CN115311607A - Epoxy resin production monitoring data management method - Google Patents
Epoxy resin production monitoring data management method Download PDFInfo
- Publication number
- CN115311607A CN115311607A CN202211231069.6A CN202211231069A CN115311607A CN 115311607 A CN115311607 A CN 115311607A CN 202211231069 A CN202211231069 A CN 202211231069A CN 115311607 A CN115311607 A CN 115311607A
- Authority
- CN
- China
- Prior art keywords
- frame
- dynamic
- frames
- moving object
- random
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 92
- 239000003822 epoxy resin Substances 0.000 title claims abstract description 49
- 229920000647 polyepoxide Polymers 0.000 title claims abstract description 48
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013523 data management Methods 0.000 title claims description 9
- 230000000149 penetrating effect Effects 0.000 claims abstract description 23
- 239000011159 matrix material Substances 0.000 claims abstract description 15
- 238000000638 solvent extraction Methods 0.000 claims abstract description 7
- 238000007726 management method Methods 0.000 claims abstract description 5
- 230000033001 locomotion Effects 0.000 claims description 56
- 239000004593 Epoxy Substances 0.000 claims description 5
- 238000013139 quantization Methods 0.000 description 18
- 238000004364 calculation method Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 230000003068 static effect Effects 0.000 description 7
- 238000005192 partition Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229960001948 caffeine Drugs 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 229920005989 resin Polymers 0.000 description 1
- 239000011347 resin Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 229920001187 thermosetting polymer Polymers 0.000 description 1
- RYYVLZVUVIJVGH-UHFFFAOYSA-N trimethylxanthine Natural products CN1C(=O)N(C)C(=O)C2=C1N=CN2C RYYVLZVUVIJVGH-UHFFFAOYSA-N 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/08—Key distribution or management, e.g. generation, sharing or updating, of cryptographic keys or passwords
- H04L9/0861—Generation of secret information including derivation or calculation of cryptographic keys or passwords
- H04L9/0863—Generation of secret information including derivation or calculation of cryptographic keys or passwords involving passwords or one-time passwords
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to the technical field of production data processing, in particular to a management method of epoxy resin production monitoring data. The method comprises the following steps: obtaining dynamic frames in all video frames; calculating the similarity of two moving objects in the two types of dynamic frames to obtain the moving objects belonging to the same moving object in the various types of dynamic frames, and recording as penetrating moving objects; obtaining a random frame, and partitioning the monitoring video data by using the random frame to obtain monitoring video data of different intervals; selecting a moving object with a random path and the largest average characteristic value from the monitoring video data in an interval, and recording the moving object as a key generation moving object; generating a key matrix by using all pixel points in a mark connected domain in an image corresponding to a first frame in all random frames corresponding to a moving object in monitoring video data by using a key; and encrypting the monitoring video data of each interval by using the key matrix corresponding to the monitoring video data of each interval. The invention can improve the safety of monitoring video data.
Description
Technical Field
The invention relates to the technical field of production data processing, in particular to a management method of epoxy resin production monitoring data.
Background
Epoxy resin is an important thermosetting resin. Has good chemical properties and physical properties, and can be widely applied to various industries. In the process of epoxy resin production, wherein production monitoring data, such as production monitoring videos, contain private information of enterprises, such as unique formulas and operation practices of epoxy resin production, information leakage in the monitoring videos can cause relatively large influence on production data of the enterprises, so that encryption processing is required.
The conventional encryption method for the monitoring video is to use the existing algorithm to generate a key to encrypt the whole video, and the encryption mode is that the whole key is commonly used, and the algorithm of the generated key is not truly random, so that the generated key is easily cracked violently when being invaded by the outside, thereby causing the data loss of the monitoring video in the production process of the epoxy resin and the leakage of production data of enterprises.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a method for managing monitoring data of epoxy resin production, which adopts the following technical scheme:
one embodiment of the invention provides a management method for monitoring data in epoxy resin production, which comprises the following steps: collecting monitoring video data in the production process of epoxy resin; obtaining the information content of each video frame in the monitoring video data; obtaining the information quantity difference value of every two adjacent video frames, and judging the dynamic frame in all the video frames based on the information quantity difference value, wherein the dynamic frame refers to the video frame with object motion in the monitoring video data;
dividing dynamic frames which are continuous frames into one type, and obtaining a mark connected domain in an image corresponding to the dynamic frame in each type of dynamic frame by using a frame difference method; calculating to obtain the object characteristics of the marked connected domain based on a two-bit data set consisting of gray values of pixel points in the marked connected domain and pixel points in the neighborhood in the image corresponding to each dynamic frame; obtaining the motion trail of the same moving object in each type of dynamic frame based on the object characteristics of the marked connected domain; obtaining the average characteristic value of each moving object in each type of dynamic frame, calculating the similarity of two moving objects in the two types of dynamic frames, obtaining the moving objects belonging to the same moving object in the various types of dynamic frames, and recording as penetrating moving objects;
classifying the motion tracks of the penetrating motion objects in each type of dynamic frame to obtain a random path penetrating the motion objects, wherein the dynamic frame corresponding to the random path is a random frame; partitioning the monitoring video data by using random frames to obtain monitoring video data of different intervals; selecting a moving object with a random path and the largest average characteristic value from the monitoring video data in an interval, and recording the moving object as a key generation moving object; generating a key matrix by using all pixel points in a mark connected domain in an image corresponding to a first frame in all random frames corresponding to a moving object in monitoring video data by using a key; and encrypting the monitoring video data of each interval by using the key matrix corresponding to the monitoring video data of each interval.
Preferably, the amount of information of each video frame is:
wherein,
an information amount indicating an nth frame video frame;
the gray value of a pixel point is represented,
represents a gray value of
The pixel point of is at
Probability of occurrence in the image corresponding to the frame video frame.
Preferably, obtaining an information amount difference value between every two adjacent video frames, and determining a dynamic frame in all the video frames based on the information amount difference value comprises: and if the difference value of the information quantity of every two adjacent video frames is not equal to 0, the video frame of the next frame in every two adjacent video frames is a dynamic frame.
Preferably, the object characteristic of the labeled connected components is:
wherein,
representing the object characteristics of the m-th mark connected domain in the n-th frame dynamic frame;
represents the first in the class
Is marking the first in the connected domain
The number of the pixel points is one,
the number of all pixel points in the mark connected domain;
represents the first in the class
Is marking the first in the connected domain
The surrounding 8 neighborhood pixels of a pixel are,
is shown as
The gray value and the second of each pixel point
The second data group of the average value of the gray values of 8 pixel points in the neighborhood around each pixel point is
The probability of occurrence in all the two-bit data sets in the connected component field is marked.
Preferably, obtaining an average feature value of each moving object in each type of dynamic frame, calculating the similarity of two moving objects in the two types of dynamic frames, and obtaining the moving objects belonging to the same moving object in the various types of dynamic frames includes:
the average characteristic value is the average value of the object characteristics of all corresponding mark connected domains of a moving object in a class of dynamic frames; the ratio of the average characteristic values of the two moving objects in different types of dynamic frames is the similarity of the two moving objects in the two types of dynamic frames, and if the similarity of the two moving objects in the two types of dynamic frames is larger than a preset threshold value, the two moving objects in the two types of dynamic frames are the same moving object.
Preferably, classifying the motion trajectory of each type of dynamic frame penetrating the moving object to obtain a random path penetrating the moving object, where the dynamic frame corresponding to the random path is a random frame and includes: and analyzing the motion track penetrating through the moving object in each type of dynamic frame by using an LOF algorithm to obtain outliers, wherein the path corresponding to the outliers is a random path.
Preferably, generating the key matrix comprises: each element in the key matrix is used as a key to generate a gray value of a pixel point in a mark connected domain of a moving object in an image corresponding to a first frame in all random frames corresponding to the monitored video data.
The embodiment of the invention at least has the following beneficial effects: the invention relates to a method for encrypting a surveillance video by using a conventional video encryption algorithm, which comprises the steps of generating pseudo-random data by using a corresponding algorithm, utilizing epoxy resin to produce a moving object which moves randomly in the surveillance video data to quantize and set a secret key for encryption, wherein the action of producing the moving object which moves randomly in the surveillance video data moving object by using the epoxy resin is influenced by external factors, the moving object does not have regularity and is random in a real sense, so that the surveillance video data produced by using the epoxy resin is continuously encrypted by using the characteristic, the randomness of the secret key is realized in the real sense, and the secret key is different and the encryption result is different due to the different random dynamic frames selected by each section of surveillance video.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for managing monitoring data of epoxy resin production according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention for achieving the predetermined objects, the following detailed description of the method for managing the monitoring data of the epoxy resin production according to the present invention, the specific implementation, structure, features and effects thereof will be provided in conjunction with the accompanying drawings and the preferred embodiments. In the following description, the different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the epoxy resin production monitoring data management method provided by the invention in detail with reference to the accompanying drawings.
Example (b):
the main application scenarios of the invention are as follows: in the process of epoxy resin production, a monitoring video records a plurality of private enterprise production data, so that encryption is needed to be carried out on the monitoring video.
The main purposes of the invention are: the method comprises the steps of extracting dynamic frames by using an epoxy resin monitoring video, identifying moving objects in all the dynamic frames, analyzing different moving tracks of all the moving objects in different dynamic frames, selecting a random dynamic frame where the random moving track is located to partition the whole monitoring video, calculating a key of a current interval for the video of each interval according to carrying information of the moving objects, and finally encrypting the video of the current interval through the key and completing transmission and storage to realize management of epoxy resin production monitoring data.
Referring to fig. 1, a flowchart of a method for managing monitoring data of epoxy resin production according to an embodiment of the present invention is shown, where the method includes the following steps:
s1, collecting monitoring video data in the production process of epoxy resin; obtaining the information content of each video frame in the monitoring video data; and obtaining the information quantity difference value of every two adjacent video frames, and judging the dynamic frame in all the video frames based on the information quantity difference value, wherein the dynamic frame refers to the video frame with object motion in the monitored video data.
The method needs to encrypt the monitoring video data of epoxy resin production, so the monitoring video data of epoxy resin production needs to be collected, the specific collection process is to collect the monitoring video data of epoxy resin production by using a monitoring camera in an epoxy resin production workshop, and the specific arrangement position method of the monitoring camera is arranged by an implementer.
Further, the difference value of the information carrying amount of the image of each frame in the epoxy resin production monitoring video data is needed to be used for extracting the dynamic frame.
The monitoring video of the epoxy resin is obtained, and the monitoring video is divided into static frames and dynamic frames due to the particularity (scene invariance), wherein the static frames are video frames without object motion in the monitoring video, and the dynamic frames are video frames with object motion in the monitoring video. The information carrying amount in all the static frames is the same, and the information carrying amount of the video in the dynamic frame is different, so that the static frames and the dynamic frames of the monitoring video are divided according to the characteristic, and the dynamic frame video is extracted, which is specifically shown as follows:
to a first order
For example, the method for determining whether the frame video is a dynamic frame is as follows:
firstly, calculating the information carrying capacity of the first frame video
：
In the formula:
an amount of information representing the video frame of the nth frame,
expressing the gray value of the pixel point, ()
），
Represents a gray value of
The pixel point of is at
The probability of occurrence in the frame image is calculated as a gray value
Is at the first of the pixel points
The frequency of occurrence in the frame image is divided by the number of pixels of the overall image.
Formula logic: first, the
The frame image is a static picture, the carried information is expressed by the gray values of different pixel points as visual information, and the carried information is constant because the frame image is a static picture, namely the distribution of the expressed gray values of the pixel points is constant, so the carried information is quantized by utilizing the calculation mode of the information entropy.
Then calculate the
Information carrying amount of image (A)
Calculation method and
information carrying amount of frame image
The same way of calculation.
Then calculate the first
Frame image and
the difference value of the information carrying amount of the frame image is used for judging the first time
Whether the frame image is a moving image, the first
Frame image and
difference value of information carrying amount of frame image
The calculation is as follows:
formula logic: to a first order
Information carrying amount of frame image
And a first
The difference value of the information carrying amount of the frame image is used for representing the second
Whether the frame image is a dynamic frame or not is judged, wherein the dynamic frame refers to that dynamic movement of different objects appears in two continuous frames of images in the production monitoring video of the epoxy resin, namely that the dynamic in the dynamic frame is relative to the previous frame of image. The surveillance video has particularity in all videos, because the angle of the surveillance camera is constant due to the fixation of the surveillance camera, and the video of each frame shot by the surveillance camera is the same when no dynamic object exists, the difference value in the videos is the same
If it is zero, it indicates
Frame image compared with the first
The frame images do not have any difference, i.e. in production monitoring video of epoxy resin
If no dynamic object appears in the frame image, even if the dynamic object appears, the frame image is the same object, because the frame image changes in the angle of the monitoring camera after moving, the gray value of the frame image changes certainly according to the optical principle, the gray value of the frame image is expressed in the above formula, and the difference value of the gray value of the frame image is certainly not 0, so that the frame image is judged to be the first frame image according to the above mode
Whether the frame image is a dynamic frame.
Then explain the
The frame image is a still frame image,
then explain the
The frame image is a dynamic frame image.
To this end, the
And finishing the judgment of the dynamic frame and the static frame of the frame image. By utilizing the method, the videos of all the frames shot in the epoxy resin monitoring video are judged, whether the monitoring videos of the epoxy resin of all the frames are dynamic frames or not can be obtained, and then the dynamic frames are marked and extracted. And at this point, extracting dynamic frames of all the epoxy resin monitoring videos.
S2, dividing the dynamic frames which are continuous frames into one type, and obtaining a mark connected domain in an image corresponding to the dynamic frame in each type of dynamic frame by using a frame difference method; calculating to obtain the object characteristics of the marked connected domain based on a two-bit data set consisting of gray values of pixel points in the marked connected domain and pixel points in the neighborhood in the image corresponding to each dynamic frame; obtaining the motion trail of the same moving object in each type of dynamic frame based on the object characteristics of the marked connected domain; and obtaining the average characteristic value of each moving object in each type of dynamic frame, calculating the similarity of the two moving objects in the two types of dynamic frames, obtaining the moving objects belonging to the same moving object in the various types of dynamic frames, and recording as penetrating moving objects.
The dynamic frames in all the epoxy resin production monitoring videos are obtained in the step S1, random dynamic frame extraction is performed on all dynamic frame images, then interval division is performed on the whole epoxy resin production monitoring videos by using the random dynamic frames, and the encryption key of the epoxy resin monitoring videos in the corresponding interval is calculated by using the characteristic parameters of the random dynamic frames in each interval.
The purpose is as follows: in the monitoring video of epoxy resin, the dynamic objects appear in two states, namely, mechanical repeated motion (motion of a conveyor belt, motion of workers on and off duty and the like) and non-mechanical motion of randomly appearing objects (a certain staff or leader performs workshop inspection, abnormal motion of the conveyor belt and the like). The latter state is unexpected factor in the production process of epoxy, uncontrollable, it is totally random relatively with holistic surveillance video, so utilize its characteristic to carry out encryption to epoxy's production surveillance video, what its real sense was accomplished is random of key, and because the difference of the random dynamic frame that every section surveillance video selected, its key also just also is different, the encryption result also is different, compare in current pseudo-random encryption and whole encryption, its security is higher, be difficult to be cracked more.
Quantizing the characteristics and motion tracks of the moving object by using the dynamic frame image; the purpose of the feature quantization of the moving object in the dynamic frame image is to determine whether the motion within the continuous dynamic frames is the same object, and the purpose of the quantization of the motion trajectory of the moving object in the dynamic frame is to determine the motion trajectory of the same moving object.
All the dynamic frames are obtained, and the motion trail of the same moving object is quantized. The specific process is to cluster the objects according to whether the objects are continuous frames or not (the motion representation of the objects is continuous in the monitored video, so the objects are clustered according to the characteristics, for example, the 1,2,3,7,8,9 and the 10 th frames are continuous frames, and the clustering result is 1,2,3 frames, 7,8,9 and the 10 th frames are classified). Then, each frame image in each category is quantized with the moving object characteristics and the moving characteristics, taking continuous dynamic frame images of any category as an example.
The quantization of its features is as follows: firstly, detecting all moving object boundaries in two continuous frame images by using a frame difference method for the dynamic frame images of two adjacent frames, and marking the detected moving object boundaries on the two continuous frame images to obtain a marked connected domain.
Then, the object characteristics in each connected domain in each frame image are quantified, and the first one in a certain class is used
In the frame image
An example of a connected domain, its object characteristics
The specific calculation method is as follows:
in the formula,
representing the object characteristics of the m-th mark connected domain in the n-th frame dynamic frame;
represents the first in the class
Is marking the first in the connected domain
A pixel point (
In which
The number of all the pixels in the labeled connected domain),
represents the first in the class
Is marking the first in the connected domain
The surrounding 8 neighborhood pixels of a pixel are,
is shown as
Gray value and the second of each pixel point
The probability of the two-bit data set of the average value of the gray values of 8 pixels in the neighborhood around each pixel in all the two-bit data sets in the first mark connected domain is calculated in the first way
The gray value and the second of each pixel point
The two-bit data of the average value of the gray values of 8 pixel points in the surrounding neighborhood of each pixel point is in the whole
The frequency count occurring in each connected field is divided by the number of all the two-bit data sets.
Formula logic: because only the boundary of a moving object between two continuous dynamic frames is detected by the frame difference method, when a plurality of same objects move simultaneously, the corresponding objects cannot be detected. So the quantization is performed in the above-mentioned manner
The characteristics of the object in each connected domain not only consider the information carried by the gray value of each pixel point in the connected domain, but also consider the information carried by the gray value of the pixel points in 8 neighborhoods around each pixel point and the pixel point to be integrally distributed, and the characteristics of the moving object in the connected domain are quantified in the spatial distribution mode.
In the above manner to
Frame dynamic frame and second
All connected domains in the frame dynamic frame are subjected to characteristic quantization of moving objects, and then the second step is used
Matching the characteristic values of all connected domains in the frame dynamic frame image to calculate the matching degree
Frame number
A connected domain and a
In the frame
Example of connected component, degree of matching of connected component
The calculation is as follows:
in the formula:
is as follows
In frame dynamic frame
The quantization characteristic values of the moving objects of the connected domain,
is as follows
In frame dynamic frame
Quantitative characteristic values of the moving objects of the connected domains.
Formula logic: in order to effectively prevent the problem that correct matching cannot be carried out under the condition that the connected domains of a plurality of moving objects are relatively the same (by utilizing the connected domains of a plurality of moving people, the connected domains are nearly the same), the matching is carried out by utilizing the difference value of the space distribution characteristics of moving objects of two connected domains in two continuous frames.
Then setting a matching threshold
For two connected domains corresponding to matching values greater than the threshold, the two connected domains are considered as the same moving object (one empirical matching threshold is
）。
By using the above method to match different connected domains in all the frame dynamic frames in the category by using the characteristic values, the distribution of the same moving object in all the dynamic frame images in the category can be obtained.
Then, the moving track of each moving object is quantified to obtain a certain moving object in any one class
For example, the motion track
The quantization is as follows:
locating the corresponding connected domain appearing in each frame in the category, and then obtaining the centroid coordinates of the connected domain corresponding to the moving object in each frame by utilizing the prior art, wherein the centroid coordinates are respectively as follows:
the coordinates of the centroid position are the motion track of the moving object
。
The motion tracks of all the moving objects can be obtained by quantifying the tracks of all the moving objects in the above mode.
At this point, the motion trail of the same object in all the dynamic frames and the feature quantization of the moving object are completed.
Selecting a random dynamic frame through the characteristics and the motion track of a quantized moving object and partitioning a monitoring video by utilizing the random dynamic frame; the specific logic is that similarity calculation is carried out on all moving objects in all epoxy resin production monitoring videos through the quantization characteristics and quantization tracks of the moving objects, and the continuous dynamic frames where the moving objects which appear for many times but have different motion tracks are located are screened out and considered as the random dynamic frames. And then partitioning all the whole monitoring videos by using the random dynamic frame. The specific method is as follows:
firstly, carrying out overall moving object identification on all the moving objects identified in the above step according to the characteristic quantization values of the moving objects, and judging whether the moving objects of different categories are the same moving object or not, wherein the identification mode is as follows:
since the feature quantization values of each moving object in each category dynamic frame are slightly different due to the motion, the average value of the feature quantization values of each moving object in each category dynamic frame is calculated first to reduce the influence of the difference, and then
Sports articles in a category
For example, the specific manner is as follows:
in the formula,
representing moving objects
In the first place
The average characteristic value in each of the classifications is,
representing moving objects
In the first place
In successive frames of a classification
The characteristic quantization values calculated for the corresponding connected components of the frame,
representing moving objects
In the first place
Total number of consecutive total frames present in each category.
By performing the average feature value calculation for all the different categories of moving objects in the above manner, the average feature value of each object in each category can be obtained.
Then, the similarity calculation of the average characteristic value is utilized for the articles in each category, whether the moving articles in all the categories are the same moving article is determined, and the second step is to calculate the similarity of the average characteristic values
Moving objects in a category
And a first
Moving objects in a category
For example, the similarity is calculated as follows:
in the formula,
is shown as
Moving objects in a category
Is determined by the average characteristic value of (a),
is shown as
Moving objects in a category
Average eigenvalues of (d).
All the above-mentioned modes are used
And performing similarity calculation on all the moving objects in each category, then identifying the moving objects in all the categories by using the similarity, identifying the same moving object in the moving objects in all the categories, and marking as a penetrating moving object.
S3, classifying the motion tracks of the penetrating motion object in each type of dynamic frame to obtain a random path penetrating the motion object, wherein the dynamic frame corresponding to the random path is a random frame; partitioning the monitoring video data by using random frames to obtain monitoring video data of different intervals; selecting a moving object with a random path and the largest average characteristic value from the monitoring video data in an interval, and recording the moving object as a key generation moving object; generating a key matrix by using all pixel points in a mark connected domain in an image corresponding to a first frame in all random frames corresponding to a moving object in monitoring video data by using a key; and encrypting the monitoring video data of each interval by using the key matrix corresponding to the monitoring video data of each interval.
Then classifying the motion trails of each penetrating motion object in different categories to find out different motion trails, wherein most operations and courses in the production process are repeated mechanical motions, so that the theoretical motion trails of the same object are the same, and each penetrating motion object changes in the motion trails due to the influence of various external factors in the motion process to generate random behaviors, so that the random behaviors are detected by using the motion trails to detect the motion trails so as to find out the different motion trails of the moving objects
For example, its random behavior is detected by first detecting what it appears in each categoryAnd extracting the motion track quantization value.
And then, analyzing the motion track quantization values which appear in all the categories of the penetrating motion object by utilizing LOF algorithm model outliers, and selecting each outlier, wherein the corresponding path is a random path.
All the penetrating moving objects are subjected to random behavior path acquisition in the mode, and the frames of the random paths are random dynamic frames which can be acquired
A random frame. (a plurality of random paths penetrating the moving object are in the same frame and are calculated once), then the whole monitoring video is segmented by using the random frames, and the classification process is as follows:
first, the position of the random frame (which is also a continuous frame because the path is obtained by continuous frame analysis) in the whole epoxy resin surveillance video is retrieved; and then segmenting the whole epoxy resin monitoring video by using random frames. By this, the data partitioning ends.
And calculating an encryption key of each interval according to the random dynamic frame of each interval and carrying out interval encryption by using the encryption key. In the above, the partition of the whole monitoring video is obtained by using the random frame, the encryption key is calculated for each interval and the encryption key is used for the interval encryption, so as to
Partition of a surveillance video, for example, its key
The calculation of (c) is as follows:
firstly, searching the average value of the characteristic quantization values of the moving objects with random paths in the random frame in the interval, and selecting the moving object corresponding to the maximum average characteristic quantization value
And is recorded as a key generation moving object.
Then using the key to generate the moving object
All of the connected components (obtained by the mid-frame difference method) present in the first frame of the random frames
Each pixel point becomes a key matrix
Each element of the key matrix being this
The gray value of each pixel point and the size of the key matrix are
Wherein
And
are respectively as
The largest two prime factors of (a).
Then use
To the first
Encrypting each frame of video of each surveillance video partition to obtain encrypted ciphertext
The encryption mode is to use
And a first
Each frame of each surveillance video partition is subjected to convolution operation.
By utilizing the method, each frame of image in the epoxy resin monitoring video is encrypted, and the encrypted data of the epoxy resin monitoring video after being encrypted integrally can be obtained. At this point, the encryption of the epoxy monitoring video is completed. And obtaining encrypted data of the epoxy resin monitoring video data, transmitting the transmission value to a monitoring video storage terminal, and storing.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A management method for monitoring data of epoxy resin production is characterized by comprising the following steps:
collecting monitoring video data in the production process of epoxy resin; obtaining the information content of each video frame in the monitoring video data; obtaining the information quantity difference value of every two adjacent video frames, and judging the dynamic frame in all the video frames based on the information quantity difference value, wherein the dynamic frame refers to the video frame with object motion in the monitoring video data;
dividing dynamic frames which are continuous frames into one type, and obtaining a mark connected domain in an image corresponding to the dynamic frame in each type of dynamic frame by using a frame difference method; calculating to obtain the object characteristics of the marked connected domain based on a two-bit data set consisting of gray values of pixel points in the marked connected domain and pixel points in the neighborhood in the image corresponding to each dynamic frame; obtaining the motion trail of the same moving object in each type of dynamic frame based on the object characteristics of the marked connected domain; obtaining the average characteristic value of each moving object in each type of dynamic frame, calculating the similarity of two moving objects in the two types of dynamic frames, obtaining the moving objects belonging to the same moving object in the various types of dynamic frames, and recording as penetrating moving objects;
classifying the motion tracks of the penetrating motion objects in each type of dynamic frame to obtain a random path penetrating the motion objects, wherein the dynamic frame corresponding to the random path is a random frame; partitioning the monitoring video data by using random frames to obtain monitoring video data of different intervals; selecting a moving object with a random path and the largest average characteristic value from the monitoring video data in an interval, and recording the moving object as a key generation moving object; generating a key matrix by using all pixel points in a mark connected domain in an image corresponding to a first frame in all random frames corresponding to a moving object in monitoring video data by using a key; and encrypting the monitoring video data of each interval by using the key matrix corresponding to the monitoring video data of each interval.
2. The epoxy resin production monitoring data management method according to claim 1, wherein the information amount of each video frame is:
wherein,
an information amount indicating an nth frame video frame;
the gray value of a pixel point is represented,
represents a gray value of
The pixel point of is at
Probability of occurrence in the image corresponding to the frame video frame.
3. The epoxy resin production monitoring data management method according to claim 1, wherein the obtaining the information amount difference value of every two adjacent video frames and the determining the dynamic frames in all the video frames based on the information amount difference value comprises: and if the difference value of the information quantity of every two adjacent video frames is not equal to 0, the video frame of the next frame in every two adjacent video frames is a dynamic frame.
4. The epoxy production monitoring data management method of claim 1, wherein the object characteristic of the tag connected domain is:
wherein,
representing the object characteristics of the m-th mark connected domain in the n-th frame dynamic frame;
represents the first in the class
Is marking the first in the connected domain
The number of the pixel points is one,
the number of all pixel points in the mark connected domain;
represents the first in the class
Is marking the first in the connected domain
The surrounding 8 neighborhood pixels of a pixel are,
is shown as
The gray value and the second of each pixel point
The second bit data group of the average value of the gray values of 8 pixel points in the neighborhood around each pixel point
The probability of occurrence in all the two-bit data sets in the connected component field is marked.
5. The epoxy resin production monitoring data management method according to claim 1, wherein the obtaining of the average feature value of each moving object in each type of dynamic frame, the calculating of the similarity of two moving objects in two types of dynamic frames, and the obtaining of moving objects belonging to the same moving object in various types of dynamic frames comprises:
the average characteristic value is the average value of the object characteristics of all corresponding mark connected domains of a moving object in a class of dynamic frames; the ratio of the average characteristic values of the two moving objects in different types of dynamic frames is the similarity of the two moving objects in the two types of dynamic frames, and if the similarity of the two moving objects in the two types of dynamic frames is larger than a preset threshold value, the two moving objects in the two types of dynamic frames are the same moving object.
6. The epoxy resin production monitoring data management method according to claim 1, wherein the classifying the motion trajectories of the motion penetrating objects in each type of dynamic frame to obtain a random path penetrating the motion penetrating objects, and the dynamic frame corresponding to the random path is a random frame including: and analyzing the motion track penetrating through the moving object in each type of dynamic frame by using an LOF algorithm to obtain outliers, wherein the path corresponding to the outliers is a random path.
7. The epoxy production monitoring data management method of claim 1, wherein the generating a key matrix comprises: each element in the key matrix is used as a key to generate a gray value of a pixel point in a mark connected domain of a moving object in an image corresponding to a first frame in all random frames corresponding to the monitored video data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211231069.6A CN115311607B (en) | 2022-10-10 | 2022-10-10 | Epoxy resin production monitoring data management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211231069.6A CN115311607B (en) | 2022-10-10 | 2022-10-10 | Epoxy resin production monitoring data management method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115311607A true CN115311607A (en) | 2022-11-08 |
| CN115311607B CN115311607B (en) | 2023-04-14 |
Family
ID=83866678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211231069.6A Active CN115311607B (en) | 2022-10-10 | 2022-10-10 | Epoxy resin production monitoring data management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115311607B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801062A (en) * | 2021-04-07 | 2021-05-14 | 平安科技（深圳）有限公司 | Live video identification method, device, equipment and medium |
| CN114926317A (en) * | 2022-07-20 | 2022-08-19 | 南通倍佳机械科技有限公司 | Fire-fighting monitoring video intelligent tamper-proofing method based on big data |
| CN115119016A (en) * | 2022-06-29 | 2022-09-27 | 王雨佳 | Information data encryption algorithm |
-
2022
- 2022-10-10 CN CN202211231069.6A patent/CN115311607B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801062A (en) * | 2021-04-07 | 2021-05-14 | 平安科技（深圳）有限公司 | Live video identification method, device, equipment and medium |
| CN115119016A (en) * | 2022-06-29 | 2022-09-27 | 王雨佳 | Information data encryption algorithm |
| CN114926317A (en) * | 2022-07-20 | 2022-08-19 | 南通倍佳机械科技有限公司 | Fire-fighting monitoring video intelligent tamper-proofing method based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 赵耿等: "一种基于隐私保护的云端视频监控移动物体检测方法", 《北京电子科技学院学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115311607B (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Improving triplet-wise training of convolutional neural network for vehicle re-identification | |
| Ntoutsi et al. | Density-based projected clustering over high dimensional data streams | |
| Yang et al. | Moving object detection for dynamic background scenes based on spatiotemporal model | |
| CN110751591B (en) | Self-adaptive color image steganography method based on fuzzy inference system | |
| US20040233987A1 (en) | Method for segmenting 3D objects from compressed videos | |
| Rao et al. | Detection of anomalous crowd behaviour using hyperspherical clustering | |
| Bakas et al. | Mpeg double compression based intra-frame video forgery detection using cnn | |
| Gupta et al. | Video authentication in digital forensic | |
| Zhou | End-to-end video violence detection with transformer | |
| Chao et al. | Track merging for effective video query processing | |
| Liu et al. | Feature Mining and Neuro-Fuzzy Inference System for Steganalysis of LSB Matching Stegangoraphy in Grayscale Images. | |
| CN115311607B (en) | Epoxy resin production monitoring data management method | |
| Singh et al. | Crowd escape event detection via pooling features of optical flow for intelligent video surveillance systems | |
| An et al. | Face recognition in multi-camera surveillance videos using dynamic Bayesian network | |
| Ouyang et al. | The comparison and analysis of extracting video key frame | |
| CN118264679A (en) | Cloud computing-based intelligent video chip data monitoring system and method | |
| Jayamohan et al. | Review of video analytics method for video surveillance | |
| Olatunji et al. | Dynamic threshold for resource tracking in observed scenes | |
| Katsenou et al. | Understanding video texture—A basis for video compression | |
| Hu et al. | Revisit Visual Representation in Analytics Taxonomy: A Compression Perspective | |
| CN102289816B (en) | Video motion characteristic extraction method based on fuzzy concept lattice | |
| Karthikeyan et al. | Automatic forged scene detection in advanced video using combined Mpeg-2 and optical flow features | |
| EP2372640A1 (en) | Methods of representing and analysing images | |
| Jin et al. | Computational and statistical tradeoffs in inferring combinatorial structures of ising model | |
| Zheng et al. | A novel method for quantifying target tracking difficulty of the infrared image sequence |
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 |