module, a cover module, and an access module that are bidirectionally connected to the operation module;

[0022] Among them, the acquisition module is bidirectionally arranged with the data snapshot module, the cover module is bidirectionally arranged with the acquisition module, and the access module is bidirectionally arranged with the cover module and the data snapshot module.

[0023] As a preferred embodiment of the present invention: The data snapshot module is inductively arranged with the operation module. The playback data display module is also bidirectionally provided with a data mining area, which contains a data writing end, a data connection end, and a data redundancy module for sensing data feedback;

[0024] And the data redundancy module is bidirectionally arranged with the data writing end,

[0025] The data connection end is connected in series through a data broadband module.

[0026] As a preferred embodiment of the present invention: The data broadband module further includes a data transmitting end for corresponding data detection, a bandwidth transmission module, a transmission information module, and a crawler scanning module;

[0027] Among them, the data transmitter is inductively set with the bandwidth transmission module, and the transmission information module is set bidirectionally with the bandwidth transmission module. One end of the crawler scanning module is inductively set with the specified data transmitter, and one end located in the bandwidth transmission module and the transmission information module is numerically controlled by an auxiliary bandwidth flow chip.

[0028] As a preferred solution of the present invention: Among them, the crawler scanning modules are inductively set through corresponding bandwidth flow chips. A maintenance unit is also provided below the crawler scanning module. The maintenance unit also includes a temperature control module, a temperature control mechanism, a data diagnosis module, and a detection module for unit data maintenance. Among them, the temperature control module and the data diagnosis module are in bidirectional control, and the data diagnosis module and the temperature control mechanism are in bidirectional induction setting. The crawler scanning module also includes a crawler speed control unit: used to control the crawler speed within a preset crawler speed range.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] 1) By using the data calculation platform, video upload module, and transceiver base station in combination, it is used to achieve the access and reading of a part of the data. After the upload is completed, based on the mutual cooperation between the video classification module and the video storage module, for the monitoring and analysis of a part of the algorithm data, the data detection and analysis of the push module and the algorithm service module are adopted. In addition, according to the data access of the corresponding collection module, it promotes the feedback display of a part of the data;

[0031] 2) By using the operation module and the programming module, during the large-scale video calculation in the later stage, the specified data can be better analyzed and summarized, and the connection of the data of the total control management module can be better achieved. The imaging module further achieves the connection and analysis of the data after the specified data is accessed by the phenomenon module;

[0032] 3) The combination of the programming module and the corresponding servo server module and servo graphics card module is used to achieve the optimization of a part of the programming data in terms of hardware, ensuring that during the stage of obtaining and analyzing specific data in the later stage, the data snapshot module and the cover module can be obtained to better assist the programming algorithm in calculating and classifying specific data information. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0034] Figure 1 It is a schematic diagram of the modules of the present invention;

[0035] Figure 2 This is a partial schematic diagram of the visualization mining module of the present invention;

[0036] Figure 3 This is a schematic diagram of the phenomenon module of the present invention;

[0037] Figure 4 This is a schematic diagram of the structure of the algorithm data mining mechanism of the present invention.

[0038] In the figure: 1. Data calculation platform; 11. Video upload module; 12. Transceiving base station; 13. Video classification module; 14. Video storage module; 15. Video data review module; 16. Exposure module; 17. Push module; 18. Algorithm service module; 19. Aggregation module; 191. Operation module; 192. Programming module; 193. Disk program cache module; 194. Data writing module; 195. Analysis module; 196. Management module; 2. Visualization mining module; 21. Play data display module; 22. Video data compression and play module; 23. Traffic timing module; 24. Data mining module; 3. Total control management module; 31. Data presentation module; 32. Editing module; 33. Correction module; 34. Data editing module; 4. Imaging module; 41. Data algorithm entry module; 42. External speaker module; 43. Logic training module; 44. Servo server module; 45. Servo graphics card module; 46. Integrated operation module; 5. Algorithm data mining mechanism; 51. Acquisition module; 52. Data snapshot module; 53. Cover module; 54. Access module; 6. Data mining area; 61. Data writing end; 62. Data connection end; 63. Data redundancy module; 7. Data broadband module; 71. Data transmitting end; 72. Bandwidth transmission module; 73. Transmission information module; 74. Crawler scanning module; 8. Maintenance unit; 81. Temperature control module; 82. Temperature control mechanism; 83. Data diagnosis module; 84. Detection module. Detailed implementation manners

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] Please refer to Figures 1-4, the present invention provides a technical solution: a large-scale video algorithm data mining platform. This data calculation platform 1 includes a video upload module 11. The upload module is bidirectionally controlled by means of a transceiver base station 12. The sensing end of the transceiver base station 12 is provided with a video classification module 13 for classifying and uploading videos. The video classification module 13 is inductively arranged with the wireless transceiver base station 12. One end of the video classification module 13 is inductively arranged through the transceiver base station 12. The video classification module 13 is bidirectionally arranged with a designated video storage module 14. The video classification module 13 and the video storage module 14 are bidirectionally arranged. There is also a video data review module 15, an exposure module 16, a push module 17, an algorithm service module 18, and a collection module 19. The push module 17 and the video storage module 14 are bidirectionally arranged. The video exposure module 16 and the algorithm service module 18 are linked and bidirectionally arranged. The algorithm service module 18 and the collection module 19 are inductively arranged. One end of the video exposure module 16 is bidirectionally arranged through a designated operation module 191. Based on the algorithm service module 18, there is also a programming module 192, a disk program cache module 193, a data writing module 194, and an analysis module 195 for video algorithm analysis.

[0041] In this embodiment: The analysis end of the analysis module 195 is inductively arranged with a designated management module 196 for storage management. The data writing module 194 and the disk program cache module 193 are bidirectionally arranged. The disk program cache module 193 is a redundant disk. The disk program cache module 193 and the analysis module 195 are inductively bidirectionally arranged. The analysis data of the analysis module 195 and the data of the management module 196 are sorted out.

[0042] Through the cooperation of the analysis module 195 and the specific management module 196, it is used for the inductive analysis of part of the video data. The adopted part of the disk program cache module 193 achieves the sorting, recommendation, and mining of part of the data.

[0043] In this embodiment: The management module 196 also includes data access. The collection module 19 has an extended visual mining module 2. The visual mining module 2 includes a playback data display module 21, a video data compression playback module 22, a traffic timing module 23, and a data mining module 24. The playback data display module 21 and the collection module 19 are bidirectionally arranged. The traffic timing module 23 and the video data compression playback module 22 are bidirectionally inductive. The traffic timing module 23 and the collection module 19 are bidirectionally arranged on the surface of the traffic timing module 23. One end of the data mining module 24 is inductively arranged and is in mutual cooperation with the data mining module 24.

[0044] Through the set collection module 19 and the visualization mining module 2, personnel can relatively easily play and transmit the video data implanted inside the server, and complete the recommended display of the visual data of the video.

[0045] In this embodiment: One end of the data mining module 24 is bidirectionally set with the management module 196. The management module 196 further includes a general control management module 3, and the general control management module 3 includes: a data presentation module 31, an editing module 32, a correction module 33, and a data editing module 34; The data presentation module 31 is bidirectionally set with the editing module 32, the correction module 33 is bidirectionally set with the editing module 32, and the data editing module 34 is bidirectionally set with the correction module 33; The data presentation module 31 further includes an imaging module 4, a data algorithm input module 41, an external playback module 42, and a logic training module 43.

[0046] The mutual cooperation of the adopted note module 32 and correction module 33 further achieves the analysis and detection of the corresponding programming data. The adopted imaging module 4 and data algorithm input module 41 display the feedback in the data analysis writing stage.

[0047] In this embodiment: The imaging module 4 is bidirectionally set with the data presentation module 31, and the data algorithm input module 41 is inductively set with the imaging module 4; The pushed data of the external playback module 42 is inductively set with the data algorithm input module 41. The logic training module 43 includes a servo server module 44 based on hardware, a servo graphics card module 45, and an integrated operation module 46. The integrated operation module 46 is bidirectionally set with the servo graphics card module 45 and is bidirectionally controlled by the servo server module 44 during this period.

[0048] After the access of the adopted servo graphics card module 45 and integrated operation module 46, multiple independent graphics cards will be unified into a complete system, which is faster in the video editing and transcoding stage and can share a large amount of video operation pressure while ensuring the stable performance of the CPU.

[0049] In this embodiment: The algorithm data mining mechanism 5 includes an acquisition module 51, a data snapshot module 52, a cover module 53, and a pickup module 54 that are bidirectionally connected to the operation module 191; The acquisition module 51 is bidirectionally set with the data snapshot module 52, the cover module 53 is bidirectionally set with the acquisition module 51, and the pickup module 54 is bidirectionally set with the cover module 53 and the data snapshot module 52.

[0050] Through the mutual cooperation of the adopted acquisition module 51 and data snapshot module 52, the writing and analysis of part of the data are further achieved. The cover module 53 can read the cover of the video in real time after the video is edited and uploaded, and then classify such videos through scanning information.

[0051] In this embodiment: The data snapshot module 52 is inductively set with the operation module 191. The data display module 21 for playing data is also bidirectionally provided with a data mining area 6. The data mining area 6 includes a data writing end 61, a data connection end 62, and a data redundancy module 63 for sensing data feedback; and the data redundancy module 63 is bidirectionally set with the data writing end 61, and the data connection end 62 is connected in series through a data broadband module 7.

[0052] By combining the data writing end 61 and the corresponding data redundancy module 63, a part of data analysis and detection is achieved. Under the state of unified feedback writing of the data broadband module 7 adopted, the specified data writing analysis is maintained.

[0053] In this embodiment: The data broadband module 7 further includes a data transmitting end 71 for corresponding data detection, a bandwidth transmission module 72, a transmission information module 73, and a crawler scanning module 74; wherein the data transmitting end 71 is inductively set with the bandwidth transmission module 72, and the transmission information module 73 is bidirectionally set with the bandwidth transmission module 72. One end of the crawler scanning module 74 is inductively set with the specified data transmitting end 71, and one end located at the bandwidth transmission module 72 and the transmission information module 73 is numerically controlled by an auxiliary bandwidth traffic chip.

[0054] By using the data transmitting end 71 and the bandwidth transmission module 72, a part of real-time data analysis and detection is achieved. The access of the data to the crawler scanning module 74 can summarize the types of videos into separate samples according to the user's preferences, so as to optimize an independent video recommendation mechanism suitable for individuals as much as possible.

[0055] In this embodiment: Among them, the crawler scanning module 74 is inductively set through corresponding bandwidth traffic chips. A maintenance unit 8 is further provided below the crawler scanning module 74. The maintenance unit 8 further includes a temperature control module 81, a temperature control mechanism 82, a data diagnosis module 83, and a detection module 84 for unit data maintenance. Among them, the temperature control module 81 and the data diagnosis module 83 are bidirectionally controlled, and the data diagnosis module 83 and the temperature control mechanism 82 are bidirectionally inductively set.

[0056] The data diagnosis module 83 and the temperature control mechanism 82 cooperate with each other to achieve a part of data writing and analysis, and further construct a suitable data analysis architecture. The crawler scanning module further includes a crawler speed control unit: used to control the crawler speed within a preset crawler speed range, and the preset crawler speed range is, for example, 10 to 60 crawl requests per minute.

[0057] When specifically used, the first step: Access to a large amount of video data information:

[0058] First, according to the recommendation of the program algorithm: After the personnel batch upload the videos based on the data of each single account and obtain the data, after accessing the information of the video upload module 11 and while the information connection status of the peer base station and the transceiver base station 12 is on, directly implement the specified upload based on the network topology of the intelligent device and upload it to the server. Based on the data connection of the video data review module 15 in the video storage module 14 of the server, under the influence of the access algorithm of the video exposure module 16, a set of evaluation systems is constructed. This evaluation system is based on the data experienced by the video. After specifically completing the data on the data side, the data of a single video will be further statistically analyzed for the feedback reading situation according to one end of the specific video exposure module 16;

[0059] In addition, the reading duration is further achieved for the feedback of the granulometer data through the detection of the personnel collection module 19. Under the feedback display status of the specified play data display module 21 and the specific video data compression play module 22, the data feedback on the traffic timing module 23 is further realized. Relying on the corresponding data mining module 24, the detection of the data can be better achieved. For the mining reaction of the video data algorithm, at this time, the user accesses the data of the lower operation module 191 and the programming module 192 and then detects one end of the disk program cache module 193, calculates the data feedback characteristics of the video, and then performs automated analysis and classification. In addition, the user compiles the video induction management program according to the lower algorithm service module 18 and the lower data writing module 194 to achieve programming storage. Under the analysis and management status of the management module 196 and the analysis module 195, combined with the mutual operation and maintenance between the specified data presentation module 31 and the editing module 32, the feedback display status of the specified traffic timing module 23 can be manually realized at the position of the background terminal. Then, according to the data mining module 24, the algorithm-analyzed data videos are classified. When the specific data feedback detection is completed, the information access of the specific algorithm service is achieved. At this time, the user uses the acquisition module 51 and the data snapshot module 52 for information writing and analysis. At this time, the user cooperates between the cover module 53 and the access module 54 to obtain the cover in real time after the video data is accessed. After the cover is obtained, the data feedback display is realized at the position of the specific data connection end 62;

[0060] The second step; Hardware support for the algorithm program:

[0061] Since the computer performance consumed by the algorithm program is relatively large, it is necessary to coordinate the feedback of specific data between the servo graphics card module 45 and multiple integrated operation modules 46 in the management module 196. At this time, the user then follows the state displayed by the feedback of the logic training module 43 and the data editing module 34; furthermore, the feedback display of data is achieved. At this time, the user then further achieves the analysis and correction of data according to the state displayed by the feedback of the corresponding upload module 73 and the bandwidth transmission module 72. At this time, the user can change the data algorithm rule: the way to change is to further achieve the data presentation reaction after writing the data between the data algorithm entry module 41 and the imaging module 4. At this time, the user then further realizes the display after writing the data according to the reaction state of the external module 42 and the corresponding data editing module 34. At this time, the user then further realizes the automatic sorting of data classification according to the feedback display state of the editing module 32 and the feedback display state of the logic training module 43. The data algorithm entry module 41 is currently designed to solve the writing port for replacing the algorithm editing program;

[0062] Step 3: Mining and analysis of data algorithms:

[0063] At this time, the user needs to further obtain the corresponding data feedback according to the corresponding data monitoring. At this time, the user then coordinates the temperature control module 81 and the temperature control mechanism 82 to be used for the real-time diagnosis of data. In addition, the cooperation between the detection module 84 and the specified data feedback stage achieves the data monitoring.

[0064] The content not described in detail in the description belongs to the prior art well-known to those skilled in the art. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A large-scale video algorithm data mining platform, characterized in that: The data inference platform (1) comprises a video uploading module (11), wherein the uploading module is bidirectionally controlled by means of a transceiver base station (12), wherein a sensing end of the transceiver base station (12) is provided with a video classification module (13) for video classification uploading, and the video classification module (13) is arranged in a sensing manner with the wireless transceiver base station (12); The video classification module (13) is set by sensing at one end of the transceiver base station (12), the video classification module (13) and the designated video storage module (14) are set bidirectionally, and the video classification module (13) and the video storage module (14) are set bidirectionally; The video classification module (13) and the video storage module (14) are bidirectionally arranged, and further include a video data review module (15), an exposure module (16), a push module (17), an algorithm service module (18) and a collection module (19); The push module (17) and the video storage module (14) are bidirectionally arranged, the video exposure module (16) and the algorithm service module (18) are bidirectionally arranged in linkage, and the algorithm service module (18) and the collection module (19) are inductively arranged; One end of the video exposure module (16) is bidirectionally configured via a designated operation module (191); The algorithm service module (18) also includes a programming module (192), a disk program cache module (193), a data writing module (194) and an analysis module (195) for video algorithm analysis.

2. A large-scale video algorithm data mining platform according to claim 1, characterized in that: The analysis end of the analysis module (195) is configured to sense a designated management module (196) for storage management, and the data writing module (194) and the disk program cache module (193) are configured to sense each other in a bidirectional manner, wherein the disk program cache module (193) is a redundant disk, and the disk program cache module (193) and the analysis module (195) are configured to sense each other in a bidirectional manner, and the analysis data of the analysis module (195) is configured to be sorted out by the management module (196).

3. A large-scale video algorithm data mining platform according to claim 2, characterized in that: The management module (196) further includes an access analysis module for data, wherein the collection module (19) has an extended presentation visualization mining module (2), and the visualization mining module (2) includes a playback data display module (21), a video data compression playback module (22), a flow timing module (23) and a data mining module (24); The playback data display module (21) and the collection module (19) are bidirectionally arranged, and are located at the flow timing module (23) and the video data compression playback module (22) for bidirectional sensing, and the flow timing module (23) and the collection module (19) are bidirectionally arranged on the surface of the flow timing module (23), and are located at one end of the data mining module (24) for sensing.

4. A large-scale video algorithm data mining platform according to claim 3, characterized in that: One end of the data mining module (24) is bidirectionally arranged with the management module (196), and the management module (196) further includes a general control management module (3), and the general control management module (3) includes: a data presentation module (31), an editing module (32), a correction module (33) and a data editing module (34); The data presentation module (31) and the editing module (32) are bidirectionally arranged, the correction module (33) and the editing module (32) are bidirectionally arranged, and the data editing module (34) and the correction module (33) are bidirectionally arranged; The data presentation module (31) further comprises an imaging module (4), a data algorithm input module (41), an external playback module (42), and a logic training module (43).

5. A large-scale video algorithm data mining platform according to claims 1-4, characterized in that: The imaging module (4) and the data presentation module (31) are bidirectionally arranged, and the data algorithm input module (41) and the imaging module (4) are inductively arranged; The data pushed by the external speaker module (42) is sensed by the data algorithm input module (41), and the logic training module (43) includes a hardware-based servo server module (44), a servo graphics card module (45) and an integrated computing module (46), wherein the integrated computing module (46) and the servo graphics card module (45) are bidirectionally configured and bidirectionally controlled by the servo server module (44).

6. A large-scale video algorithm data mining platform according to claim 1, characterized in that: The algorithm data mining mechanism (5) comprises an acquisition module (51) bidirectionally connected to the operation module (191), a data snapshot module (52), a cover module (53) and an access module (54); The acquisition module (51) and the data snapshot module (52) are bidirectionally arranged, the cover module (53) and the acquisition module (51) are bidirectionally arranged, and the access module (54) and the cover module (53) and the data snapshot module (52) are bidirectionally arranged.

7. A large-scale video algorithm data mining platform according to claim 6, characterized in that: The data snapshot module (52) and the operation module (191) are inductively arranged, and the playback data display module (21) is also bidirectionally arranged with a data mining area (6), the data mining area (6) comprising a data writing end (61), a data connection end (62) and a data redundancy module (63) for sensing data feedback; Furthermore, the data redundancy module (63) and the data writing end (61) are bidirectionally arranged, and the data connection end (62) is serially connected via the data broadband module (7).

8. A large-scale video algorithm data mining platform according to claim 7, characterized in that: The data broadband module (7) further comprises a data transmitting end (71) for corresponding data detection, a bandwidth transmission module (72), a transmission information module (73) and a crawler scanning module (74); The data transmitting end (71) is inductively arranged with the bandwidth transmission module (72), and the transmission information module (73) is bidirectionally arranged with the bandwidth transmission module (72), wherein one end of the crawler scanning module (74) is inductively arranged with the designated data transmitting end (71), and an auxiliary bandwidth flow chip is numerically controlled between one end of the bandwidth transmission module (72) and the transmission information module (73).

9. A large-scale video algorithm data mining platform according to claim 8, characterized in that: The crawler scanning module (74) is provided with a sensing arrangement between corresponding bandwidth flow chips, and a maintenance unit (8) is provided below the crawler scanning module (74). The maintenance unit (8) also includes a temperature control module (81), a temperature control mechanism (82), a data diagnosis module (83) and a detection module (84) for unit data maintenance, wherein the temperature control module (81) and the data diagnosis module (83) are bidirectionally controlled, and the data diagnosis module (83) and the temperature control mechanism (82) are provided with a bidirectional sensing arrangement.

10. A large-scale video algorithm data mining platform according to claim 8, characterized in that The crawler scanning module also includes a crawler speed control unit: used to control the crawler speed to be within a preset crawler speed range.

Images