CN103310373A - Bank counter service evaluation system based on audio and video recognition and method thereof - Google Patents
Bank counter service evaluation system based on audio and video recognition and method thereof Download PDFInfo
- Publication number
- CN103310373A CN103310373A CN2013100797781A CN201310079778A CN103310373A CN 103310373 A CN103310373 A CN 103310373A CN 2013100797781 A CN2013100797781 A CN 2013100797781A CN 201310079778 A CN201310079778 A CN 201310079778A CN 103310373 A CN103310373 A CN 103310373A
- Authority
- CN
- China
- Prior art keywords
- video
- service
- audio
- svm classifier
- evaluation result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a bank counter service evaluation system based on audio and video recognition and a method thereof. The system comprises video acquiring equipment, audio acquiring equipment and a background processing unit, wherein the video acquiring equipment and the audio acquiring equipment are connected with an analog-digital converter through corresponding interfaces respectively; the analog-digital converter is connected with a service evaluation unit through a digital signal processor; the service evaluation unit comprises an SVM (Support Vector Machine) classifier training module, an SVM classifier recognition module and a database; each typical sample is input into the SVM classifier training module respectively; the SVM classifier training module is used for outputting an SVM classifier model; an audio sample signal to be judged is input into the SVM classifier recognition module; and the SVM classifier recognition module is used for loading a trained SVM classifier, and outputting a service evaluation result. According to the system, acquired videos and audios are processed and recognized, and the evaluation of customers on the bank counter service quality is acquired intelligently, so that the workload of workers is reduced, and the time of the customers is saved; and the system is based on the recognition accuracy of an SVM method.
Description
Technical field
The present invention relates to a kind of bank's cabinet face service evaluation system and method based on audio frequency, video identification.
Background technology
Develop rapidly and the improving constantly of people's income level along with socioeconomic, the competition between each large-scale commerce bank of China also is growing more intense.Commercial bank is centered by customer value and is the commerce services tissue that client continues to provide the ability of creating the wealth and system's solution.Bank service is that bank is platform with the channel, is all activities, process or the memory that carrier is created client's consumption experience with the product, the bank service quality then is that bank is the regulation that consumption experience that the client creates can satisfy the bank service build-in attribute, reaches the degree of user expectation.For commercial bank, bank's cabinet face service quality is to judge the important dimension of current operation result, is again the important weathervane that embodies following investment value.Therefore, service quality is the important component part of bank core competitive power, not only is subjected to the great attention of commercial bank, also is subjected to supervision department, society, bank client and investor's common concern simultaneously.
In order to improve bank's cabinet face service quality, improve customer experience, the classic method that each large-scale commerce bank adopts is: the service quality evaluation table of placing papery at the cabinet face, after client's business handling is intact, the requirement client fills in or fills in this evaluation form voluntarily by the client, regularly these evaluation tables of collection are manually added up by the staff, thereby obtain client's cabinet face quality of service feedback information.Yet cabinet face-port mouth is more in bank, and when the business handling amount was big, the workload that the staff adds up the papery evaluation table was very big, and makes mistakes easily, can not react client's service evaluation objectively; On the other hand, thisly traditional finish the quality time that mode that evaluation table fills in needs to delay the client to a certain extent by the client, under a lot of situations, the client is not very willing to take time to finish filling in of evaluation table, and this has brought big obstacle to the feedback information that bank obtains service quality.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, a kind of bank's cabinet face service evaluation system and method based on audio frequency, video identification is provided, video when obtaining client's transacting business by video, audio collecting device, audio frequency, again by video, the audio frequency gathered are handled and are identified, obtain the client intelligently to the evaluation of bank's cabinet face service quality, need not the staff and artificially collect the customer evaluation table, reduce staff's workload, and improved service evaluation result's accuracy and fiduciary level.
The objective of the invention is to be achieved through the following technical solutions: a kind of bank's cabinet face service evaluation system based on audio frequency, video identification, it comprises the one or more video capture device of hall of bank or cabinet face, one or more background processing unit of being located at the audio collecting device of bank's cabinet face and being located at administrative center be located at, background processing unit comprises video processing module and audio processing modules, and video processing module comprises video interface, analog to digital converter a, video digital signal processor and Video service evaluation unit; Audio processing modules comprises audio interface, analog to digital converter b, audio digital signal processor and audio service evaluation unit; Video capture device links to each other with analog to digital converter a by video interface, and analog to digital converter a is connected with the Video service evaluation unit by video digital signal processor; Audio collecting device is connected with analog to digital converter b by audio interface, and analog to digital converter b is connected with the audio service evaluation unit by audio digital signal processor;
Described Video service evaluation unit and audio service evaluation unit include for the svm classifier device training module that typical sample is carried out features training, are used for the svm classifier device identification module that video to be discriminated and audio samples are identified and the database that is used for the storage typical sample; Described typical sample comprises efficiency of service video typical sample, efficiency of service audio frequency typical sample, service quality video typical sample, service quality audio frequency typical sample, attitude video typical sample and attitude audio frequency typical sample, each typical sample is imported svm classifier device training module respectively, svm classifier device training module output svm classifier device model; The audio samples signal to be discriminated of the video sample signal to be discriminated of video digital signal processor output and audio digital signal processor output is imported svm classifier device identification module respectively, svm classifier device identification module loads and to have trained the svm classifier device of finishing, and output is based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency.
Further, described svm classifier device training module comprises data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device drill circuit, the input end of data pre-process circuit connects the typical sample signal, the output terminal of data pre-process circuit links to each other with an input end of svm classifier device drill circuit with the proper vector normalization circuit by characteristic extracting circuit successively, and another input end of svm classifier device drill circuit is imported the corresponding service evaluation result of this sample.
Further, described svm classifier device identification module comprises the data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device identification circuit, the input end of data pre-process circuit connects sample signal to be discriminated, the output terminal of data pre-process circuit links to each other with the input end of proper vector normalization circuit with svm classifier device identification circuit by characteristic extracting circuit successively, be loaded with in the svm classifier device identification circuit and train the svm classifier device of finishing, the proper vector of proper vector normalization circuit output inputs in the svm classifier device finishes identification, and the svm classifier device is exported the service evaluation result of sample to be discriminated.
A kind of bank's cabinet face service evaluation method based on audio frequency, video identification, it may further comprise the steps:
S1: video capture device and audio collecting device are gathered video and the audio frequency of client when cabinet face transacting business respectively;
S2: video file inputs to analog to digital converter a by video interface and carries out analog to digital conversion, is converted to enter video digital signal processor behind the video signal and handle, and transfers to the Video service evaluation unit and carry out service evaluation after digital signal processing; Audio file inputs to analog to digital converter b by audio interface and carries out analog to digital conversion, is converted to enter audio digital signal processor behind the audio digital signals and handle, and transfers to the audio service evaluation unit and carry out service evaluation after digital signal processing;
The service evaluation that described Video service evaluation unit carries out may further comprise the steps:
The training of S211:SVM sorter, it comprises following substep:
S2111: respectively efficiency of service video typical sample, service quality video typical sample and the attitude video typical sample that is pre-stored in the database carried out the data pre-service;
S2112: feature extraction, composition characteristic vector;
S2113: normalization proper vector;
S2114: confirm the service evaluation result that this video typical sample is corresponding;
S2115: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2116: obtain svm classifier device model and storage;
The identification of S212:SVM sorter, it comprises following substep:
S2121: the video sample signal to be discriminated to video digital signal processor output carries out the data pre-service;
S2122: feature extraction, composition characteristic vector;
S2123: normalization proper vector;
S2124: load and trained the svm classifier device of finishing;
S2125: proper vector inputed in the svm classifier device identify;
S2126: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this video sample signal to be discriminated, service quality evaluation result and attitude evaluation result;
The service evaluation that described audio service evaluation unit carries out may further comprise the steps:
The training of S221:SVM sorter, it comprises following substep:
S2211: respectively efficiency of service audio frequency typical sample, service quality audio frequency typical sample and the attitude audio frequency typical sample that is pre-stored in the database carried out the data pre-service;
S2212: feature extraction, composition characteristic vector;
S2213: normalization proper vector;
S2214: confirm the service evaluation result that this audio frequency typical sample is corresponding;
S2215: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2216: obtain svm classifier device model and storage;
The identification of S222:SVM sorter, it comprises following substep:
S2221: the audio samples signal to be discriminated to audio digital signal processor output carries out the data pre-service;
S2222: feature extraction, composition characteristic vector;
S2223: normalization proper vector;
S2224: load and trained the svm classifier device of finishing;
S2225: proper vector inputed in the svm classifier device identify;
S2226: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this audio samples signal to be discriminated, service quality evaluation result and attitude evaluation result;
S3: comprehensively based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency, obtain the integrated service evaluation.
Further, described step S3 comprises following substep:
S301: the weighting coefficient that efficiency of service evaluation result, service quality evaluation result, attitude evaluation result are set;
S302: will estimate score and multiply by its corresponding weighting coefficient respectively based on efficiency of service evaluation score, service quality evaluation score, the attitude of video, addition obtains the service evaluation score based on video; Will based on the efficiency of service evaluation result score of audio frequency, service quality evaluation as a result score, attitude evaluation result score multiply by its corresponding weighting coefficient respectively, addition obtains the service evaluation score based on audio frequency;
S303: arrange based on the service evaluation result of video and weighting coefficient based on the service evaluation result of audio frequency;
S304: multiply by its corresponding weighting coefficient respectively based on the service evaluation score of video and service evaluation score based on audio frequency, addition obtains the integrated service evaluation.
The invention has the beneficial effects as follows:
Video, audio frequency when (1) obtaining client's transacting business by video, audio collecting device, again by video, the audio frequency gathered are handled and are identified, obtain the client intelligently to the evaluation of bank's cabinet face service quality, intelligent degree height, need not the staff and collect customer evaluation table and complicate statistics, reduce staff's workload, and improved service evaluation result's accuracy and fiduciary level;
(2) need not the client and fill in the service evaluation table, saved client's time, can avoid the client to produce the dislike mood;
(3) support vector machine SVM method is adopted in the identification of video and audio file, has further improved recognition accuracy, helps bank to obtain the objective service quality evaluation of client exactly.
Description of drawings
Fig. 1 is bank of the present invention cabinet face service evaluation system structural representation block diagram;
Fig. 2 is the structural representation block diagram of svm classifier device training module;
Fig. 3 is the structural representation block diagram of svm classifier device identification module;
Fig. 4 is bank of the present invention cabinet face service evaluation method flow diagram.
Embodiment
Below in conjunction with accompanying drawing technical scheme of the present invention is described in further detail, but protection scope of the present invention is not limited to the following stated.
As shown in Figure 1, a kind of bank's cabinet face service evaluation system based on audio frequency, video identification, it comprises the one or more video capture device of hall of bank or cabinet face, one or more background processing unit of being located at the audio collecting device of bank's cabinet face and being located at administrative center be located at, background processing unit comprises video processing module and audio processing modules, and video processing module comprises video interface, analog to digital converter a, video digital signal processor and Video service evaluation unit; Audio processing modules comprises audio interface, analog to digital converter b, audio digital signal processor and audio service evaluation unit; Video capture device links to each other with analog to digital converter a by video interface, and analog to digital converter a is connected with the Video service evaluation unit by video digital signal processor; Audio collecting device is connected with analog to digital converter b by audio interface, and analog to digital converter b is connected with the audio service evaluation unit by audio digital signal processor;
Described Video service evaluation unit and audio service evaluation unit include for the svm classifier device training module that typical sample is carried out features training, are used for the svm classifier device identification module that video to be discriminated and audio samples are identified and the database that is used for the storage typical sample; Described typical sample comprises efficiency of service video typical sample, efficiency of service audio frequency typical sample, service quality video typical sample, service quality audio frequency typical sample, attitude video typical sample and attitude audio frequency typical sample, each typical sample is imported svm classifier device training module respectively, svm classifier device training module output svm classifier device model; The audio samples signal to be discriminated of the video sample signal to be discriminated of video digital signal processor output and audio digital signal processor output is imported svm classifier device identification module respectively, svm classifier device identification module loads and to have trained the svm classifier device of finishing, and output is based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency.
As shown in Figure 2, svm classifier device training module comprises data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device drill circuit, the input end of data pre-process circuit connects the typical sample signal, the output terminal of data pre-process circuit links to each other with an input end of svm classifier device drill circuit with the proper vector normalization circuit by characteristic extracting circuit successively, and another input end of svm classifier device drill circuit is imported the corresponding service evaluation result of this sample.
As shown in Figure 3, svm classifier device identification module comprises the data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device identification circuit, the input end of data pre-process circuit connects sample signal to be discriminated, the output terminal of data pre-process circuit links to each other with the input end of proper vector normalization circuit with svm classifier device identification circuit by characteristic extracting circuit successively, be loaded with in the svm classifier device identification circuit and train the svm classifier device of finishing, the proper vector of proper vector normalization circuit output inputs in the svm classifier device finishes identification, and the svm classifier device is exported the service evaluation result of sample to be discriminated.
As shown in Figure 4, a kind of bank's cabinet face service evaluation method based on audio frequency, video identification, it may further comprise the steps:
S1: video capture device and audio collecting device are gathered video and the audio frequency of client when cabinet face transacting business respectively;
S2: video file inputs to analog to digital converter a by video interface and carries out analog to digital conversion, is converted to enter video digital signal processor behind the video signal and handle, and transfers to the Video service evaluation unit and carry out service evaluation after digital signal processing; Audio file inputs to analog to digital converter b by audio interface and carries out analog to digital conversion, is converted to enter audio digital signal processor behind the audio digital signals and handle, and transfers to the audio service evaluation unit and carry out service evaluation after digital signal processing;
The service evaluation that described Video service evaluation unit carries out may further comprise the steps:
The training of S211:SVM sorter, it comprises following substep:
S2111: respectively efficiency of service video typical sample, service quality video typical sample and the attitude video typical sample that is pre-stored in the database carried out the data pre-service;
S2112: feature extraction, composition characteristic vector;
S2113: normalization proper vector;
S2114: confirm the service evaluation result that this video typical sample is corresponding;
S2115: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2116: obtain svm classifier device model and storage;
The identification of S212:SVM sorter, it comprises following substep:
S2121: the video sample signal to be discriminated to video digital signal processor output carries out the data pre-service;
S2122: feature extraction, composition characteristic vector;
S2123: normalization proper vector;
S2124: load and trained the svm classifier device of finishing;
S2125: proper vector inputed in the svm classifier device identify;
S2126: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this video sample signal to be discriminated, service quality evaluation result and attitude evaluation result;
The service evaluation that described audio service evaluation unit carries out may further comprise the steps:
The training of S221:SVM sorter, it comprises following substep:
S2211: respectively efficiency of service audio frequency typical sample, service quality audio frequency typical sample and the attitude audio frequency typical sample that is pre-stored in the database carried out the data pre-service;
S2212: feature extraction, composition characteristic vector;
S2213: normalization proper vector;
S2214: confirm the service evaluation result that this audio frequency typical sample is corresponding;
S2215: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2216: obtain svm classifier device model and storage;
The identification of S222:SVM sorter, it comprises following substep:
S2221: the audio samples signal to be discriminated to audio digital signal processor output carries out the data pre-service;
S2222: feature extraction, composition characteristic vector;
S2223: normalization proper vector;
S2224: load and trained the svm classifier device of finishing;
S2225: proper vector inputed in the svm classifier device identify;
S2226: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this audio samples signal to be discriminated, service quality evaluation result and attitude evaluation result;
S3: comprehensively based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency, obtain the integrated service evaluation.
Further, described step S3 comprises following substep:
S301: the weighting coefficient that efficiency of service evaluation result, service quality evaluation result, attitude evaluation result are set;
S302: will estimate score and multiply by its corresponding weighting coefficient respectively based on efficiency of service evaluation score, service quality evaluation score, the attitude of video, addition obtains the service evaluation score based on video; Will based on the efficiency of service evaluation result score of audio frequency, service quality evaluation as a result score, attitude evaluation result score multiply by its corresponding weighting coefficient respectively, addition obtains the service evaluation score based on audio frequency;
S303: arrange based on the service evaluation result of video and weighting coefficient based on the service evaluation result of audio frequency;
S304: multiply by its corresponding weighting coefficient respectively based on the service evaluation score of video and service evaluation score based on audio frequency, addition obtains the integrated service evaluation.
Claims (5)
1. bank's cabinet face service evaluation system based on audio frequency, video identification, it is characterized in that: it comprises the one or more video capture device of hall of bank or cabinet face, one or more background processing unit of being located at the audio collecting device of bank's cabinet face and being located at administrative center be located at, background processing unit comprises video processing module and audio processing modules, and video processing module comprises video interface, analog to digital converter a, video digital signal processor and Video service evaluation unit; Audio processing modules comprises audio interface, analog to digital converter b, audio digital signal processor and audio service evaluation unit; Video capture device links to each other with analog to digital converter a by video interface, and analog to digital converter a is connected with the Video service evaluation unit by video digital signal processor; Audio collecting device is connected with analog to digital converter b by audio interface, and analog to digital converter b is connected with the audio service evaluation unit by audio digital signal processor;
Described Video service evaluation unit and audio service evaluation unit include for the svm classifier device training module that typical sample is carried out features training, are used for the svm classifier device identification module that video to be discriminated and audio samples are identified and the database that is used for the storage typical sample; Described typical sample comprises efficiency of service video typical sample, efficiency of service audio frequency typical sample, service quality video typical sample, service quality audio frequency typical sample, attitude video typical sample and attitude audio frequency typical sample, each typical sample is imported svm classifier device training module respectively, svm classifier device training module output svm classifier device model; The audio samples signal to be discriminated of the video sample signal to be discriminated of video digital signal processor output and audio digital signal processor output is imported svm classifier device identification module respectively, svm classifier device identification module loads and to have trained the svm classifier device of finishing, and output is based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency.
2. according to claim 1 a kind of based on audio frequency, bank's cabinet face service evaluation system of video identification, it is characterized in that: described svm classifier device training module comprises the data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device drill circuit, the input end of data pre-process circuit connects the typical sample signal, the output terminal of data pre-process circuit links to each other with an input end of svm classifier device drill circuit with the proper vector normalization circuit by characteristic extracting circuit successively, and another input end of svm classifier device drill circuit is imported the corresponding service evaluation result of this sample.
3. according to claim 1 a kind of based on audio frequency, bank's cabinet face service evaluation system of video identification, it is characterized in that: described svm classifier device identification module comprises the data pre-process circuit, characteristic extracting circuit, proper vector normalization circuit and svm classifier device identification circuit, the input end of data pre-process circuit connects sample signal to be discriminated, the output terminal of data pre-process circuit links to each other with the input end of proper vector normalization circuit with svm classifier device identification circuit by characteristic extracting circuit successively, be loaded with in the svm classifier device identification circuit and train the svm classifier device of finishing, the proper vector of proper vector normalization circuit output inputs in the svm classifier device finishes identification, and the svm classifier device is exported the service evaluation result of sample to be discriminated.
4. bank's cabinet face service evaluation method based on audio frequency, video identification, it is characterized in that: it may further comprise the steps:
S1: video capture device and audio collecting device are gathered video and the audio frequency of client when cabinet face transacting business respectively;
S2: video file inputs to analog to digital converter a by video interface and carries out analog to digital conversion, is converted to enter video digital signal processor behind the video signal and handle, and transfers to the Video service evaluation unit and carry out service evaluation after digital signal processing; Audio file inputs to analog to digital converter b by audio interface and carries out analog to digital conversion, is converted to enter audio digital signal processor behind the audio digital signals and handle, and transfers to the audio service evaluation unit and carry out service evaluation after digital signal processing;
The service evaluation that described Video service evaluation unit carries out may further comprise the steps:
The training of S211:SVM sorter, it comprises following substep:
S2111: respectively efficiency of service video typical sample, service quality video typical sample and the attitude video typical sample that is pre-stored in the database carried out the data pre-service;
S2112: feature extraction, composition characteristic vector;
S2113: normalization proper vector;
S2114: confirm the service evaluation result that this video typical sample is corresponding;
S2115: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2116: obtain svm classifier device model and storage;
The identification of S212:SVM sorter, it comprises following substep:
S2121: the video sample signal to be discriminated to video digital signal processor output carries out the data pre-service;
S2122: feature extraction, composition characteristic vector;
S2123: normalization proper vector;
S2124: load and trained the svm classifier device of finishing;
S2125: proper vector inputed in the svm classifier device identify;
S2126: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this video sample signal to be discriminated, service quality evaluation result and attitude evaluation result;
The service evaluation that described audio service evaluation unit carries out may further comprise the steps:
The training of S221:SVM sorter, it comprises following substep:
S2211: respectively efficiency of service audio frequency typical sample, service quality audio frequency typical sample and the attitude audio frequency typical sample that is pre-stored in the database carried out the data pre-service;
S2212: feature extraction, composition characteristic vector;
S2213: normalization proper vector;
S2214: confirm the service evaluation result that this audio frequency typical sample is corresponding;
S2215: with proper vector and this service evaluation result respectively as the input and output of svm classifier device drill circuit, training svm classifier device;
S2216: obtain svm classifier device model and storage;
The identification of S222:SVM sorter, it comprises following substep:
S2221: the audio samples signal to be discriminated to audio digital signal processor output carries out the data pre-service;
S2222: feature extraction, composition characteristic vector;
S2223: normalization proper vector;
S2224: load and trained the svm classifier device of finishing;
S2225: proper vector inputed in the svm classifier device identify;
S2226: the output according to the svm classifier device obtains the corresponding efficiency of service evaluation result of this audio samples signal to be discriminated, service quality evaluation result and attitude evaluation result;
S3: comprehensively based on the efficiency of service evaluation result of video, service quality evaluation result, attitude evaluation result with based on efficiency of service evaluation result, service quality evaluation result, the attitude evaluation result of audio frequency, obtain the integrated service evaluation.
5. a kind of bank's cabinet face service evaluation method based on audio frequency, video identification according to claim 4, it is characterized in that: described step S3 comprises following substep:
S301: the weighting coefficient that efficiency of service evaluation result, service quality evaluation result, attitude evaluation result are set;
S302: will estimate score and multiply by its corresponding weighting coefficient respectively based on efficiency of service evaluation score, service quality evaluation score, the attitude of video, addition obtains the service evaluation score based on video; Will based on the efficiency of service evaluation result score of audio frequency, service quality evaluation as a result score, attitude evaluation result score multiply by its corresponding weighting coefficient respectively, addition obtains the service evaluation score based on audio frequency;
S303: arrange based on the service evaluation result of video and weighting coefficient based on the service evaluation result of audio frequency;
S304: multiply by its corresponding weighting coefficient respectively based on the service evaluation score of video and service evaluation score based on audio frequency, addition obtains the integrated service evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100797781A CN103310373A (en) | 2013-03-13 | 2013-03-13 | Bank counter service evaluation system based on audio and video recognition and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100797781A CN103310373A (en) | 2013-03-13 | 2013-03-13 | Bank counter service evaluation system based on audio and video recognition and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103310373A true CN103310373A (en) | 2013-09-18 |
Family
ID=49135555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100797781A Pending CN103310373A (en) | 2013-03-13 | 2013-03-13 | Bank counter service evaluation system based on audio and video recognition and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103310373A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106128448A (en) * | 2016-08-30 | 2016-11-16 | 金嵩 | A kind of novel court sound collector |
| CN108898115A (en) * | 2018-07-03 | 2018-11-27 | 北京大米科技有限公司 | data processing method, storage medium and electronic equipment |
| CN109740020A (en) * | 2018-12-26 | 2019-05-10 | 秒针信息技术有限公司 | Data processing method, device, storage medium and processor |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080183646A1 (en) * | 2002-12-06 | 2008-07-31 | Harris Christopher K | Effective multi-class support vector machine classification |
| CN201387622Y (en) * | 2009-04-30 | 2010-01-20 | 罗杰 | Customer evaluation information acquisition and processing system |
-
2013
- 2013-03-13 CN CN2013100797781A patent/CN103310373A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080183646A1 (en) * | 2002-12-06 | 2008-07-31 | Harris Christopher K | Effective multi-class support vector machine classification |
| CN201387622Y (en) * | 2009-04-30 | 2010-01-20 | 罗杰 | Customer evaluation information acquisition and processing system |
Non-Patent Citations (3)
| Title |
|---|
| 孙华丽 等: "基于支持向量机的移动电话顾客满意度评价系统", 《计算机仿真》 * |
| 庄越挺 等: "《网上多媒体信息分析与检索》", 30 September 2002, 清华大学出版社 * |
| 王嘉鋆: "基于支持向量机的科技公共服务平台满意度评价研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106128448A (en) * | 2016-08-30 | 2016-11-16 | 金嵩 | A kind of novel court sound collector |
| CN108898115A (en) * | 2018-07-03 | 2018-11-27 | 北京大米科技有限公司 | data processing method, storage medium and electronic equipment |
| CN108898115B (en) * | 2018-07-03 | 2021-06-04 | 北京大米科技有限公司 | Data processing method, storage medium and electronic device |
| CN109740020A (en) * | 2018-12-26 | 2019-05-10 | 秒针信息技术有限公司 | Data processing method, device, storage medium and processor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2007142656A3 (en) | Data center uninterruptible power distribution architecture | |
| WO2009012451A3 (en) | Power supply for a data center | |
| CN110626662A (en) | Image recognition-based garbage self-classification method and device | |
| WO2001053940A3 (en) | Method and apparatus for determining the order of streaming modules | |
| CN103310373A (en) | Bank counter service evaluation system based on audio and video recognition and method thereof | |
| CN102750366B (en) | Video search system and method based on natural interactive import and video search server | |
| CN101719696B (en) | Centralized bus protection device and method applied to digitized transformer station | |
| CN201387385Y (en) | Integration piezoelectricity multi-channel scanning structural health monitoring system based on computer bus | |
| CN202075878U (en) | Multimedia teaching control system | |
| CN201878155U (en) | Device for acquiring signals and monitoring data and ground satellite comprehensive test system with same | |
| CN114169778A (en) | Enterprise work task distribution system based on artificial intelligence | |
| WO2004095213A8 (en) | Method and system of processing billing data | |
| CN101620723A (en) | Offline intelligent image information processing method and intelligent image information processing system | |
| CN210697629U (en) | Signal acquisition device, mobile terminal and signal analysis system | |
| CN106357199A (en) | Energy-saving motor control system and energy-saving method | |
| CN102609513B (en) | Multi-data-source database system for condition monitoring of electric transmission and transformation equipment | |
| CN202600936U (en) | Remote controller with voice function | |
| CN106094642A (en) | A kind of monitoring system of big data platform | |
| CN110007636A (en) | A kind of internet-of-things terminal based on power grid acquires equipment in real time | |
| CN106056234A (en) | Transformer capacity determination method and device | |
| WO2018148881A1 (en) | Interactive teaching robot and processing method therefor | |
| CN202041287U (en) | Analysis circuit applied to spectral analysis equipment | |
| CN206209708U (en) | A kind of intelligent mathematical computing device | |
| CN206819731U (en) | A kind of novel voice identification equipment | |
| CN105346904A (en) | Intelligent cable weighing data collection device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130918 |
|
| RJ01 | Rejection of invention patent application after publication |