CN111209057A - Method and device for quickly adjusting sensing terminal - Google Patents
Method and device for quickly adjusting sensing terminal Download PDFInfo
- Publication number
- CN111209057A CN111209057A CN202010026725.3A CN202010026725A CN111209057A CN 111209057 A CN111209057 A CN 111209057A CN 202010026725 A CN202010026725 A CN 202010026725A CN 111209057 A CN111209057 A CN 111209057A
- Authority
- CN
- China
- Prior art keywords
- background server
- sensing terminal
- performance index
- detection data
- parameter configuration
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44505—Configuring for program initiating, e.g. using registry, configuration files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3692—Test management for test results analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method for quickly adjusting a sensing terminal, wherein the sensing terminal uploads detection data to a background server by the most loose parameter configuration; the background server summarizes all the service performances into a performance index; the background server tests the detection data and generates a test record; the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record; obtaining the parameter group when the performance index is maximum according to the functional relation; synchronizing the parameter set to the perceiving terminal; the invention also provides a device for quickly adjusting the sensing terminal, which can determine parameters only by running once and save time.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for quickly adjusting a sensing terminal.
Background
At present, the intelligent sensing service mostly adopts a form of a sensing terminal and a background server, as shown in fig. 1, and the sensing terminal is, for example, a face snapshot camera/a face recognition camera/a structured camera, etc. The detection of information such as images and sounds is realized by adopting a machine learning technology, and the detection result (such as a face image/identification information) is transmitted to a background server. The result detected by the perception terminal is controlled by a plurality of groups of parameters. For example, the face snapshot camera is controlled by a face credibility threshold, a face quality score threshold, a minimum detection face threshold, a face angle threshold and the like. When the minimum detected face threshold value is set to be lower, the detected face is smaller, otherwise, the detected face is detected when the face is large, and the small face is ignored.
The background server may be located at the cloud or a separately installed server. And the background server receives the detection data from the plurality of sensing terminals and processes the detection data to form service and service. If the background server receives the face information from the sensing terminal, the face is identified, and the track information of the person can be formed by combining the location information of the sensing terminal. Meanwhile, the background server manages and controls all the sensing terminals, namely, parameters of the sensing terminals can be remotely set through the background, and then the detection results of the sensing terminals are changed. The different uploading results of the sensing terminal can lead the business/service of the background server to generate different expressions. For example, if the face detection of the sensing terminal is not strict, fuzzy faces, side faces and small faces exist in the faces uploaded by the sensing terminal, and the algorithm of the background server is prone to false recognition or non-recognition when processing such faces, so that the subsequent result is inaccurate. When the face detection of the sensing terminal is strict, the number of faces uploaded by the sensing terminal is small, and the background server does not have enough data to generate a result. Therefore, the performance of the background server is best only if the sensing terminal is allowed to set appropriate parameters.
To make the background server perform best, there are generally two ways to implement: firstly, the method comprises the following steps: providing a plurality of groups of general parameter configurations, and setting a group of general parameter configurations aiming at the scene according to the characteristics of the service scene; II, secondly: and manually adjusting parameters, namely adjusting the parameters on the basis of default parameter configuration until the platform performance cannot be improved.
The prior art has the following disadvantages:
the method for providing multiple groups of general parameter configurations has the advantage that the general parameter configurations cannot configure the sensing terminal into the optimal parameter configuration in the scene because the scenes of all the sensing terminals are different.
The manual parameter adjustment method takes a long time because only one offset can be adjusted at a time, and when the configuration parameters are multiple, finding the optimal configuration becomes a theoretically feasible method, but cannot be implemented.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method and an apparatus for quickly adjusting a sensing terminal, which only needs to operate once to determine parameters, thereby saving time.
In a first aspect, the present invention provides a method comprising:
step 2, the background server summarizes all the business performances into a performance index;
step 3, testing the detection data and generating a test record by the background server;
step 4, the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record;
step 5, obtaining the parameter group when the performance index is maximum according to the functional relation;
and 6, synchronizing the parameter group to the perception terminal.
Further, the step 1 is further specifically: the sensing terminal uploads detection data to the background server in the most loose parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results.
Further, the step 4 is further specifically: and the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal in a machine learning mode according to the test record.
In a second aspect, the present invention provides an apparatus comprising:
the sensing terminal uploads the detection data to the background server according to the most relaxed parameter configuration;
the index module is used for summarizing all the service performances into one performance index by the background server;
the generating module is used for testing the detection data and generating a test record by the background server;
the function module is used for determining the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record by the background server;
the calculation module is used for solving the parameter group when the performance index is maximum according to the functional relation;
and the synchronization module synchronizes the parameter group to the perception terminal.
Further, the uploading module further specifically includes: the sensing terminal uploads detection data to the background server in the most loose parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results.
Further, the function module is further specifically: and the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal in a machine learning mode according to the test record.
One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:
compared with the sensing terminal configuration general parameters, the method and the device provided by the embodiment of the application can easily traverse all value ranges of all parameters and determine the optimal sensing terminal configuration parameters in all possible combinations; compared with the mode of manually adjusting parameters, observing and measuring after running for a period of time and then further adjusting and measuring, the method only needs to run once to determine the parameters, thereby saving time;
when a plurality of parameters are to be configured, the point with the maximum performance is difficult to find by changing a single parameter, and the method can realize the purpose;
the parameter adjusting process of the invention is automatically completed without depending on manual data processing, and when a large number of sensing terminals of different types are accessed to the same background server (namely, Param group parameters are different), or when the same type of sensing terminals are accessed to the same background server in different scenes, the workload of configuring the sensing terminals can be obviously reduced.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
The invention will be further described with reference to the following examples with reference to the accompanying drawings.
FIG. 1 is a schematic diagram of a prior art frame of the present invention;
FIG. 2 is a schematic diagram of data conversion according to the present invention;
FIG. 3 is a flow chart of a method according to one embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an apparatus according to a second embodiment of the present invention;
Detailed Description
The technical scheme in the embodiment of the application has the following general idea:
a sensing terminal uploads the detection data to a background server with the most loose parameter configuration;
the background server collects all the service performances into a performance index;
the third background server uses the most loose detection data test data and generates a test record;
the four background servers determine the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record;
fifthly, parameter groups when the performance indexes are maximum are obtained according to the functional relation;
and sixthly, synchronizing the parameter group to the perception terminal.
The scheme is described in detail as follows:
firstly, the sensing terminal uploads the detection data to a background server with the most loose parameter configuration. Therefore, a plurality of groups of detection data can be generated on the posterior server, and each group of detection data comprises two parts:
the first part is a detection result, if the sensing terminal is a face snapshot camera, the detection result is a face picture, and if the sensing terminal is a vehicle detection camera, the detection result is a vehicle picture; for convenient presentation, we use Data to represent the test results
The second part is an algorithm output result corresponding to the detection result, for example, the algorithm outputs the size of a face pixel, the face angle, the face reliability and the like when a face picture is detected. For convenience of description, we use Param ═ { s, r, q } to represent a set of three-parameter algorithm output results. The actual device parameters may be more than 3
In summary, with the most relaxed parameter configuration, several sets of detection Data can be obtained and expressed as DevResult _0 ═ { Data0, Param0 }; param0 ═ { s0, r0, q0}
DevResult_1＝{Data1,Param1}；Param0＝{s1,r1,q1}
……
DevResult_n＝{Datan,Paramn}；Param0＝{sn,rn,qn}
Secondly, the background server summarizes all the service performances into a performance index.
After the background processes the detection result, one or more performance indexes are generated. If the background server identifies the face uploaded by the sensing terminal, the higher the identification accuracy is required to be, the better the identification accuracy is, the identification accuracy is one of the performance indexes, for convenient expression, the identification accuracy is represented by C, the clearer the face is required to be, the picture definition is used as one of the performance indexes, and for convenient expression, the definition is represented by V. We can thus specify an overall performance metric, form (C, V). And (C, V) ═ a × C + (1-a) × V, wherein a is a decimal number between 0 and 1.
Thirdly, the background server uses the most loose result test data and generates test records
The background generates a set of performance results C, V each time it tests a set of detection data DevResult. When tested using n sets of DevResults, n sets of performance results { C0, V0} … { Cn, Vn } are formed. Using performance indicator function performance (C, V), n sets of performance indicators performance _0 … performance _ n may be generated. N sets of test records are then generated, each set of test records may be denoted as TestResult ═ { Param, form }, i.e., TestResult ═ { s, r, q, form }, and the data transformation process is described in fig. 2:
fourthly, the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record
The functional relationship f (s, r, q) between the parameters and the performance index can be calculated from the generated test record, and is generally calculated by machine learning. Such as decision trees, neural networks.
Fifthly, parameter group when performance index is maximum is obtained according to function relation
And calculating values of s _ best, r _ best and q _ best when the Perform is maximum according to the determined functional relation f (s, r and q) which is equal to the Perform.
Sixthly, synchronizing the parameter group to the perception terminal
And setting the parameter group Param _ best obtained by calculation to the perception terminal, wherein the parameter group Param _ best is { s _ best, r _ best and q _ best }. And the sensing terminal detects according to the set parameter group and uploads the detection result.
Example one
The present embodiment provides a method, as shown in fig. 3, comprising;
the method comprises the following steps that 1, a sensing terminal uploads detection data to a background server in the most loose parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results;
step 2, the background server summarizes all the business performances into a performance index;
step 3, testing the detection data and generating a test record by the background server;
step 4, the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal in a machine learning mode according to the test record;
step 5, obtaining the parameter group when the performance index is maximum according to the functional relation;
and 6, synchronizing the parameter group to the perception terminal.
Based on the same inventive concept, the application also provides a device corresponding to the method in the first embodiment, which is detailed in the second embodiment.
Example two
In the present embodiment, there is provided an apparatus, as shown in fig. 4, comprising: in a second aspect, the present invention provides an apparatus comprising:
the sensing terminal uploads detection data to the background server according to the most relaxed parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results;
the index module is used for summarizing all the service performances into one performance index by the background server;
the generating module is used for testing the detection data and generating a test record by the background server;
the function module is used for determining the functional relation between the parameter configuration and the performance index of the sensing terminal by the background server in a machine learning mode according to the test record;
the calculation module is used for solving the parameter group when the performance index is maximum according to the functional relation;
and the synchronization module synchronizes the parameter group to the perception terminal.
Since the apparatus described in the second embodiment of the present invention is an apparatus used for implementing the method of the first embodiment of the present invention, based on the method described in the first embodiment of the present invention, a person skilled in the art can understand the specific structure and the deformation of the apparatus, and thus the details are not described herein. All the devices adopted in the method of the first embodiment of the present invention belong to the protection scope of the present invention.
Although specific embodiments of the invention have been described above, it will be understood by those skilled in the art that the specific embodiments described are illustrative only and are not limiting upon the scope of the invention, and that equivalent modifications and variations can be made by those skilled in the art without departing from the spirit of the invention, which is to be limited only by the appended claims.
Claims (6)
1. A method for quickly adjusting a sensing terminal is characterized in that: the method comprises the following steps:
step 1, a perception terminal uploads detection data to a background server in the most loose parameter configuration;
step 2, the background server summarizes all the business performances into a performance index;
step 3, testing the detection data and generating a test record by the background server;
step 4, the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record;
step 5, obtaining the parameter group when the performance index is maximum according to the functional relation;
and 6, synchronizing the parameter group to the perception terminal.
2. The method of claim 1, wherein the method comprises: the step 1 is further specifically as follows: the sensing terminal uploads detection data to the background server in the most loose parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results.
3. The method of claim 1, wherein the method comprises: the step 4 is further specifically as follows: and the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal in a machine learning mode according to the test record.
4. The utility model provides a device of quick adjustment perception terminal which characterized in that: the method comprises the following steps:
the sensing terminal uploads the detection data to the background server according to the most relaxed parameter configuration;
the index module is used for summarizing all the service performances into one performance index by the background server;
the generating module is used for testing the detection data and generating a test record by the background server;
the function module is used for determining the functional relation between the parameter configuration and the performance index of the sensing terminal according to the test record by the background server;
the calculation module is used for solving the parameter group when the performance index is maximum according to the functional relation;
and the synchronization module synchronizes the parameter group to the perception terminal.
5. The apparatus for fast adjusting the sensing terminal according to claim 4, wherein: the uploading module is further embodied as follows: the sensing terminal uploads detection data to the background server in the most loose parameter configuration, wherein the detection data comprise a plurality of groups of detection results and corresponding algorithm output results.
6. The apparatus for fast adjusting the sensing terminal according to claim 4, wherein: the function module is further specifically: and the background server determines the functional relation between the parameter configuration and the performance index of the sensing terminal in a machine learning mode according to the test record.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010026725.3A CN111209057B (en) | 2020-01-10 | 2020-01-10 | Method and device for quickly adjusting sensing terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010026725.3A CN111209057B (en) | 2020-01-10 | 2020-01-10 | Method and device for quickly adjusting sensing terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111209057A true CN111209057A (en) | 2020-05-29 |
| CN111209057B CN111209057B (en) | 2022-09-09 |
Family
ID=70787105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010026725.3A Active CN111209057B (en) | 2020-01-10 | 2020-01-10 | Method and device for quickly adjusting sensing terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111209057B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024016344A1 (en) * | 2022-07-22 | 2024-01-25 | 北京小米移动软件有限公司 | Reception parameter adjusting methods and apparatuses, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105553769A (en) * | 2015-12-15 | 2016-05-04 | 北京奇虎科技有限公司 | Data collecting-analyzing system and method |
| US20160182801A1 (en) * | 2014-12-23 | 2016-06-23 | Bryant Genepang Luk | Modifying image parameters using wearable device input |
| CN105893258A (en) * | 2016-03-31 | 2016-08-24 | 中电海康集团有限公司 | Performance optimizing method and tool based on artificial fish school algorithm |
| CN108537932A (en) * | 2018-05-10 | 2018-09-14 | 北京世拓方舟科技有限公司 | A kind of intelligent quick identity verification system of hidden |
| CN110598648A (en) * | 2019-09-17 | 2019-12-20 | 江苏慧眼数据科技股份有限公司 | Video face detection method, video face detection unit and system |
-
2020
- 2020-01-10 CN CN202010026725.3A patent/CN111209057B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160182801A1 (en) * | 2014-12-23 | 2016-06-23 | Bryant Genepang Luk | Modifying image parameters using wearable device input |
| CN105553769A (en) * | 2015-12-15 | 2016-05-04 | 北京奇虎科技有限公司 | Data collecting-analyzing system and method |
| CN105893258A (en) * | 2016-03-31 | 2016-08-24 | 中电海康集团有限公司 | Performance optimizing method and tool based on artificial fish school algorithm |
| CN108537932A (en) * | 2018-05-10 | 2018-09-14 | 北京世拓方舟科技有限公司 | A kind of intelligent quick identity verification system of hidden |
| CN110598648A (en) * | 2019-09-17 | 2019-12-20 | 江苏慧眼数据科技股份有限公司 | Video face detection method, video face detection unit and system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024016344A1 (en) * | 2022-07-22 | 2024-01-25 | 北京小米移动软件有限公司 | Reception parameter adjusting methods and apparatuses, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111209057B (en) | 2022-09-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3863222A1 (en) | Service quality evaluation model training method and device | |
| WO2018141252A1 (en) | Facial tracking method, apparatus, storage medium and electronic device | |
| US11683234B2 (en) | Server outlier detection | |
| CN110659397B (en) | Behavior detection method and device, electronic equipment and storage medium | |
| WO2019214309A1 (en) | Model test method and device | |
| CN110633610A (en) | Student state detection algorithm based on YOLO | |
| US10388005B2 (en) | Sensor system for data enhancement | |
| US20200272524A1 (en) | Method and system for auto-setting of image acquisition and processing modules and of sharing resources in large scale video systems | |
| CN110221979A (en) | Performance test methods, device, equipment and the storage medium of application program | |
| US20220237917A1 (en) | Video comparison method and apparatus, computer device, and storage medium | |
| CN110969045B (en) | Behavior detection method and device, electronic equipment and storage medium | |
| CN111915020A (en) | Method and device for updating detection model and storage medium | |
| WO2022135539A1 (en) | Method and apparatus for processing device configuration parameters, method and apparatus for data analysis, computing device, computer readable storage medium, and computer program product | |
| WO2023056896A1 (en) | Definition determination method and apparatus, and device | |
| CN111209057B (en) | Method and device for quickly adjusting sensing terminal | |
| CN110503092B (en) | Improved SSD monitoring video target detection method based on field adaptation | |
| CN110716803A (en) | Computer system, resource allocation method and image identification method thereof | |
| KR101942892B1 (en) | Client terminal performing the hybrid machine vision and method of thereof | |
| CN108596068B (en) | Method and device for recognizing actions | |
| US11823359B2 (en) | Systems and methods for leveling images | |
| CN111768729A (en) | VR scene automatic explanation method, system and storage medium | |
| CN112995650B (en) | Method and device for detecting video continuity of camera | |
| CN115474108A (en) | Event monitoring system and method based on edge calculation | |
| CN111553408B (en) | Automatic test method for video recognition software | |
| KR20220056599A (en) | Learning-data generation method and apparatus for improving cctv image quality |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Method and device for quickly adjusting sensing terminal Effective date of registration: 20230217 Granted publication date: 20220909 Pledgee: Bank of China Limited Xiamen hi tech Park sub branch Pledgor: XIAMEN RUIWEI INFORMATION TECHNOLOGY CO.,LTD. Registration number: Y2023980032829 |