CN109167997B - Video quality diagnosis system and method - Google Patents
Video quality diagnosis system and method Download PDFInfo
- Publication number
- CN109167997B CN109167997B CN201811135467.1A CN201811135467A CN109167997B CN 109167997 B CN109167997 B CN 109167997B CN 201811135467 A CN201811135467 A CN 201811135467A CN 109167997 B CN109167997 B CN 109167997B
- Authority
- CN
- China
- Prior art keywords
- detection
- camera
- video
- diagnosis
- video quality
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/002—Diagnosis, testing or measuring for television systems or their details for television cameras
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the field of real-time video monitoring, and particularly relates to a video quality diagnosis system and method. Firstly, a video source acquisition unit acquires a video frame of a camera to be detected and sends the video frame to a video quality diagnosis unit; the video quality diagnosis unit diagnoses the video frames according to the received video frames and the detection threshold value of each type of detection items of each camera to be detected through a detection algorithm corresponding to each type of detection items; a user views the diagnosis result of the camera with the problem through a terminal page; the user confirms the diagnosis result, and if the video quality is normal and the system diagnosis is abnormal, the diagnosis result is added into a false detection result library; and a timing self-learning module in the video quality diagnosis unit corrects the detection threshold of the camera according to the data in the false detection result base. The method can greatly reduce the false detection rate and the missing detection rate of the video quality diagnosis and accurately judge the video quality of the camera.
Description
The invention relates to a divisional application, wherein the application date of the original application is 2017, 03 and 30, the application number is 201710200994.5, and the name is a video quality diagnosis method and system based on a timing self-learning module.
Technical Field
The invention belongs to the field of real-time video monitoring, and particularly relates to a video quality diagnosis system and method.
Background
With the rapid development of the construction of the safe city, the safe city has the characteristics of large scale, multiple point locations, wide area and the like, and is extremely inconvenient to maintain. The usability of the image is identified only by human eyes, so that the labor intensity is high, and the later evidence obtaining is difficult due to lag and unreliability. And also results in huge resource waste for the back-end video analysis, storage, transmission, etc. system.
The video quality diagnosis system accurately analyzes, judges and alarms aiming at common camera faults, video signal interference, video quality degradation and the like of video images such as black screen, blurring, abnormal brightness, noise, color cast, shaking, freezing, PTZ motion detection, artificial shielding and the like. The management level of the city is improved, and the image of the city is updated.
The video quality diagnosis system on the market at present mainly has a set of threshold value for each algorithm. However, in the use process, the texture of the image and the information which can be obtained by the image are different due to the position of the camera. If the same set of threshold values is adopted, more false detection rates and missed detection rates are caused. In addition, in the use process of the camera, the quality of the camera is reduced due to aging and the like, but the purpose of needing to be overhauled is not achieved. Therefore, parameters of the video quality of the camera need to be intelligently adjusted in the diagnosis process so as to avoid the situation that the system cannot accurately judge the video quality of the camera.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a video quality diagnosis system, which solves the problem that the video quality diagnosis error detection rate and omission factor are high in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
comprises a video source acquisition unit (10), a video quality diagnosis unit (20), a data storage unit (30) and a management unit (40),
the video source acquisition unit (10) is used for acquiring a video frame of a camera to be detected;
the video quality diagnosis unit (20) is used for diagnosing the acquired video frames by using the detection threshold value of each type of detection item and the detection algorithm corresponding to each type of detection item to obtain a diagnosis result;
a data storage unit (30) for storing the diagnosis result;
a management unit (40) for interacting with a user, the user being able to view the diagnosis of the camera through the management unit (40);
the video quality diagnosis system also comprises a false detection result library, the user confirms the diagnosis result, and if the video quality is normal and the system diagnosis is abnormal, the diagnosis result is added into the false detection result library.
The video quality diagnosis unit (20) comprises a timing self-learning module (21) and a video quality diagnosis algorithm module (22), wherein the timing self-learning module (21) comprises a picture reading module (211), an image feature extraction module (212), a machine learning module (213) and a threshold value updating module (214),
the image reading module (211) is used for reading the image of the camera in the false detection result library and sending the image of the camera to the image feature extraction module (212);
the image feature extraction module (212) is used for analyzing the picture features and sending the analyzed picture features to the machine learning module (213);
the machine learning module (213) is used for fusing the analyzed picture characteristics with the characteristic data of the original camera to perform machine learning, and obtaining a new detection threshold value of each type of detection items of the false detection camera;
a threshold update module (214) for updating the detection threshold of the camera.
The invention also provides a video diagnosis method, which is characterized by comprising the following steps:
s1, when the detection start time is reached, the video source acquisition unit (10) acquires the video frame of the camera to be detected and sends the video frame to the video quality diagnosis unit (20);
s2, the video quality diagnosis unit (20) diagnoses the video frames according to the received video frames and the detection threshold value of each type of detection items of each camera to be detected through the detection algorithm corresponding to each type of detection items;
s3, after the diagnosis is finished, the diagnosis result is stored in the data storage unit (30);
s4, the user views the diagnosis result of the camera with the problem through the terminal page;
s5, the user confirms the diagnosis result, and if the video quality is normal and the system diagnosis is abnormal, the diagnosis result is added into a false detection result library;
s6, the timing self-learning module (21) in the video quality diagnosis unit (20) corrects the detection threshold value of the camera according to the data in the false detection result base.
Before step S1, the user needs to set the camera to be detected, the time period for detection, and the detection interval, and the system automatically determines the detection start time according to the time period for detection and the detection interval.
The specific operation steps of step S6 are as follows:
s61, the user sets the timing self-learning time, and after the set self-learning time is reached, the system starts a timing self-learning module (21); the timing self-learning module (21) comprises a picture reading module (211), an image feature extraction module (212), a machine learning module (213) and a threshold updating module (214);
s62, the picture reading module (211) reads the picture of the camera in the false detection result base and sends the picture of the camera to the image feature extraction module (212);
s63, the image feature extraction module (212) analyzes the image features and sends the analyzed image features to the machine learning module (213);
s64, the machine learning module (213) combines the analyzed picture features with the feature data of the original camera, namely the picture feature data corresponding to the default detection threshold parameters of the camera, to perform machine learning, and a new detection threshold of each type of detection items of the false-detection camera is obtained;
s65, the threshold value updating module (214) updates the detection threshold value of the camera.
The picture features at least comprise picture brightness features, picture boundary features and picture histogram features.
The detection items comprise black screen detection, fuzzy detection, brightness abnormity detection, noise detection, color cast detection, shake detection, freezing detection, PTZ motion detection and artificial shielding detection.
The video source acquisition unit (10) in step S1 acquires the video frames of the cameras to be detected by polling and concurrently.
The invention also provides a self-learning method for video quality diagnosis, which is characterized by comprising the following steps of:
s1, the picture reading module (211) reads the picture of the camera in the false detection result base and sends the picture of the camera to the image feature extraction module (212); the picture sources of the cameras in the false detection result library are as follows: the user confirms the diagnosis result, and if the video quality is normal and the system diagnosis is abnormal, the diagnosis result is added into the false detection result library;
s2, the image feature extraction module (212) analyzes the picture features and sends the analyzed picture features to the machine learning module (213); the picture characteristics at least comprise picture brightness characteristics, picture boundary characteristics and picture histogram characteristics;
s3, the machine learning module (213) combines the analyzed picture features with the feature data of the original camera, namely the picture feature data corresponding to the default detection threshold parameters of the camera, to perform machine learning, and obtain a new detection threshold of each type of detection items of the false-detection camera; the detection items include: black screen detection, fuzzy detection, brightness anomaly detection, noise detection, color cast detection, jitter detection, freeze detection, PTZ motion detection and artificial shielding detection;
s4, the threshold updating module (214) updates the detection threshold of the camera.
The invention has the advantages that:
(1) the timing self-learning module is arranged in the video quality diagnosis unit, the user terminal checks the diagnosis result of the camera with problems through the terminal page, the user confirms the diagnosis result, if the diagnosis result is normal and the system detection is abnormal, the diagnosis result is added into the false detection result base, and the timing self-learning module corrects the detection threshold value of the camera according to the data in the false detection result base.
(2) The video source acquisition unit acquires the video frames of the cameras to be detected in a polling and concurrent mode, so that each server can simultaneously detect multiple paths of cameras, and the detection efficiency is greatly improved.
(3) The self-learning method for video quality diagnosis corrects the detection threshold of the camera according to the data in the false detection result base, so the method can greatly reduce the false detection rate and the omission factor of the video quality diagnosis and accurately judge the video quality of the camera.
Drawings
FIG. 1 is a general flow diagram of a video quality diagnostic method of the present invention;
FIG. 2 is a flow chart of the timing self-learning module of the present invention correcting the detection threshold of the camera;
FIG. 3 is a block diagram of the video quality diagnostic system of the present invention;
fig. 4 is a block diagram showing the structure of a video quality diagnosis unit according to the present invention.
The reference numerals in the figures have the following meanings:
10-video source acquisition unit 20-video quality diagnosis unit 21-timing self-learning module
22-video quality diagnostic algorithm module 211-picture reading module
212-image feature extraction module 213-machine learning module
214-threshold update module 30-data storage unit 40-management unit
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the video quality diagnosis method based on the timing self-learning module includes the following steps:
s0, the user sets the camera to be detected, the detection time period and the detection interval, and the system automatically determines the detection start time according to the detection time period and the detection interval.
If the user sets the detection time period to be 8: 00-15: 00 in one day and the detection interval is once every 2 hours, the system automatically judges that the detection is started at four time points of 8:00, 10:00, 12:00 and 14: 00.
S1, when the detection start time is reached, the video source obtaining unit 10 obtains the video frame of the camera to be detected, and sends the video frame to the video quality diagnosing unit 20;
s2, the video quality diagnosis unit 20 diagnoses the video frames according to the received video frames and the detection threshold value of each type of detection items of each camera to be detected through the detection algorithm corresponding to each type of detection items;
the detection items comprise black screen detection, fuzzy detection, brightness abnormity detection, noise detection, color cast detection, shake detection, freezing detection, PTZ motion detection and artificial shielding detection;
s3, after the diagnosis is finished, the diagnosis result is stored in the data storage unit 30;
s4, the user views the diagnosis result of the camera with the problem through the terminal page;
s5, the user confirms the diagnosis result, and if the video quality is normal and the system diagnosis is abnormal, the diagnosis result is added into a false detection result library;
s6, the timing self-learning module 21 in the video quality diagnosis unit 20 corrects the detection threshold of the camera according to the data in the false detection result library.
The black screen detection, the fuzzy detection, the brightness abnormity detection, the noise detection, the color cast detection, the shake detection, the freezing detection, the PTZ motion detection and the artificial shielding detection are all realized by the detection method in the prior art.
The camera black screen detection comprises the steps of firstly converting an image into a gray image, carrying out pixel detection according to image pixels, wherein the pixel value is 0-255, the brightness is from dark to light, the color in the corresponding image is from black to white, if the length of a video image is 1920 and the width of the video image is 1080, the total number of the pixels is 1920 multiplied by 1080, if the number of the pixel values of 0 exceeds 80%, the camera black screen is considered, and the 80% is a detection threshold value of the black screen detection.
Fuzzy detection and artificial shielding detection, wherein when a video frame is fuzzy or shielded, the image boundary is not clear, the image definition is detected by using a boundary detection theory, the more fuzzy the image is, the smaller the gradient value of the image is, a detection threshold value is set to obtain the gradient value of the image, and if the gradient value reaches the detection threshold value, the image is considered to be fuzzy.
And detecting abnormal brightness, wherein the color distribution histogram of the normal video image is in uniform distribution, when the brightness of the image is abnormal, the color distribution histogram of the image jumps suddenly, and the abnormal brightness condition of the image can be judged according to the color distribution histogram of the image.
Noise detection, wherein if a point of non-noise point exists in the video image, the point of non-noise point and the gray scale convolution value of the surrounding point in at least one direction are very small according to the noise template; on the contrary, if the point is a noise point, the convolution values in the four directions are all large, and the minimum value of the convolution of the target point in the four directions is obtained, so that whether the point is the noise point can be judged according to the set detection threshold value.
The color cast detection, the color cast of the video image not only has direct relation with the average value of the image chromaticity, but also is related with the chromaticity distribution characteristic of the image, if the chromaticity distribution in the two-dimensional histogram on the a-b chromaticity coordinate plane is basically a single peak value or is distributed more intensively, and the chromaticity average value is larger, the color cast generally exists in the situation, the color cast is more serious when the chromaticity average value is larger, therefore, the ratio of the image average chromaticity D and the chromaticity center distance M, namely the color cast factor is adopted to measure the color cast degree.
And (4) freezing detection, namely analyzing the three frames of pictures at intervals by using image pixels, and judging the pictures to be frozen if the change of the three frames of pictures is less than a set detection threshold value.
PTZ motion and shake detection, and a gray projection algorithm follow the change rule of image sequence gray distribution, and the relative movement distance and the association degree between the images are determined by using the projection curves of the two images. Firstly, the gray video frame image is intercepted, then the line projection and the line projection are respectively carried out, and the projected vector is further utilized to carry out the correlation operation, thus obtaining the displacement of the line direction and the line direction. And if the line displacement is in accordance with the moving distance of the PTZ, the PTZ is considered to move normally.
As shown in fig. 2, the specific operation steps of the timing self-learning module 21 in the video quality diagnosis unit 20 to correct the detection threshold of the camera according to the data in the false detection result base are as follows:
s61, the user sets the timing self-learning time, and after the set self-learning time is reached, the system starts the timing self-learning module 21; the timing self-learning module 21 comprises a picture reading module 211, an image feature extraction module 212, a machine learning module 213 and a threshold updating module 214;
s62, the picture reading module 211 reads the picture of the camera in the false detection result library, and sends the picture of the camera to the image feature extraction module 212;
s63, the image feature extraction module 212 analyzes the image features and sends the analyzed image features to the machine learning module 213;
the picture features comprise picture brightness features, picture boundary features, picture histogram features and the like.
The image feature extraction module 212 analyzes the image features through the detection methods in the prior art, namely, black screen detection, blur detection, luminance anomaly detection, noise detection, color cast detection, shake detection, freeze detection, PTZ motion detection, and artificial occlusion detection.
S64, the machine learning module 213 combines the analyzed picture features with the original camera feature data, i.e. the picture feature data corresponding to the default detection threshold parameters of the camera, to perform machine learning, so as to obtain new detection thresholds of each type of detection items of the false-detection camera;
the machine learning is to learn with old data and new data, and the new detection threshold is generated by machine learning, which is SVM support vector machine learning.
S65, the threshold update module 214 updates the detection threshold of the camera.
The video source acquiring unit 10 in step S1 acquires the video frames of the cameras to be detected by polling and concurrently.
If 10 servers are used for diagnosing the video quality of 1 ten thousand cameras, each server can simultaneously detect 5 paths of cameras through testing, then each server averagely needs to detect 1000 cameras, one server corresponds to 1-1000 cameras, the cameras can be simultaneously detected in at most five paths, five paths are one time, and the detection is sequentially carried out until the detection of 1000 is finished.
As shown in fig. 3, the video quality diagnosis system based on the timing self-learning module comprises a video source obtaining unit 10, a video quality diagnosis unit 20, a data storage unit 30 and a management unit 40, wherein the video source obtaining unit 10 is used for obtaining video frames of a camera to be detected; the video quality diagnosis unit 20 is configured to diagnose the obtained video frame by using a detection threshold of each type of detection item and a detection algorithm corresponding to each type of detection item to obtain a diagnosis result; the data storage unit 30 is used for storing the diagnosis result; the management unit 40 is used for interacting with a user, and the user can view the diagnosis result of the camera through the management unit 40.
The video source obtaining unit 10 is configured to interface with a front-end camera, and obtain a video frame of the front-end camera by using a standard onvif protocol, a GB/T28181 protocol, and a camera private SDK.
For small or individual items, the data storage unit 30 employs a Mysql database; for large scale or combined implementations with other safe city systems, the data storage unit 30 employs an Oracle database.
As shown in fig. 4, the video quality diagnosis unit 20 includes a timing self-learning module 21 and a video quality diagnosis algorithm module 22, the timing self-learning module 21 includes a picture reading module 211, an image feature extraction module 212, a machine learning module 213, and a threshold updating module 214, the picture reading module 211 is configured to read pictures of cameras in the false detection result library and send the pictures of the cameras to the image feature extraction module 212; the image feature extraction module 212 is configured to analyze image features and send the analyzed image features to the machine learning module 213; the machine learning module 213 is configured to perform machine learning by fusing the analyzed image features with feature data of the original camera, and obtain a new detection threshold for each type of detection items of the false-detection camera; the threshold update module 214 is used to update the detection threshold of the camera.
Based on the above framework, the invention can realize the following technical indexes:
fuzzy detection: the detection rate is 100 percent, the omission factor is 1 percent, and the false detection rate is 2 percent;
and (3) detecting brightness abnormality: the detection rate is 100 percent, the omission factor is 1 percent, and the false detection rate is 3 percent;
noise detection: the detection rate is 100 percent, the omission factor is 1 percent, and the false detection rate is 5 percent;
and (3) black screen detection: the detection rate is 100%, the omission factor is 0%, and the false detection rate is 0%;
color cast detection: the detection rate is 100 percent, the omission factor is less than 0 percent, and the false detection rate is less than 1 percent;
and (3) jitter detection: the detection rate is 100 percent, the omission factor is less than 3 percent, and the false detection rate is less than 1 percent;
and (3) freezing detection: the detection rate is 100 percent, the omission factor is less than 0 percent, and the false detection rate is less than 2 percent;
PTZ motion detection: the detection rate is 100 percent, the omission factor is less than 0 percent, and the false detection rate is less than 0 percent;
and (3) artificial shielding detection: the detection rate is 100 percent, the omission factor is 1 percent, and the false detection rate is 3 percent;
the video quality diagnosis method and system based on the timing self-learning module can be applied to the field of video monitoring operation and maintenance in a large scale, improve the accuracy of video quality diagnosis of the camera and guarantee the timeliness and accuracy of video application based on the monitoring camera.
The invention is not to be considered as limited to the specific embodiments shown and described, but is to be understood to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.
Claims (1)
1. A video quality diagnosis system comprising a video source acquisition unit (10), a video quality diagnosis unit (20), a data storage unit (30), and a management unit (40),
the video source acquisition unit (10) is used for acquiring a video frame of a camera to be detected;
the video quality diagnosis unit (20) is used for diagnosing the acquired video frames by using the detection threshold value of each type of detection item and the detection algorithm corresponding to each type of detection item to obtain a diagnosis result;
a data storage unit (30) for storing the diagnosis result;
a management unit (40) for interacting with a user, the user being able to view the diagnosis of the camera through the management unit (40);
the video quality diagnosis unit (20) comprises a timing self-learning module (21), the timing self-learning module (21) comprises a picture reading module (211), an image feature extraction module (212), a machine learning module (213) and a threshold value updating module (214), wherein,
the image reading module (211) is used for reading the image of the camera in the false detection result library and sending the image of the camera to the image feature extraction module (212); the false detection result library is used for storing the pictures of the cameras with normal video quality and abnormal system diagnosis;
the image feature extraction module (212) is used for analyzing the picture features and sending the analyzed picture features to the machine learning module (213);
the machine learning module (213) is used for fusing the analyzed picture characteristics with the characteristic data of the original camera to perform machine learning, and obtaining a new detection threshold value of each type of detection items of the false detection camera;
a threshold update module (214) for updating the detection threshold of the camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811135467.1A CN109167997B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosis system and method |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710200994.5A CN106851263B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosing method based on timing self-learning module and system |
| CN201811135467.1A CN109167997B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosis system and method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710200994.5A Division CN106851263B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosing method based on timing self-learning module and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109167997A CN109167997A (en) | 2019-01-08 |
| CN109167997B true CN109167997B (en) | 2020-04-28 |
Family
ID=59142444
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710200994.5A Active CN106851263B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosing method based on timing self-learning module and system |
| CN201811135467.1A Active CN109167997B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosis system and method |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710200994.5A Active CN106851263B (en) | 2017-03-30 | 2017-03-30 | Video quality diagnosing method based on timing self-learning module and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN106851263B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107197233A (en) * | 2017-06-23 | 2017-09-22 | 安徽大学 | Monitor video quality of data evaluating method and device based on edge calculations model |
| CN107396089A (en) * | 2017-07-03 | 2017-11-24 | 安徽大学 | A kind of video monitoring system monitoring reliability method based on cloud side computation model |
| CN107705334B (en) * | 2017-08-25 | 2020-08-25 | 北京图森智途科技有限公司 | Camera abnormity detection method and device |
| CN110108448B (en) * | 2019-04-30 | 2020-09-29 | 惠州市德赛西威智能交通技术研究院有限公司 | Automatic defect detection method for dynamic logo |
| CN110868585B (en) * | 2019-09-30 | 2021-02-19 | 安徽云森物联网科技有限公司 | Real-time video quality diagnosis system and method |
| CN110677725B (en) * | 2019-10-31 | 2022-04-22 | 飞思达技术（北京）有限公司 | Audio and video anomaly detection method and system based on Internet television service |
| CN110866503B (en) * | 2019-11-19 | 2024-01-05 | 圣点世纪科技股份有限公司 | Abnormality detection method and abnormality detection system for finger vein equipment |
| CN112153373A (en) * | 2020-09-23 | 2020-12-29 | 平安国际智慧城市科技股份有限公司 | Fault identification method and device for bright kitchen range equipment and storage medium |
| CN112949390B (en) * | 2021-01-28 | 2024-03-15 | 浙江大华技术股份有限公司 | Event detection method and device based on video quality |
| CN113850144A (en) * | 2021-08-30 | 2021-12-28 | 湖南黑鲸数据科技有限公司 | Video reliability automatic inspection system based on image recognition |
| CN114928740A (en) * | 2021-11-25 | 2022-08-19 | 广东利通科技投资有限公司 | Video quality detection method, device, equipment, storage medium and program product |
| CN114677574B (en) * | 2022-05-26 | 2022-10-21 | 杭州宏景智驾科技有限公司 | Method and system for diagnosing image fault for automatic driving |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202085261U (en) * | 2010-12-14 | 2011-12-21 | 广东鑫程电子科技有限公司 | Intelligent video diagnosing and monitoring system |
| CN102395043A (en) * | 2011-11-11 | 2012-03-28 | 北京声迅电子股份有限公司 | Video quality diagnosing method |
| CN103281559A (en) * | 2013-05-31 | 2013-09-04 | 于京 | Method and system for detecting quality of video |
| CN105430384A (en) * | 2015-12-10 | 2016-03-23 | 青岛海信网络科技股份有限公司 | Video quality diagnosis method and system |
| CN106454250A (en) * | 2016-11-02 | 2017-02-22 | 北京弘恒科技有限公司 | Intelligent recognition and early warning processing information platform |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102196295B (en) * | 2010-03-05 | 2015-05-20 | 上海风格信息技术股份有限公司 | Method for self-adaptively regulating alarm threshold of parameter of digital television system |
| CN202135267U (en) * | 2010-12-14 | 2012-02-01 | Tcl集团股份有限公司 | Television system possessing functions of error detection and report |
| CN103037239A (en) * | 2011-10-09 | 2013-04-10 | 三星电子（中国）研发中心 | System and method for testing receiving performance of digital video receiving terminal |
| CN202773015U (en) * | 2012-06-19 | 2013-03-06 | 广州市浩云安防科技股份有限公司 | Image quality diagnosis device for video monitoring system |
| CN103024439B (en) * | 2012-12-27 | 2014-12-17 | 贵阳海信电子有限公司 | Detection method and system for smart televisions |
| US8842185B1 (en) * | 2013-03-14 | 2014-09-23 | Microsoft Corporation | HDMI image quality analysis |
| CN104023209B (en) * | 2014-06-12 | 2018-07-13 | 浙江宇视科技有限公司 | A kind of adaptive video diagnostic method and device |
| CN104539936B (en) * | 2014-11-12 | 2017-09-12 | 广州中国科学院先进技术研究所 | There are the system and method for snow noise in a kind of monitoring video |
| CN104469345B (en) * | 2014-12-10 | 2016-05-11 | 北京理工大学 | A kind of video method for diagnosing faults based on image processing |
-
2017
- 2017-03-30 CN CN201710200994.5A patent/CN106851263B/en active Active
- 2017-03-30 CN CN201811135467.1A patent/CN109167997B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202085261U (en) * | 2010-12-14 | 2011-12-21 | 广东鑫程电子科技有限公司 | Intelligent video diagnosing and monitoring system |
| CN102395043A (en) * | 2011-11-11 | 2012-03-28 | 北京声迅电子股份有限公司 | Video quality diagnosing method |
| CN103281559A (en) * | 2013-05-31 | 2013-09-04 | 于京 | Method and system for detecting quality of video |
| CN105430384A (en) * | 2015-12-10 | 2016-03-23 | 青岛海信网络科技股份有限公司 | Video quality diagnosis method and system |
| CN106454250A (en) * | 2016-11-02 | 2017-02-22 | 北京弘恒科技有限公司 | Intelligent recognition and early warning processing information platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106851263A (en) | 2017-06-13 |
| CN106851263B (en) | 2018-11-02 |
| CN109167997A (en) | 2019-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109167997B (en) | Video quality diagnosis system and method | |
| JP4718253B2 (en) | Image abnormality detection device for surveillance camera | |
| CN109299703B (en) | Method and device for carrying out statistics on mouse conditions and image acquisition equipment | |
| JP5241782B2 (en) | Surveillance camera system having a camera abnormality detection device | |
| JP4203736B2 (en) | Image motion detection apparatus and computer program | |
| US10713798B2 (en) | Low-complexity motion detection based on image edges | |
| US8922674B2 (en) | Method and system for facilitating color balance synchronization between a plurality of video cameras and for obtaining object tracking between two or more video cameras | |
| CN113660427A (en) | Image analysis system and method applied to video monitoring tester | |
| US20130002864A1 (en) | Quality checking in video monitoring system | |
| US20220122360A1 (en) | Identification of suspicious individuals during night in public areas using a video brightening network system | |
| CN110826522A (en) | Method and system for monitoring abnormal human behavior, storage medium and monitoring equipment | |
| CN105577983A (en) | Apparatus and method of detecting motion mask | |
| CN109151463B (en) | Video quality diagnosis system and video quality analysis method | |
| JP7446060B2 (en) | Information processing device, program and information processing method | |
| CN102740121A (en) | Video quality diagnosis control system and method applied to video monitoring network | |
| CN111601011A (en) | Automatic alarm method and system based on video stream image | |
| CN109905670A (en) | A kind of multi-stage platform monitoring system | |
| CN112906488A (en) | Security protection video quality evaluation system based on artificial intelligence | |
| CN108174198B (en) | Video image quality diagnosis analysis detection device and application system | |
| KR102511287B1 (en) | Image-based pose estimation and action detection method and appratus | |
| CN116416678A (en) | Method for realizing motion capture and intelligent judgment by using artificial intelligence technology | |
| KR101396838B1 (en) | Video stabilization method and system by selection one of various motion models | |
| JP5710230B2 (en) | Monitoring system and monitoring method | |
| JP3625442B2 (en) | Object detection method, object detection apparatus, and object detection program | |
| KR20220151130A (en) | Image processing method and device, electronic equipment and medium |
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 |