CN113191286B - Image data quality detection and optimization method, system, equipment and medium - Google Patents
Image data quality detection and optimization method, system, equipment and medium Download PDFInfo
- Publication number
- CN113191286B CN113191286B CN202110502065.6A CN202110502065A CN113191286B CN 113191286 B CN113191286 B CN 113191286B CN 202110502065 A CN202110502065 A CN 202110502065A CN 113191286 B CN113191286 B CN 113191286B
- Authority
- CN
- China
- Prior art keywords
- picture
- stream
- target
- image
- detected
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention provides an image data quality detection and adjustment method, a system, equipment and a medium, wherein the method is used for acquiring an image to be detected comprising a video stream and a picture stream, analyzing the image to be detected to determine detection information, determining a detection result according to the detection information, adjusting and optimizing image equipment for shooting the image to be detected, and accurately detecting the image data quality problem of image data by comprehensively analyzing the video stream and the picture stream, so that the method meets the data quality improvement requirement of large data image application, improves the data integrity, the function availability and the customer experience degree of an image equipment using customer, and can adjust and optimize the image equipment in time when the image data quality is abnormal, so that the image data shot later by the image equipment meets the requirement.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method, a system, an apparatus, and a medium for detecting and optimizing image data quality.
Background
Along with the development of technology and the transformation of people to security concepts, the security field is more and more applied to image equipment, and under the same security application scene, along with the increase of the image equipment, the acquired image data volume is also geometrically increased, and the application scene and the reliability of the image data are also greatly enhanced.
Because of the lack of a method for detecting the quality of the image data and timely finding out and optimizing the abnormal quality of the image data, only the image data collected by the image equipment is simply stored, so that more image data with low quality can be found in the process of applying the image data, the integral data integrity of the collected image data is poor, the usability of functions is low, and the customer experience of using customers by the image equipment is reduced.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, an object of the present invention is to provide a method, a system, an apparatus and a medium for detecting and optimizing image data quality, which are used for solving the problems that due to lack of a method for detecting image data quality, in the process of image data application, more image data with low quality is found, resulting in poor data integrity and low functional availability of the whole acquired image data, and the customer experience of using the image device with customers is reduced.
To achieve the above and other related objects, the present invention provides a method for detecting and optimizing image data quality, comprising:
acquiring an image to be detected, wherein the image to be detected comprises a video stream and a picture stream;
Analyzing the image to be detected, and determining detection information, wherein the detection information comprises video stream quality and picture stream quality;
and determining a detection result according to the video stream quality and the picture stream quality, and optimizing the image equipment for shooting the image to be detected according to the detection result.
Optionally, the detection information further includes a picture stream modeling rate, and the determining manner of the picture stream modeling rate includes:
determining a plurality of target pictures from the picture stream;
obtaining target picture structural information of each target picture, and removing target pictures of which the target picture structural information does not comprise target attributes;
and acquiring the number of the target pictures and the total number of the target picture structuring information of the target pictures comprising the target attribute, and determining the picture stream modeling rate.
Optionally, the detection information further includes a target missing rate, and the determining manner of the target missing rate includes:
determining a plurality of target video frame pictures from the video stream;
acquiring target video frame structural information of each target video frame and target picture structural information of each target picture;
performing similarity comparison on the target video frame picture and the target picture to obtain the number of target video frame structural information failing to be compared;
And determining the target beat missing rate according to the number of the target video frame structural information and the total number of the target video frame structural information which are failed to be compared.
Optionally, performing similarity comparison on the target video frame picture and the target picture includes:
acquiring the time of a target video frame picture of a target video frame, and determining a picture preferred time period according to the time of the target video frame picture, wherein the picture preferred time period comprises a first preset time period before the time of the target video frame picture and/or a second preset time period after the time of the target video frame picture;
determining a target picture corresponding to the target video frame according to the picture preferable time period;
and carrying out similarity comparison on the target video frame and the target picture.
Optionally, the initial time of the picture is earlier than the initial time of the video, and the end time of the picture is later than the end time of the video;
the initial time of the picture is the earliest time in the belonging time of each target picture, the end time of the picture is the latest time in the belonging time of each target picture, the initial time of the video is the earliest time in the belonging time of each target video frame picture, and the end time of the video is the latest time in the belonging time of each target video frame picture.
Optionally, optimizing the image device for capturing the image to be detected includes:
and determining an abnormality reason according to the detection result, and adjusting the image equipment, wherein the adjustment comprises at least one of parameter adjustment and position adjustment.
Optionally, after tuning the image device that captures the image to be detected, the method further includes:
acquiring an adjusted image to be detected, which is shot by the adjusted image equipment;
analyzing the adjusted image to be detected, and determining the adjusted detection information and the adjusted detection result;
and determining an adjustment result according to the detection result and the adjusted detection result.
The invention also provides an image data quality detection and adjustment system, which comprises:
the image acquisition module is used for acquiring an image to be detected, wherein the image to be detected comprises a video stream and a picture stream;
the analysis module is used for analyzing the image to be detected and determining detection information, wherein the detection information comprises video stream quality and picture stream quality;
and the tuning module is used for determining a detection result according to the quality of the video stream and the quality of the picture stream, and tuning the image equipment for shooting the image to be detected according to the detection result.
The invention also provides a device comprising a processor, a memory and a communication bus;
the communication bus is used for connecting the processor and the memory;
the processor is configured to execute a computer program stored in the memory to implement the image data quality detection tuning method according to any one of the embodiments described above.
The present invention also provides a computer-readable storage medium having stored thereon a computer program for causing the computer to execute the image data quality detection tuning method according to any one of the above embodiments.
The image data quality detection and optimization method, system, equipment and medium provided by the invention have the following beneficial effects:
the image equipment shooting the image to be detected is optimized according to the detection result, the image data quality problem of the image data can be accurately detected through comprehensive analysis of the video stream and the picture stream, the data quality improvement requirement of large data image application is met, the data integrity, the function availability and the customer experience degree of an image equipment using customer are improved, and the image equipment can be timely optimized when the image data quality is abnormal, so that the image data shot by the image equipment subsequently meets the requirement.
Drawings
Fig. 1 is a flow chart of an image data quality detecting and optimizing method according to an embodiment.
Fig. 2 is a schematic diagram of a scheme framework of the image data quality detection tuning method.
Fig. 3 is a schematic diagram of a structure of an image data quality detecting and optimizing system according to a second embodiment.
Fig. 4 is a schematic hardware structure of an apparatus according to an embodiment.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complicated.
In the following description, numerous details are set forth in order to provide a more thorough explanation of embodiments of the present invention, it will be apparent, however, to one skilled in the art that embodiments of the present invention may be practiced without these specific details, in other embodiments, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring the embodiments of the present invention.
Referring to fig. 1, an embodiment of the present invention provides a method for detecting and optimizing image data quality, including:
s101: and acquiring an image to be detected.
It should be noted that the image to be detected includes a video stream and a picture stream.
The image to be detected may be a real-time image acquired by the image device in real time, or may be a historical image (such as an image delayed for several minutes or days), which is not limited herein.
When the image to be detected is a real-time image, the detection result obtained by the image data quality detection method can be approximately seen as a real-time detection result, so that the image data quality of the image data acquired by the image device can meet the requirements of customers by being beneficial to the adjustment of the image acquisition device of the image data according to the detection result in time by a person skilled in the art.
In some embodiments, the image to be detected is an image of a period of time, the length of which can be set by one skilled in the art as desired.
The image data quality detection can be directly based on self detection of the image equipment, or can be carried out by randomly or pointedly selecting partial image data as an image to be detected after uniformly storing the image data collected by a plurality of image equipment, so as to carry out the image data quality detection.
S102: and analyzing the image to be detected and determining detection information.
The detection information includes video stream quality and picture stream quality.
Alternatively, the detection information may be determined according to a number of to-be-detected picture stream target pictures determined from the picture stream and a number of video frame target pictures determined from the video stream.
The quality of the video stream may also be determined by other relevant criteria established by those skilled in the art, without limitation.
Optionally, when the quality of the video stream is greater than a preset video stream quality threshold, the quality of the video stream is normal, otherwise, the quality of the video stream is abnormal.
Alternatively, when the quality of the video stream is determined according to a plurality of parameters such as resolution, definition, etc., the quality of the video stream may be determined by performing weighted average on each parameter to obtain a comparable value of the quality of the video stream.
Optionally, the resolution and the definition can be set with the grid lines respectively, when the resolution and the definition of the target video frame picture exceed the grid lines, the quality of the video stream is normal, otherwise, the quality of the video stream is abnormal.
Optionally, the determining method of the quality of the picture stream includes:
and obtaining the picture stream target picture structuring information of each initial picture stream target picture, obtaining the picture stream target picture structuring information quantity with unqualified structuring attribute and the picture stream target picture structuring information total quantity, and determining the picture stream quality.
The method can carry out segment statistics on the structural attribute of the picture stream target picture structural information extracted based on the picture stream, and further determine the quality of the obtained picture stream.
The structural attribute can be determined according to an acquisition scene of the image to be detected and an application scene of the image data, and classification of the structural attribute of the structural information of the target picture can be realized according to the existing related technical means, and whether the structural attribute is qualified or not is determined.
Alternatively, the evaluation criterion for whether the structured attribute is acceptable may be determined by a person skilled in the art according to at least one of a collection scene of the image to be detected and an application scene of the image data.
It should be noted that, the above-mentioned determining the quality of the picture stream according to the number of the target picture structuralization information of the picture stream and the total number of the target picture structuralization information of the picture stream, and in some scenes, the quality of the picture stream may also be determined according to the number of the target picture structuralization information of the picture stream and the total number of the target picture structuralization information of the picture stream, which are not limited herein.
In some embodiments, the quality of the picture stream may be determined according to the failure rate of the structural attribute of the picture stream target picture, if the failure rate of the structural attribute of the picture stream target picture is lower than the failure rate threshold, the quality of the picture stream is normal, otherwise, the quality of the picture stream is abnormal.
Optionally, the failure rate of the structural attribute of the picture stream target picture can be determined by the following manner:
the picture stream target picture structured attribute disqualification rate=the number of picture stream target picture structured information/the total number of picture stream target picture structured information that the picture stream structured attribute is disqualified is 100%.
In some embodiments, the quality of the picture stream may be determined according to the qualification rate of the picture stream target picture structure attribute, if the failure rate of the picture stream target picture structure attribute is lower than the qualification rate threshold, the quality of the picture stream is abnormal, otherwise, the quality of the picture stream is normal.
Optionally, the qualification rate of the structural attribute of the picture stream target picture can be determined by the following manner:
the qualification rate of the picture stream target picture structural attribute=the number of picture stream target picture structural information/the total number of picture stream target picture structural information qualified by the picture stream structural attribute is 100%.
In some embodiments, the detection information further includes a picture stream modeling rate, and the determining of the picture stream modeling rate includes:
determining a plurality of initial picture stream target pictures from a picture stream;
obtaining picture stream target picture structure information of each initial picture stream target picture, removing the initial picture stream target picture which does not comprise the target attribute from the picture stream target picture structure information, and taking the initial picture stream target picture comprising the target attribute as the picture stream target picture to be detected;
and acquiring the number of the initial picture stream target pictures and the number of the picture stream target pictures to be detected, and determining the picture stream modeling rate.
Optionally, the determining method of the modeling rate of the picture stream includes:
and acquiring the number of the picture stream target picture structuring information of the initial picture stream target picture comprising the target attribute and the total number of the picture stream target picture structuring information, and determining the picture stream modeling rate.
The total number of the picture stream target picture structuring information is the sum of the number of the picture stream target picture structuring information of each picture stream target picture.
The number statistics can be carried out on the initial picture stream target picture and the picture stream target picture structural information based on the initial picture stream target picture and the picture stream target picture structural information extracted by the picture stream, and the picture stream modeling rate is determined.
Optionally, the picture stream modeling rate is determined as follows:
picture stream modeling rate = picture stream target picture structuring information quantity of picture stream target pictures to be detected/initial picture stream target picture quantity 100%.
Optionally, the initial picture stream target picture excluding the target attribute from the picture stream target picture structuring information includes, but is not limited to: and removing the picture with the missing influence attribute from the initial picture stream target picture, wherein the missing influence attribute comprises at least one of a characteristic value, identity identification information and the like.
The picture stream modeling rate may also be determined according to the number of picture stream target picture structuring information of the picture stream target picture to be detected and the number of initial picture stream target pictures, where the determining manner of the number of picture stream target picture structuring information of the picture stream target picture to be detected may be:
When the initial picture stream target picture is subjected to structural information analysis, the attribute information of the initial picture stream target picture can be obtained, if the attribute information of the initial picture stream target picture does not comprise target attributes, such as feature values, identity identification information and the like, the initial picture stream target picture is not a picture required by modeling, the initial picture stream target picture can be removed in advance, and then picture stream target picture structural information of each initial picture stream target picture (picture stream target picture to be detected) with other attributes comprising the target attributes is determined and summed up, so that the total amount of picture stream target picture structural information to be detected is obtained. The target attribute may be preset by those skilled in the art according to the scene requirement, and is not limited herein.
Optionally, if the picture stream modeling rate is greater than the modeling rate threshold, the picture stream modeling rate is normal, otherwise, the picture stream modeling rate is abnormal.
In some embodiments, the detection information further includes a target miss rate, and the determining method of the target miss rate includes:
determining a plurality of video frame target pictures from a video stream;
obtaining video frame target picture structural information of each video frame target picture;
comparing the video frame target pictures with the to-be-detected picture stream target pictures to obtain the number of video frame target pictures failing to be compared;
And determining the target skip rate according to the number of the video frame target pictures and the total number of the video frame target pictures which are failed to be compared.
Optionally, the image to be detected may also extract the video frame target picture and the video frame target picture structural information in the video stream by using a preset video stream extraction device, and extract the initial picture stream target picture and the picture stream target picture structural information in the picture stream by using a preset picture stream extraction device.
Optionally, the collection of the video frame target picture structural information and the picture stream target picture structural information can be realized by adopting the existing related technology. The target may be a face, a vehicle, a human body, a non-motor vehicle, etc. in the image stream or the video stream, for example, if the shooting scene is a road, the vehicle is taken as the target, the image of the region including the target of the vehicle is selected from several images in the image stream as the initial image stream target image, and the image of the region including the target of the vehicle is selected from several video frames in the video stream as the video frame target image, where the acquisition modes of the initial image stream target image and the video frame target image may also be implemented by adopting related technologies.
Optionally, there is an intersection between the time of the to-be-detected picture stream target picture and the time of the video frame target picture, that is, the time of at least a part of to-be-detected picture stream target pictures corresponding to the to-be-detected image is the same as or similar to the time of the video frame target pictures corresponding to the to-be-detected image.
Optionally, the target picture of the initial picture stream may be a target picture corresponding to all pictures extracted from the picture stream of the whole image to be detected, or may be a target picture corresponding to a part of pictures in the picture stream, which is not limited herein. Similarly, the target pictures of the video frames may be target pictures corresponding to all video frames extracted from the video stream of the whole image to be detected, or may be target pictures corresponding to a part of video frames in the video stream, which is not limited herein.
Alternatively, the time representation may be determined by a person skilled in the art according to needs, for example, may be represented by the time of the image to be detected in its original image, may be represented by the world time of the capturing of the video frame or picture, and so on. It should be noted that, the reference dimension of the time of the to-be-detected picture stream target picture and the reference dimension of the time of the video frame target picture need to be kept consistent or can be converted.
Optionally, the target pictures of the picture stream to be detected may be determined by continuous pictures in the picture stream, or may be determined by discontinuous pictures, for example, an image of an area where a target in one picture is located is obtained from the picture stream at a preset number of intervals as an initial picture stream target picture, or an image of an area where a target in one picture is located is obtained from the picture stream at a certain time interval as an initial picture stream target picture. Similarly, the video frame target picture may be determined by continuous video frames in the video stream, or may be determined by discontinuous video frames, for example, by acquiring a video frame from the video stream at intervals of a preset number of video frames to determine the video frame target picture, or acquiring a video frame from the video stream at intervals of a preset number of video frames to determine the video frame target picture.
Optionally, the determining method of the target beat rate includes:
determining a plurality of video frame target pictures from a video stream;
obtaining video frame target picture structural information of each video frame target picture;
comparing the video frame target picture structural information with the picture stream target picture structural information of the picture stream target picture to be detected to obtain the number of video frame target picture structural information failing to be compared;
And determining the target skip rate according to the number of the video frame target picture structural information and the total number of the video frame target picture structural information which are failed to be compared.
Optionally, a specific method for determining the target beat rate includes: obtaining video frame target picture structured information of each video frame target picture, determining the total number of the video frame target picture structured information, obtaining picture stream target picture structured information of each picture stream target picture to be detected, comparing similarity based on target attributes, obtaining the number of video frame target picture structured information failing to be compared, and determining a target skip rate according to the number of the video frame target picture structured information failing to be compared and the total number of the video frame target picture structured information. The specific implementation of similarity comparison can adopt characteristic value comparison.
The number of the video frame target picture structural information which is failed to be compared is determined after the similarity comparison and screening of the target attributes is carried out based on the video frame target picture structural information which is respectively extracted by the video stream and the picture stream target picture structural information of the target picture of the picture stream to be detected, so that the target skip rate is determined.
Alternatively, the target beat rate may be determined by:
target skip rate= (1-number of failed video frame target picture structured information/total number of video frame target picture structured information) 100%.
The total number of the video frame target picture structural information is the sum of the number of the video frame target picture structural information of each video frame target picture.
If the target missing rate is smaller than the missing rate threshold, the target missing rate is normal, otherwise, the target missing rate is abnormal.
Optionally, in the process of comparing the video frame target picture structural information with the picture stream target picture structural information of the picture stream target picture to be detected, similarity comparison can be performed between the video frame target picture and the picture stream target picture to be detected corresponding to the video frame target picture. The corresponding relationship between the video frame target picture and the picture to be detected and the picture stream target picture can be determined according to the time positions of the video frame target picture and the picture stream target picture in the image to be detected. The video frame target picture can be compared with one picture stream target picture to be detected or can be compared with a plurality of picture stream target pictures to be detected.
In some embodiments, when the similarity comparison is performed, because of time delay in extracting the target pictures of the picture stream to be detected, in order to ensure that the target pictures of the video frame extracted by the video stream are matched with the target pictures with similarity larger than the similarity threshold value in the multiple target pictures of the picture stream to be detected extracted as much as possible, that is, in order to make the target pictures of the video frame and the target pictures of the picture stream to be detected successfully compared as much as possible, the initial time of the pictures can be required to be earlier than the initial time of the video, and the end time of the pictures is later than the end time of the video. The picture initial time is the earliest time in the time of each to-be-detected picture stream target picture, the picture end time is the latest time in the time of each to-be-detected picture stream target picture, the video initial time is the earliest time in the time of each video frame target picture, and the video end time is the latest time in the time of each video frame target picture. Therefore, the problem that the target miss rate is determined to have larger error due to failure in comparison of target pictures of video frames caused by systematic errors such as time delay and the like in picture extraction can be effectively avoided.
In some embodiments, when the similarity is compared, because of time delay in extracting the target pictures of the image stream to be detected, in order to ensure that the target pictures of the video stream to be detected are matched with the target pictures with the similarity not less than the similarity threshold value in the multiple target pictures of the image stream to be detected, the time period [ V ] of the video stream to be detected for extracting the target pictures of the video frame is as long as possible start ，V end ]Time period [ I ] of to-be-detected picture stream target picture required to be contained in picture stream snapshot extraction start ，I end ]At the same time its duration is not less than time threshold D t 。
The relation between the time period of the video frame target picture and the time period of the picture stream target picture to be detected can be seen by the following formula:
V start ≥I start +D t *0.5 formula (1)
V end +D t *0.5≤I end Formula (2)
Wherein V is start For video initiation time, V end For video end time, I start For the initial time of the picture, I end For the end time of the picture, D t Is a time threshold.
It should be understood that the coefficient "0.5" in the above formula (1) and formula (2) may be other different coefficients, for example, 0.5 in the formula (1) is replaced with 0.3, 0.5 in the formula (2) is replaced with 0.7, etc. However, the sum of the two coefficients in the formula (1) and the formula (2) is 1 or less.
It should be noted that the time threshold may be set by those skilled in the art according to need, and is not limited herein.
In some embodiments, performing similarity comparison on the video frame target picture and the to-be-detected picture stream target picture includes:
acquiring the time of a video frame target picture of the video frame target picture, and determining a picture preferable time period according to the time of the video frame target picture, wherein the picture preferable time period comprises a first preset time period before the time of the video frame target picture and/or a second preset time period after the time of the video frame target picture;
determining at least one picture stream target picture to be detected corresponding to the video frame target picture according to the picture optimal selection time period;
and carrying out similarity comparison on the video frame target picture and the picture stream target picture to be detected.
For example, when a single video stream extracts a video frame target picture and a picture stream snap shots extracts a to-be-detected picture stream target picture to perform similarity matching, a snap shot time T of the to-be-detected picture stream target picture can be selected i At video frame target picture snap time T v Front and rear time window [ W ] start ，W end ]Within the scope, i.e. there is the following relationship:
T v ＝(W start +W end )x 0.5；
or alternatively, the first and second heat exchangers may be,
W start ≤T v ≤W end ；W start ≤T i ≤W end 。
wherein W is start Is T v At some point in time before, W end Is T v Some point in time thereafter.
In some embodiments, if the detection parameters include at least one of video stream quality, picture stream modeling rate, target skip rate, etc., and each detection parameter is normal, then the image data quality is normal, otherwise, the image data quality is abnormal.
S103: and determining a detection result according to the quality of the video stream and the quality of the picture stream.
Alternatively, the detection result includes normal, or abnormal. The detection result also includes the condition of each detection information such as video stream quality, picture stream quality, and the like.
If the detection information includes at least one of a picture stream modeling rate, a target skip rate, etc. in addition to the video quality and the picture stream quality, the detection result is determined according to at least one of the video stream quality, the picture stream modeling rate, the target skip rate, etc.
Optionally, the detection result includes a specific value or level of each detection information.
In some embodiments, the image data quality detection tuning method further comprises:
and generating a detection report according to the detection result.
S104: and according to the detection result, optimizing the image equipment for shooting the image to be detected.
In some embodiments, optimizing an image device that captures an image to be detected includes:
and determining an abnormality reason according to the detection result, and adjusting the image equipment, wherein the adjustment comprises at least one of parameter adjustment and position adjustment.
If the detection result includes abnormality, that is, at least one detection parameter is abnormal, further analysis is needed to determine the cause of the abnormality and perform image equipment tuning.
Optionally, a mapping relationship or a database may be constructed in advance according to common abnormal conditions and abnormal reasons of each detection information and a common adjustment manner, so as to determine the abnormal reasons and adjust strategies in time when the quality of the image data is abnormal.
Alternatively, a training set may be constructed according to a preset plurality of sets of abnormal conditions, abnormal reasons and adjustment modes, and training may be performed through a neural network to obtain an adjustment policy model, and the abnormal reasons and adjustment policies may be determined based on the adjustment policy model according to the specific abnormal conditions of the current detection parameters.
Optionally, the method may also be based on a detection report, and in combination with an image to be detected, manually analyze an abnormality cause, input an adjustment instruction to the image device in a communication manner, and further control the image device to adjust, where the adjustment may be directly performed on a configuration parameter of the image device, or may be performed on an image device adjusting device capable of adjusting a physical position such as a position and an angle of the image device, so as to drive the position and the angle adjustment of the image device. That is, the image apparatus includes image apparatus adjusting means for adjusting the image apparatus by acquiring an adjustment instruction.
Optionally, the configuration parameters of the image equipment can be corrected remotely, and the position and the angle of the image equipment can be adjusted on site.
In some embodiments, the image to be detected may be analyzed first to obtain the quality of the video stream and the quality of the image stream, and then whether to obtain the modeling rate and the target missing rate of the image stream may be determined according to the quality of the video stream and the quality of the image stream.
If the quality of the image stream and the quality of the video stream of the new image to be detected are normal after the image equipment is optimized, the modeling rate of the image stream and the target skip rate are obtained. The picture stream modeling rate and the target missing rate may be obtained in any order, for example, the picture stream modeling rate is obtained first, the target missing rate is obtained only after the picture stream modeling rate is normal, or the target missing rate is obtained first, and the picture stream modeling rate is obtained only after the target missing rate is normal.
In some embodiments, the quality of the image stream and the quality of the video stream of the new image to be detected are normal, the modeling rate of the image stream and the target miss rate are obtained, and the image equipment is optimized by combining the abnormal conditions of the modeling rate of the image stream and the target miss rate and comprehensively judging.
In some embodiments, after tuning the image device capturing the image to be detected, the method further includes:
acquiring an adjusted image to be detected, which is shot by the adjusted image equipment;
analyzing the adjusted image to be detected, and determining the adjusted detection information and the adjusted detection result;
and determining an adjustment result according to the detection result and the adjusted detection result.
Optionally, the adjustment result includes a comparison of the detection results before and after adjustment, for example, some detection information is improved, some detection information is reduced, etc.
In some embodiments, the image data quality detection tuning method further comprises:
and determining the next adjustment mode of the image equipment according to the adjustment result.
Sometimes, the first tuning of the image device may not necessarily obtain a forward tuning result, that is, the quality of the image data of the image captured by the image device may not be improved after the tuning, and at this time, the next tuning mode may be formulated in combination with the last tuning mode.
And if the adjustment result meets the preset condition, stopping tuning, otherwise, tuning the image equipment again.
Wherein the preset conditions include, but are not limited to, each detection parameter meeting a preset pass threshold.
That is, after the adjustment of the image device is completed, the image data quality detection is performed again on the image captured by the adjusted image device, so as to obtain a detection report including detection parameters before and after the adjustment. Based on the two detection reports before and after adjustment, analysis and comparison of the image data quality before and after adjustment can be realized, and the adjustment effect of the image equipment is checked.
Optionally, the image data quality detection method provided by the embodiment of the invention can be applied to real-time detection and real-time adjustment of the image data quality of security image equipment.
According to the image data quality detection and adjustment method provided by the embodiment of the invention, the image to be detected is obtained by acquiring the image to be detected comprising the video stream and the image stream, the image to be detected is analyzed to determine detection information, the detection result is determined according to the image stream quality, the video stream quality and the like in the detection information, the image equipment for shooting the image to be detected is adjusted according to the detection result, the image data quality problem of the image data is possibly and accurately detected by comprehensively analyzing the video stream and the image stream, the data quality improvement requirement of large data image application is met, the data integrity, the function availability and the customer experience degree of an image equipment user are improved, and the image equipment can be adjusted in time when the image data quality is abnormal, so that the image data shot later by the image equipment meets the requirement.
Optionally, the image data quality detection method can also automatically analyze possible reasons for causing the data quality problem of the image equipment, improve the manual analysis efficiency and guide the manual tuning equipment.
Optionally, the image data quality detection method can automatically analyze and compare the image data quality conditions before and after the image equipment is tuned, and verify the tuning result of the image equipment.
The above-described image data quality detection method is further exemplified by a specific embodiment.
Referring to fig. 2, fig. 2 is a schematic diagram of a scheme framework of an image data quality detection method, the image device 201 collects images, at least a part of the images collected by the image device 201 is obtained as images to be detected, the images to be detected include video streams and picture streams, data extraction is performed on the images to be detected, picture stream targets, video frame targets and initial picture stream targets are extracted, statistical analysis is performed on the above data, statistical analysis methods include but are not limited to modeling success analysis, picture quality analysis target contrast analysis, video quality analysis and the like, detection parameters are determined, the detection parameters include at least one of video stream quality, picture stream modeling rate and target miss rate, a detection report (see the following table 1) is output, the detection report includes the above detection parameters, the image to be detected is combined according to the detection report, an abnormal cause is determined, the image device is adjusted, the adjustment of the image device can be performed by combining a remote adjustment device parameter, a site adjustment device position and an angle, and a new capture image is obtained after the new capture image is newly acquired from the image capture table after the new capture adjustment is completed (see the following table 1). By comparing the two detection reports in the tables 1 and 2, the target missing rate is obviously reduced under the condition that the picture stream quality, the video stream quality and the picture stream modeling rate are not obviously reduced, the amplitude reduction reaches about 30%, and the adjusting effect of the image equipment is obvious.
TABLE 1
TABLE 2
The image data quality detection method provided by the embodiment comprehensively analyzes the video stream quality and the picture stream quality of the image equipment to obtain a more accurate image data quality detection result, and also meets the data quality improvement requirement of large data image application.
Example two
Referring to fig. 3, an embodiment of the present invention provides an image data quality detection and optimization system 300, including:
the image acquisition module 301 is configured to acquire an image to be detected, where the image to be detected includes a video stream and a picture stream;
the analysis module 302 is configured to analyze an image to be detected, determine detection information, and the detection information includes video stream quality and picture stream quality;
and the tuning module 303 is configured to determine a detection result according to the video stream quality and the picture stream quality, and tune the image device capturing the image to be detected according to the detection result.
In this embodiment, the image data quality detection and optimization system executes the image data quality detection and optimization method described in any one of the above embodiments, and specific functions and technical effects may be referred to the above embodiments and are not described herein.
Referring to fig. 4, the embodiment of the present application further provides an apparatus 400, the apparatus 400 including a processor 401, a memory 402, and a communication bus 403;
the communication bus 403 is used to connect the processor 401 and the memory 402;
the processor 401 is configured to execute a computer program stored in the memory 402, so as to implement the image data quality detection tuning method according to any one of the embodiments.
The embodiment of the present application further provides a non-volatile readable storage medium, where one or more modules (programs) are stored, where the one or more modules are applied to a device, and the device may be caused to execute instructions (instructions) of a step included in the embodiment one of the embodiment of the present application.
The embodiment of the application also provides a computer-readable storage medium, on which a computer program is stored, the computer program being configured to cause the computer to execute the image data quality detection tuning method according to any one of the embodiments.
It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.
Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications and variations of the invention be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.
Claims (9)
1. An image data quality detection and optimization method, comprising the steps of:
acquiring an image to be detected, wherein the image to be detected comprises a video stream and a picture stream;
analyzing the image to be detected, and determining detection information, wherein the detection information comprises video stream quality, picture stream quality and target beat missing rate;
determining a detection result according to the video stream quality, the picture stream quality and the target skip rate, and optimizing an image device for shooting the image to be detected according to the detection result,
the method for determining the quality of the video stream comprises the steps of carrying out weighted average according to resolution and definition to obtain a comparison value of the quality of the video stream, and obtaining the quality of the video stream;
The method for determining the quality of the picture stream comprises the steps of determining according to the failure rate of the structural attribute of the picture stream target picture or the qualification rate of the structural attribute of the picture stream target picture;
the method for determining the target skip rate comprises the steps of determining a plurality of video frame target pictures from the video stream; obtaining video frame target picture structural information of each video frame target picture; comparing the video frame target picture structural information with the picture stream target picture structural information of the picture stream target picture to be detected to obtain the number of video frame target picture structural information failing to be compared; and determining the target skip rate according to the number of the video frame target picture structural information and the total number of the video frame target picture structural information which are failed to be compared.
2. The image data quality detection tuning method of claim 1, wherein the detection information further includes a picture stream modeling rate, and the determining of the picture stream modeling rate includes:
determining a plurality of initial picture stream target pictures from the picture stream;
obtaining picture stream target picture structuring information of each initial picture stream target picture, removing the initial picture stream target picture which does not comprise target attribute from the picture stream target picture structuring information, and taking the initial picture stream target picture which comprises the target attribute as a picture stream target picture to be detected;
And acquiring the initial picture stream target picture number and the picture stream target picture number to be detected, and determining the picture stream modeling rate.
3. The image data quality detection tuning method of claim 1, wherein comparing the video frame target picture and the picture stream target picture to be detected comprises:
acquiring the time of a video frame target picture, and determining a picture preferable time period according to the time of the video frame target picture, wherein the picture preferable time period comprises a first preset time period before the time of the video frame target picture and/or a second preset time period after the time of the video frame target picture;
determining the picture stream target picture to be detected corresponding to the video frame target picture according to the picture optimal selection time period;
and carrying out similarity comparison on the video frame target picture and the picture stream target picture to be detected.
4. The image data quality detecting and optimizing method according to claim 3, wherein,
the picture initial time is earlier than the video initial time, and the picture end time is later than the video end time;
the picture initial time is the earliest time in the time of the target picture of the picture stream to be detected, the picture end time is the latest time in the time of the target picture of the picture stream to be detected, the video initial time is the earliest time in the time of the target picture of the video frame, and the video end time is the latest time in the time of the target picture of the video frame.
5. The image data quality detection tuning method as claimed in any one of claims 1 to 4, wherein tuning an image device that captures the image to be detected comprises:
and determining an abnormality reason according to the detection result, and adjusting the image equipment, wherein the adjustment comprises at least one of parameter adjustment and position adjustment.
6. The image data quality detecting and optimizing method according to any one of claims 1 to 4, characterized by further comprising, after optimizing an image device that captures the image to be detected:
acquiring an adjusted image to be detected, which is shot by the adjusted image equipment;
analyzing the adjusted image to be detected, and determining the adjusted detection information and the adjusted detection result;
and determining an adjustment result according to the detection result and the adjusted detection result.
7. An image data quality detection tuning system, comprising:
the image acquisition module is used for acquiring an image to be detected, wherein the image to be detected comprises a video stream and a picture stream;
the analysis module is used for analyzing the image to be detected and determining detection information, wherein the detection information comprises video stream quality, picture stream quality and target skip rate;
The adjusting and optimizing module is used for determining a detection result according to the video stream quality, the picture stream quality and the target skip rate, and adjusting and optimizing image equipment for shooting the image to be detected according to the detection result; wherein, the liquid crystal display device comprises a liquid crystal display device,
the method for determining the quality of the video stream comprises the steps of carrying out weighted average according to resolution and definition to obtain a comparison value of the quality of the video stream, and obtaining the quality of the video stream;
the method for determining the quality of the picture stream comprises the steps of determining according to the failure rate of the structural attribute of the picture stream target picture or the qualification rate of the structural attribute of the picture stream target picture;
the method for determining the target skip rate comprises the steps of determining a plurality of video frame target pictures from the video stream; obtaining video frame target picture structural information of each video frame target picture; comparing the video frame target picture structural information with the picture stream target picture structural information of the picture stream target picture to be detected to obtain the number of video frame target picture structural information failing to be compared; and determining the target skip rate according to the number of the video frame target picture structural information and the total number of the video frame target picture structural information which are failed to be compared.
8. An electronic device comprising a processor, a memory, and a communication bus;
the communication bus is used for connecting the processor and the memory;
the processor is configured to execute a computer program stored in the memory to implement the image data quality detection tuning method according to any one of claims 1 to 6.
9. A computer-readable storage medium, having stored thereon a computer program for causing the computer to execute the image data quality detection tuning method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110502065.6A CN113191286B (en) | 2021-05-08 | 2021-05-08 | Image data quality detection and optimization method, system, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110502065.6A CN113191286B (en) | 2021-05-08 | 2021-05-08 | Image data quality detection and optimization method, system, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113191286A CN113191286A (en) | 2021-07-30 |
| CN113191286B true CN113191286B (en) | 2023-04-25 |
Family
ID=76988685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110502065.6A Active CN113191286B (en) | 2021-05-08 | 2021-05-08 | Image data quality detection and optimization method, system, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113191286B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015044947A1 (en) * | 2013-09-30 | 2015-04-02 | Yanai Danielle | Image and video processing and optimization |
| CN110348378A (en) * | 2019-07-10 | 2019-10-18 | 北京旷视科技有限公司 | A kind of authentication method, device and storage medium |
| WO2019232972A1 (en) * | 2018-06-04 | 2019-12-12 | 上海商汤智能科技有限公司 | Driving management method and system, vehicle-mounted intelligent system, electronic device and medium |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2347599B1 (en) * | 2008-11-04 | 2015-02-25 | Telefonaktiebolaget L M Ericsson (publ) | Method and system for determining a quality value of a video stream |
| CN106911901A (en) * | 2017-01-11 | 2017-06-30 | 北京光年无限科技有限公司 | A kind of data processing method and system for intelligent robot |
| CN107172361B (en) * | 2017-07-12 | 2019-11-15 | 维沃移动通信有限公司 | A kind of method and mobile terminal of pan-shot |
| CN108346149B (en) * | 2018-03-02 | 2021-03-12 | 北京郁金香伙伴科技有限公司 | Image detection and processing method and device and terminal |
| CN108764091B (en) * | 2018-05-18 | 2020-11-17 | 北京市商汤科技开发有限公司 | Living body detection method and apparatus, electronic device, and storage medium |
| CN108391077A (en) * | 2018-05-22 | 2018-08-10 | 广州长鹏光电科技有限公司 | A kind of simple Classic Course video automatic shooting record device |
| CN109190469B (en) * | 2018-07-27 | 2020-06-23 | 阿里巴巴集团控股有限公司 | Detection method and device, computing equipment and storage medium |
| CN109949414A (en) * | 2019-01-31 | 2019-06-28 | 顺丰科技有限公司 | The construction method and device of indoor map |
| CN111008954A (en) * | 2019-11-04 | 2020-04-14 | 联想(北京)有限公司 | Information processing method and device, electronic equipment and storage medium |
| CN110910628B (en) * | 2019-12-02 | 2021-02-12 | 支付宝(杭州)信息技术有限公司 | Interactive processing method and device for vehicle damage image shooting and electronic equipment |
| CN111402297A (en) * | 2020-03-25 | 2020-07-10 | 杭州博雅鸿图视频技术有限公司 | Target tracking detection method, system, electronic device and storage medium |
| CN112257638A (en) * | 2020-10-30 | 2021-01-22 | 重庆紫光华山智安科技有限公司 | Image comparison method, system, equipment and computer readable storage medium |
| CN112492297B (en) * | 2020-11-17 | 2022-06-14 | 腾讯科技（深圳）有限公司 | Video processing method and related equipment |
-
2021
- 2021-05-08 CN CN202110502065.6A patent/CN113191286B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015044947A1 (en) * | 2013-09-30 | 2015-04-02 | Yanai Danielle | Image and video processing and optimization |
| WO2019232972A1 (en) * | 2018-06-04 | 2019-12-12 | 上海商汤智能科技有限公司 | Driving management method and system, vehicle-mounted intelligent system, electronic device and medium |
| CN110348378A (en) * | 2019-07-10 | 2019-10-18 | 北京旷视科技有限公司 | A kind of authentication method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113191286A (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI709091B (en) | Image processing method and device | |
| WO2020094091A1 (en) | Image capturing method, monitoring camera, and monitoring system | |
| CN110298231B (en) | Method and system for judging goal of basketball game video | |
| CN109299703B (en) | Method and device for carrying out statistics on mouse conditions and image acquisition equipment | |
| US9596520B2 (en) | Method and system for pushing information to a client | |
| CN103839308A (en) | Population obtaining method, device and system | |
| WO2021031954A1 (en) | Object quantity determination method and apparatus, and storage medium and electronic device | |
| US20150104067A1 (en) | Method and apparatus for tracking object, and method for selecting tracking feature | |
| CN109922311B (en) | Monitoring method, device, terminal and storage medium based on audio and video linkage | |
| CN111464819A (en) | Live image detection method, device, equipment and storage medium | |
| CN112686165A (en) | Method and device for identifying target object in video, electronic equipment and storage medium | |
| CN111586432B (en) | Method and device for determining air-broadcast live broadcast room, server and storage medium | |
| CN111191506A (en) | Personnel flow statistical method and device, computer equipment and storage medium | |
| CN111881320A (en) | Video query method, device, equipment and readable storage medium | |
| US20190027004A1 (en) | Method for performing multi-camera automatic patrol control with aid of statistics data in a surveillance system, and associated apparatus | |
| US20090310855A1 (en) | Event detection method and video surveillance system using said method | |
| CN113191286B (en) | Image data quality detection and optimization method, system, equipment and medium | |
| CN113158773B (en) | Training method and training device for living body detection model | |
| US9286664B2 (en) | System and method for blind image deconvolution | |
| CN108960165A (en) | A kind of stadiums population surveillance method based on intelligent video identification technology | |
| CN112837640A (en) | Screen dynamic picture testing method, system, electronic equipment and storage medium | |
| CN114565952A (en) | Pedestrian trajectory generation method, device, equipment and storage medium | |
| CN112257545A (en) | Violation real-time monitoring and analyzing method and device and storage medium | |
| Zheng et al. | No-reference quality assessment of variable frame-rate videos using temporal bandpass statistics | |
| US10283166B2 (en) | Video indexing method and device using the same |
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 |