CN115497031A - Performance detection method and system of PVC sheath material based on image processing - Google Patents
Performance detection method and system of PVC sheath material based on image processing Download PDFInfo
- Publication number
- CN115497031A CN115497031A CN202211417756.7A CN202211417756A CN115497031A CN 115497031 A CN115497031 A CN 115497031A CN 202211417756 A CN202211417756 A CN 202211417756A CN 115497031 A CN115497031 A CN 115497031A
- Authority
- CN
- China
- Prior art keywords
- video frame
- processed
- sheath material
- monitoring video
- material monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N25/00—Investigating or analyzing materials by the use of thermal means
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/809—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of classification results, e.g. where the classifiers operate on the same input data
-
- 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- 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)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Analytical Chemistry (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Chemical & Material Sciences (AREA)
- Medical Informatics (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a performance detection method and system of a PVC sheath material based on image processing, and relates to the technical field of data processing. In the invention, the state of a plurality of PVC sheath materials produced in batch in a temperature control space is subjected to video monitoring to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheath materials. And analyzing and identifying each to-be-processed sheath material monitoring video frame included in the to-be-processed sheath material monitoring video by using the target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each to-be-processed sheath material monitoring video frame. And fusing initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed to form a corresponding target heat resistance identification result, wherein the target heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC jacket materials. Based on the above, the detection efficiency of the heat resistance can be provided to some extent.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for detecting the performance of a PVC sheath material based on image processing.
Background
Polyvinyl chloride (PVC), which is abbreviated as PVC in english, is a polymer obtained by polymerizing Vinyl Chloride Monomer (VCM) in the presence of an initiator such as peroxide and azo compound, or in the presence of light or heat according to a radical polymerization mechanism. Among them, vinyl chloride homopolymers and vinyl chloride copolymers are collectively called vinyl chloride resins. In addition, PVC is used in many applications, and for example, PVC can be used as a sheathing material to form a cable sheath. Considering that the PVC sheath material has certain heat resistance in the application environment, the PVC sheath material also needs to be detected before being put into use, but in the prior art, the detection efficiency is not high.
Disclosure of Invention
In view of the above, the present invention provides a method and a system for detecting performance of a PVC sheathing material based on image processing, so as to provide a certain detection efficiency of heat resistance performance.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
a performance detection method of a PVC sheath material based on image processing comprises the following steps:
performing video monitoring on the state of a plurality of PVC sheathing materials produced in batch in a temperature control space to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheathing materials, wherein the sheath material monitoring videos to be processed comprise a plurality of sheath material monitoring video frames to be processed;
analyzing and identifying each to-be-processed sheathing material monitoring video frame included in the to-be-processed sheathing material monitoring video by utilizing a target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each to-be-processed sheathing material monitoring video frame, wherein the initial heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC sheathing materials in corresponding time;
and fusing initial heat resistance identification results corresponding to each frame of the monitoring video frame of the sheath material to be processed to form target heat resistance identification results corresponding to the plurality of PVC sheath materials, wherein the target heat resistance identification results are used for reflecting the heat resistance degree of the plurality of PVC sheath materials.
In some preferred embodiments, in the method for detecting performance of PVC sheathing materials based on image processing, the step of performing video monitoring on the state of a plurality of PVC sheathing materials produced in batch in a temperature control space to form a sheath material monitoring video to be processed corresponding to the plurality of PVC sheathing materials includes:
after a plurality of PVC sheath materials produced in batch are placed in a temperature control space, temperature control is carried out on the temperature control space through temperature control equipment arranged in the temperature control space, so that the temperature of the temperature control space is maintained at a target temperature or is sequentially increased to the target temperature;
the method comprises the steps of carrying out video monitoring on states of a plurality of PVC sheathing materials in the temperature control space in the process of carrying out temperature control on the temperature control space in batch production to form to-be-processed sheathing material monitoring videos corresponding to the plurality of PVC sheathing materials, wherein each frame of to-be-processed sheathing material monitoring video frame included by the to-be-processed sheathing material monitoring videos has the shape information of the plurality of PVC sheathing materials.
In some preferred embodiments, in the method for detecting performance of PVC sheathing material based on image processing, the step of analyzing the neural network by using a target video frame, analyzing and identifying each frame of sheathing material monitoring video frame to be processed included in the sheathing material monitoring video to be processed, so as to output an initial heat resistance identification result corresponding to each frame of sheathing material monitoring video frame to be processed includes:
analyzing and outputting a plurality of initial sheath material monitoring video frame area sets to be processed corresponding to the sheath material monitoring video frames to be processed according to the video frame area matching degree between the sheath material monitoring video frame areas to be processed included in the sheath material monitoring video frames to be processed;
performing characteristic matching classification processing on the initial sheath material monitoring video frame region set to be processed according to the video frame region pixel key point characteristic information corresponding to each initial sheath material monitoring video frame region set to be processed, so as to form a plurality of target sheath material monitoring video frame region sets to be processed corresponding to the sheath material monitoring video frames to be processed;
utilizing a target video frame analysis neural network to respectively perform heat resistance identification processing on the to-be-processed sheath material monitoring video frame region included in each target to-be-processed sheath material monitoring video frame region set so as to form a region joint heat resistance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set;
and fusing the combined heat resistance identification results of the areas corresponding to the target to-be-processed sheath material monitoring video frame area set to output the initial heat resistance identification results corresponding to the to-be-processed sheath material monitoring video frames.
In some preferred embodiments, in the performance detection method of the PVC sheathing material based on image processing, the monitoring video frame region of the sheathing material to be processed includes a supervised monitoring video frame region of the sheathing material to be processed having identification information of the video frame region and an unsupervised monitoring video frame region of the sheathing material to be processed having no identification information of the video frame region;
the method comprises the following steps of analyzing and outputting a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to-be-processed sheath material monitoring video frames according to the video frame region matching degree between to-be-processed sheath material monitoring video frame regions included by the to-be-processed sheath material monitoring video frames, and comprises the following steps of:
constructing a plurality of corresponding first to-be-processed sheath material monitoring video frame region sets according to video frame region identification information of the surveillance to-be-processed sheath material monitoring video frame regions, wherein the number of the first to-be-processed sheath material monitoring video frame region sets is equal to the number of identification categories corresponding to the video frame region identification information;
according to the matching degree of video frame regions between the unsupervised to-be-processed sheath material monitoring video frame regions and the supervised to-be-processed sheath material monitoring video frame regions included in each first to-be-processed sheath material monitoring video frame region set, carrying out set distribution processing on the unsupervised to-be-processed sheath material monitoring video frame regions so as to realize adjustment on the plurality of first to-be-processed sheath material monitoring video frame region sets and form a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to the to-be-processed sheath material monitoring video frames.
In some preferred embodiments, in the method for detecting performance of a PVC sheathing material based on image processing, the step of performing set distribution processing on the unsupervised to-be-processed sheathing material surveillance video frame regions according to a video frame region matching degree between the unsupervised to-be-processed sheathing material surveillance video frame region and a surveillance to-be-processed sheathing material surveillance video frame region included in each set of the first to-be-processed sheathing material surveillance video frame regions to realize adjustment of the sets of the first to-be-processed sheathing material surveillance video frame regions to form a plurality of initial to-be-processed sheathing material surveillance video frame region sets corresponding to the to-be-processed sheathing material surveillance video frames includes:
calculating and outputting a video frame region matching degree mean value between the unsupervised to-be-processed sheath material monitoring video frame region and each first to-be-processed sheath material monitoring video frame region set according to the video frame region matching degree between the unsupervised to-be-processed sheath material monitoring video frame region and the video frame region matching degree between the supervised to-be-processed sheath material monitoring video frame regions included in each first to-be-processed sheath material monitoring video frame region set;
if the video frame area matching degree mean value is greater than or equal to a pre-configured video frame area matching degree reference value, distributing the corresponding non-supervised to-be-processed sheath material monitoring video frame area to a corresponding first to-be-processed sheath material monitoring video frame area set to realize the adjustment of the first to-be-processed sheath material monitoring video frame area set;
if the video frame area matching degree mean value is smaller than the video frame area matching degree reference value, determining a new first to-be-processed sheath material monitoring video frame area set according to the corresponding unsupervised to-be-processed sheath material monitoring video frame area;
and marking the adjusted first to-be-processed sheath material monitoring video frame region set to form a corresponding initial to-be-processed sheath material monitoring video frame region set.
In some preferred embodiments, in the performance detection method of a PVC sheathing material based on image processing, the step of performing feature matching classification processing on the initial sheathing material surveillance video frame region set to be processed according to the video frame region pixel key point feature information corresponding to each initial sheathing material surveillance video frame region set to be processed to form a plurality of target sheathing material surveillance video frame region sets to be processed corresponding to sheathing material surveillance video frames to be processed includes:
respectively analyzing a representative parameter of the pixel key point characteristic information corresponding to each initial to-be-processed sheathing material monitoring video frame region set according to the pixel key point characteristic information of the video frame region corresponding to each initial to-be-processed sheathing material monitoring video frame region set;
analyzing and outputting the video frame region matching degree between the initial sheath material monitoring video frame region sets to be processed according to the pixel key point characteristic information representative parameters corresponding to each initial sheath material monitoring video frame region set to be processed, and performing characteristic matching classification processing on the initial sheath material monitoring video frame region sets to be processed according to the video frame region matching degree between the initial sheath material monitoring video frame region sets to be processed so as to form a plurality of target sheath material monitoring video frame region sets to be processed corresponding to the sheath material monitoring video frames to be processed.
In some preferred embodiments, in the method for detecting performance of a PVC sheathing material based on image processing, the step of analyzing, according to the feature information of the pixel key point of the video frame region corresponding to each initial sheathing material monitoring video frame region set to be processed, the feature information of the pixel key point corresponding to each initial sheathing material monitoring video frame region set to be processed respectively represents a parameter includes:
analyzing and outputting video frame region pixel key points of each initial to-be-processed sheath material monitoring video frame region set according to the same pixel value quantity of each pixel point pixel value in each initial to-be-processed sheath material monitoring video frame region set;
and performing result fusion processing on data mining results corresponding to the key point feature information of the video frame region corresponding to each initial sheath material monitoring video frame region set according to corresponding key pixel value parameters of the pixel points to output representative parameters of the key point feature information of the pixel corresponding to each initial sheath material monitoring video frame region set, wherein the key pixel value parameters of the pixel points are obtained by analyzing the number of the same pixel values corresponding to the key point of the pixel of each video frame region.
In some preferred embodiments, in the performance detection method of a PVC sheathing material based on image processing, the step of analyzing a neural network by using a target video frame, and performing heat resistance identification processing on a sheathing material surveillance video frame region to be processed included in each target sheathing material surveillance video frame region set to be processed, respectively, to form a region joint heat resistance identification result corresponding to each target sheathing material surveillance video frame region set to be processed, includes:
respectively analyzing image information of sheath material monitoring video frame areas to be processed in each target sheath material monitoring video frame area set to determine corresponding video frame pixel units based on image information analysis results analyzed by the image information, and then marking the video frame pixel units to form video frame pixel content identification information corresponding to each target sheath material monitoring video frame area set to be processed;
according to the corresponding key pixel point content identification information and the video frame pixel content identification information, carrying out heat resistance identification processing by using a target video frame analysis neural network to form a region joint heat resistance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set; and analyzing and forming the key pixel point content identification information according to the determined video frame region pixel key points after pixel key point identification processing is carried out on the jacket material monitoring video frame region to be processed, which is included in each target jacket material monitoring video frame region set to be processed.
In some preferred embodiments, in the method for detecting performance of PVC sheathing materials based on image processing, the step of fusing initial heat resistance recognition results corresponding to each frame of the monitoring video frame of the sheathing material to be processed to form target heat resistance recognition results corresponding to the multiple PVC sheathing materials includes:
determining video frame time sequence information corresponding to each frame of the to-be-processed sheathing material monitoring video frame, and analyzing and outputting a weight coefficient corresponding to each frame of the to-be-processed sheathing material monitoring video frame based on the video frame time sequence information corresponding to each frame of the to-be-processed sheathing material monitoring video frame, wherein the weight coefficient and the video frame time sequence information have a positive correlation corresponding relation;
and weighting and fusing initial heat-resistant performance identification results corresponding to each frame of the jacket material monitoring video frame to be processed according to the weight coefficient corresponding to each frame of the jacket material monitoring video frame to be processed, so as to form target heat-resistant performance identification results corresponding to the plurality of PVC jacket materials.
The embodiment of the invention also provides a performance detection system of the PVC sheathing material based on image processing, which comprises a processor and a memory, wherein the memory is used for storing computer programs, and the processor is used for executing the computer programs so as to realize the method.
The embodiment of the invention provides a performance detection method and system of a PVC sheath material based on image processing, which is used for carrying out video monitoring on the state of a plurality of PVC sheath materials produced in batch in a temperature control space so as to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheath materials. And analyzing and identifying each to-be-processed sheathing material monitoring video frame included in the to-be-processed sheathing material monitoring video by using the target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each to-be-processed sheathing material monitoring video frame. And fusing initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed to form a corresponding target heat resistance identification result, wherein the target heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC jacket materials. Based on the steps, the target video frame analysis neural network is used for carrying out overall identification analysis on the sheath material monitoring videos to be processed corresponding to the plurality of PVC sheath materials, and compared with the conventional technical means for analyzing the heat resistance of each PVC sheath material independently, the detection efficiency of the heat resistance can be improved to a certain extent.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
Fig. 1 is a block diagram of a system for detecting performance of a PVC sheathing material based on image processing according to an embodiment of the present invention.
Fig. 2 is a schematic flow chart illustrating steps included in a method for detecting the performance of a PVC sheathing material based on image processing according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of modules included in the performance detection apparatus for a PVC jacket material based on image processing according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a performance detection system for a PVC sheathing material based on image processing. Wherein the performance detection system may include a memory and a processor.
In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The memory can have stored therein at least one software function (computer program) which can be present in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, so as to implement the performance detection method for PVC sheathing material based on image processing provided by the embodiments of the present invention (as described later).
For example, in some embodiments, the Memory may be, but is not limited to, random Access Memory (RAM), read Only Memory (ROM), programmable Read-Only Memory (PROM), erasable Read-Only Memory (EPROM), electrically Erasable Read-Only Memory (EEPROM), and the like. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components.
For example, in some embodiments, the performance detection system for PVC jacket material based on image processing may be a server with data processing capability. In some embodiments, the image processing-based performance detection system for PVC sheathing material may further include a communication unit.
With reference to fig. 2, an embodiment of the present invention further provides a performance detection method for a PVC sheathing material based on image processing, which is applicable to the performance detection system for a PVC sheathing material based on image processing. The method steps defined by the flow related to the method for detecting the performance of the PVC sheathing material based on the image processing can be realized by the system for detecting the performance of the PVC sheathing material based on the image processing.
The specific process shown in FIG. 2 will be described in detail below.
And S110, performing video monitoring on the state of a plurality of PVC sheath materials produced in batch in a temperature control space to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheath materials.
In the embodiment of the invention, the performance detection system of the PVC sheath material based on image processing can perform video monitoring on the state of a plurality of PVC sheath materials produced in batch in a temperature control space so as to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheath materials. The to-be-processed sheath material monitoring video comprises a plurality of frames of to-be-processed sheath material monitoring video.
And step S120, analyzing and identifying each to-be-processed sheath material monitoring video frame included in the to-be-processed sheath material monitoring video by utilizing a target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each to-be-processed sheath material monitoring video frame.
In the embodiment of the invention, the performance detection system of the PVC sheathing material based on image processing can analyze the neural network by using the target video frame, analyze and identify each frame of sheathing material monitoring video frame to be processed included in the sheathing material monitoring video to be processed, and output the initial heat resistance performance identification result corresponding to each frame of sheathing material monitoring video frame to be processed. And the initial heat resistance identification result is used for reflecting the heat resistance degree of the PVC sheath materials in corresponding time.
And S130, fusing initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed to form target heat resistance identification results corresponding to the plurality of PVC jacket materials.
In the embodiment of the invention, the performance detection system of the PVC sheathing material based on image processing can fuse the initial heat resistance identification results corresponding to each frame of the monitoring video frame of the sheathing material to be processed to form the target heat resistance identification results corresponding to a plurality of PVC sheathing materials. And the target heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC sheath materials.
Based on the steps, the target video frame analysis neural network is used for carrying out overall identification analysis on the sheath material monitoring video to be processed corresponding to the plurality of PVC sheath materials, and compared with the conventional technical means for analyzing the heat resistance of each PVC sheath material independently, the detection efficiency of the heat resistance can be improved to a certain extent, so that the problem of low detection efficiency in the prior art is solved. In addition, the initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed are fused, so that the precision of the detection result can be provided to a certain extent.
For example, in some embodiments, for step S110, the specific implementation process is as follows:
after a plurality of PVC sheath materials produced in batch are placed in a temperature control space, temperature control is carried out on the temperature control space through temperature control equipment arranged in the temperature control space, so that the temperature of the temperature control space is maintained at a target temperature or is sequentially increased to the target temperature (a specific temperature control mode can be configured according to actual application requirements);
the method comprises the steps of carrying out video monitoring on states of multiple PVC sheathing materials in the temperature control space in the process of carrying out temperature control on the temperature control space in batch production to form to-be-processed sheathing material monitoring videos corresponding to the multiple PVC sheathing materials, wherein each frame of to-be-processed sheathing material monitoring video frame included in the to-be-processed sheathing material monitoring videos has shape information (such as deformation degree of shape outline and the like) of the multiple PVC sheathing materials.
For example, in some embodiments, for step S120, the specific implementation process is:
analyzing and outputting a plurality of initial sheath material monitoring video frame region sets corresponding to the sheath material monitoring video frames to be processed according to the video frame region matching degree between the sheath material monitoring video frame regions to be processed included in the sheath material monitoring video frames to be processed (exemplarily, image information possibly possessed by dissimilar sheath material monitoring video frame regions to be processed, such as different information of the shape and the like of a PVC sheath material, can be allocated to different initial sheath material monitoring video frame region sets, on the contrary, image information possibly possessed by similar sheath material monitoring video frame regions to be processed, such as the same information of the shape and the like of the PVC sheath material, can be allocated to the same initial sheath material monitoring video frame region set to be processed);
performing feature matching classification on the initial sheath material monitoring video frame region sets to be processed according to the video frame region key point feature information corresponding to each initial sheath material monitoring video frame region set to be processed, so as to form a plurality of target sheath material monitoring video frame region sets to be processed corresponding to the sheath material monitoring video frames to be processed (for example, if the video frame region key point feature information corresponding to two initial sheath material monitoring video frame region sets to be processed is matched, the two initial sheath material monitoring video frame region sets to be processed can be combined into one target sheath material monitoring video frame region set to be processed);
analyzing a neural network by using a target video frame, and respectively performing heat resistance performance identification processing on a to-be-processed sheath material monitoring video frame region included in each target to-be-processed sheath material monitoring video frame region set to form a region joint heat resistance performance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set (through the two steps, the to-be-processed sheath material monitoring video frame can be divided into a plurality of target to-be-processed sheath material monitoring video frame region sets, so that heat resistance performance identification processing can be performed independently, so that interference is not generated when heat resistance performance identification processing is performed together due to high matching degree between the to-be-processed sheath material monitoring video frame regions in one target to-be-processed sheath material monitoring video frame region set, and thus the reliability of the identification result is ensured;
and fusing the combined heat resistance identification results of the areas corresponding to each target to-be-processed sheathing material monitoring video frame area set (such as weighted sum calculation, wherein during weighted sum calculation, a positively correlated weighting coefficient can be determined according to the number of sheathing material monitoring video frame areas to be processed, which are included in one target to-be-processed sheathing material monitoring video frame area set, and in addition, the area size of each sheathing material monitoring video frame area to be processed can be consistent, that is, the sheathing material monitoring video frames to be processed can be divided in equal size to form a plurality of sheathing material monitoring video frame areas to be processed, which correspond to the sheathing material monitoring video frames to be processed), so as to output the initial heat resistance identification results corresponding to the sheathing material monitoring video frames to be processed.
For example, the to-be-processed sheath material monitoring video frame region includes a supervised to-be-processed sheath material monitoring video frame region having video frame region identification information (which may be an artificial marker or the like) and an unsupervised to-be-processed sheath material monitoring video frame region having no video frame region identification information, and based on this, in some embodiments, for the video frame region matching degree between the to-be-processed sheath material monitoring video frame regions included in the to-be-processed sheath material monitoring video frame, a step of analyzing and outputting a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to the to-be-processed sheath material monitoring video frame is specifically implemented as follows:
constructing a plurality of corresponding first to-be-processed sheathing material monitoring video frame region sets according to video frame region identification information of the surveillance to-be-processed sheathing material monitoring video frame regions, wherein the number of the first to-be-processed sheathing material monitoring video frame region sets is equal to the number of identification categories corresponding to the video frame region identification information (namely, one video frame region identification information can construct one first to-be-processed sheathing material monitoring video frame region set, namely, every two surveillance to-be-processed sheathing material monitoring video frame regions with the same video frame region identification information are distributed into one first to-be-processed sheathing material monitoring video frame region set, and every two surveillance to-be-processed sheathing material monitoring video frame regions with different video frame region identification information are distributed into different first to-be-processed sheathing material monitoring video frame region sets);
according to the matching degree of video frame regions between the unsupervised to-be-processed sheath material monitoring video frame regions and the supervised to-be-processed sheath material monitoring video frame regions included in each first to-be-processed sheath material monitoring video frame region set, carrying out set distribution processing on the unsupervised to-be-processed sheath material monitoring video frame regions so as to realize adjustment on the plurality of first to-be-processed sheath material monitoring video frame region sets and form a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to the to-be-processed sheath material monitoring video frames.
For example, in some embodiments, the step of performing set distribution processing on the unsupervised to-be-processed jacket material surveillance video frame region according to a video frame region matching degree between the unsupervised to-be-processed jacket material surveillance video frame region and a supervised to-be-processed jacket material surveillance video frame region included in each set of the first to-be-processed jacket material surveillance video frame regions to adjust the sets of the plurality of first to-be-processed jacket material surveillance video frame regions and form a plurality of initial to-be-processed jacket material surveillance video frame region sets corresponding to the to-be-processed jacket material surveillance video frame specifically includes:
calculating and outputting a video frame region matching degree mean value between the unsupervised to-be-processed sheath material monitoring video frame region and each first to-be-processed sheath material monitoring video frame region according to the video frame region matching degree between the unsupervised to-be-processed sheath material monitoring video frame region and the supervised to-be-processed sheath material monitoring video frame region included in each first to-be-processed sheath material monitoring video frame region set (exemplarily, for each first to-be-processed sheath material monitoring video frame region set, performing mean value calculation on the video frame region matching degree between each frame of the supervised to-be-processed sheath material monitoring video frame region included in the first to-be-processed sheath material monitoring video frame region set and the unsupervised to-be-processed sheath material monitoring video frame region to obtain a video frame region matching degree mean value between the first to-be-processed sheath material monitoring video frame region set and the unsupervised to-be-processed sheath material monitoring video frame region);
if the video frame region matching degree mean value is greater than or equal to the preconfigured video frame region matching degree reference value, allocating the corresponding unsupervised to-be-processed sheath material surveillance video frame region to the corresponding first to-be-processed sheath material surveillance video frame region set (for example, if there are a plurality of first to-be-processed sheath material surveillance video frame region sets corresponding to the video frame region matching degree mean value greater than or equal to the preconfigured video frame region matching degree reference value, the first to-be-processed sheath material surveillance video frame region set corresponding to the video frame region matching degree mean value with the maximum value can be allocated), and implementing adjustment on the first to-be-processed sheath material surveillance video frame region set;
if the average value of the video frame area matching degrees is smaller than the reference value of the video frame area matching degrees, determining a new first to-be-processed sheath material monitoring video frame area set according to the corresponding unsupervised to-be-processed sheath material monitoring video frame area;
and marking the adjusted first to-be-processed sheath material monitoring video frame region set to form a corresponding initial to-be-processed sheath material monitoring video frame region set.
For example, in some embodiments, the calculation process of the video frame region matching degree between the unsupervised to-be-processed sheath material monitoring video frame region and the supervised to-be-processed sheath material monitoring video frame region specifically includes:
extracting PVC sheath material particle profiles from the unsupervised to-be-processed sheath material monitoring video frame region to form a first PVC sheath material particle profile set corresponding to the unsupervised to-be-processed sheath material monitoring video frame region, wherein the first PVC sheath material particle profile set comprises the particle profile of each PVC sheath material in the unsupervised to-be-processed sheath material monitoring video frame region; extracting PVC sheath material particle profiles in the surveillance video frame region of the sheath material to be processed to form a second PVC sheath material particle profile set corresponding to the surveillance video frame region of the sheath material to be processed, wherein the second PVC sheath material particle profile set comprises the particle profile of each PVC sheath material in the surveillance video frame region of the sheath material to be processed;
respectively serializing the first PVC sheath material particle profile set and the second PVC sheath material particle profile set based on the same serialization rule (the serialization rule can refer to a mode of traversing particle profiles in PVC sheath material particle profile sets, such as traversing according to a path based on distribution positions in corresponding video frame regions), so as to obtain a first PVC sheath material particle profile sequence corresponding to the first PVC sheath material particle profile set and a second PVC sheath material particle profile sequence corresponding to the second PVC sheath material particle profile set;
for each particle profile in the first PVC sheath material particle profile sequence, calculating a ratio between a two-dimensional profile area corresponding to the particle profile and a two-dimensional profile perimeter, or calculating a ratio between a three-dimensional profile extract corresponding to the particle profile and a three-dimensional profile perimeter; performing curve fitting according to the ratio corresponding to each particle profile in the first PVC sheath material particle profile sequence to obtain a corresponding first ratio variation trend representative curve;
for each particle profile in the particle profile sequence of the second PVC sheath material, calculating a ratio between a two-dimensional profile area corresponding to the particle profile and a two-dimensional profile perimeter, or calculating a ratio between a three-dimensional profile extract corresponding to the particle profile and a three-dimensional profile perimeter; performing curve fitting according to the ratio corresponding to each particle profile in the second PVC sheath material particle profile sequence to obtain a corresponding second ratio variation trend representative curve;
determining each first curve peak segment from the first ratio variation trend representative curve, and then performing screening processing on the first curve peak segments according to a ratio corresponding to the curve peak segment corresponding to each first curve peak segment to form at least one target first curve peak segment (for example, a part of the first curve peak segment with the largest corresponding ratio may be marked as the target first curve peak segment), where the width of each first curve peak segment is equal, such as equal to a pre-configured first width threshold; determining each second curve peak segment from the second ratio variation trend representative curve, and then performing screening processing on the second curve peak segments according to a ratio corresponding to a curve peak corresponding to each second curve peak segment to form at least one target second curve peak segment, for example, a part of the second curve peak segment with the largest ratio may be marked as the target second curve peak segment), where the width of each second curve peak segment is equal, such as equal to a preconfigured first width threshold;
determining each first curve valley segment from the first ratio variation trend representative curve, and then performing a screening process on the first curve valley segments according to a ratio corresponding to a curve valley point corresponding to each first curve valley segment to form at least one target first curve valley segment (for example, a part of the first curve valley segment with the smallest ratio may be marked as the target first curve valley segment), where the width of each first curve valley segment is equal, for example, equal to a pre-configured second width threshold, and the second width threshold may be smaller than the first width threshold; determining each second curve valley segment from the second ratio variation trend representative curve, and then performing screening processing on the second curve valley segments according to a ratio corresponding to a curve valley point corresponding to each second curve valley segment to form at least one target second curve valley segment, for example, a part of the second curve valley segments with the smallest ratio may be marked as the target second curve valley segments), wherein the width of each second curve valley segment is equal, for example, equal to the second width threshold;
respectively calculating the similarity of the band tracks between each first curve peak band and each second curve peak band, and then fusing the similarity of the band tracks between each first curve peak band and each second curve peak band (such as calculating an average value, or performing weighted summation calculation according to the average value of the corresponding ratio of the corresponding peak points, namely taking the average value as a corresponding weighting coefficient) to obtain the similarity of the corresponding first band tracks; respectively calculating the band track similarity between each first curve valley section and each second curve valley section, and then fusing the band track similarity between each first curve valley section and each second curve valley section (such as average value calculation, or weighted summation calculation according to the average value of the ratio corresponding to the corresponding valley point, namely taking the negative correlation coefficient corresponding to the average value as the corresponding weighting coefficient) to obtain the corresponding second band track similarity;
and performing fusion calculation (such as mean value calculation) on the first waveband track similarity and the second waveband track similarity to obtain the video frame region matching degree between the unsupervised to-be-processed sheath material monitoring video frame region and the supervised to-be-processed sheath material monitoring video frame region.
For example, in some embodiments, the step of performing feature matching classification processing on the initial sheath material surveillance video frame region set to be processed according to the video frame region pixel key point feature information corresponding to each initial sheath material surveillance video frame region set to be processed to form a plurality of target sheath material surveillance video frame region sets to be processed corresponding to the sheath material surveillance video frames to be processed includes:
respectively analyzing a representative parameter of the pixel key point characteristic information corresponding to each initial to-be-processed sheathing material monitoring video frame region set according to the pixel key point characteristic information of the video frame region corresponding to each initial to-be-processed sheathing material monitoring video frame region set;
according to the pixel key point characteristic information representative parameters corresponding to each initial to-be-processed sheathing material monitoring video frame region set, analyzing and outputting the video frame region matching degree between the initial to-be-processed sheathing material monitoring video frame region sets (exemplarily, the pixel key point characteristic information representative parameters may be subjected to matching degree calculation to obtain the video frame region matching degree, for example, the pixel key point characteristic information representative parameters may be represented by vectors, and the matching degree calculation may be a vector product), and then according to the video frame region matching degree between the initial to-be-processed sheathing material monitoring video frame region sets, performing feature matching classification on the initial to-be-processed sheathing material monitoring video frame region sets (that is, two initial to-be-processed sheathing material monitoring video frame region sets with high video frame region matching degrees may be merged to form a corresponding target to-be-processed sheathing material monitoring video frame region set), so as to form a plurality of target to-be-processed sheathing material monitoring video frame region sets corresponding to the to-be-processed sheathing material monitoring video frames.
For example, in some embodiments, the step of respectively analyzing the pixel key point feature information representative parameters corresponding to each initial sheath material monitoring video frame region set to be processed according to the pixel key point feature information of the video frame region corresponding to each initial sheath material monitoring video frame region set to be processed includes:
analyzing and outputting the video frame region pixel key points of each initial to-be-processed sheathing material monitoring video frame region set according to the same pixel value quantity of each pixel value in each initial to-be-processed sheathing material monitoring video frame region set (for example, each video frame region pixel point of which the corresponding same pixel value quantity is less than a quantity reference value can be determined as a video frame region pixel key point, or the video frame region pixel key points can be determined based on other modes);
according to key parameters of pixel values of corresponding pixels, performing result fusion processing on data mining results corresponding to key point feature information of pixels in video frame regions corresponding to each initial sheath material monitoring video frame region set to be processed (that is, inside the initial sheath material monitoring video frame region set to be processed, performing fusion processing on data mining results corresponding to key point feature information of pixels in video frame regions, that is, performing weighted summation calculation based on key parameters of pixel values of corresponding pixels, where the key point feature information of pixels in video frame regions may refer to a splicing result of pixel information of the key points of pixels in video frame regions and pixel difference information between the key points of pixels in video frame regions and adjacent pixels, and then encoding and performing real-time sequence data mining on the splicing result to form corresponding data mining results) to output representative parameters of the key point feature information of pixels corresponding to each initial sheath material monitoring video frame region set to be processed, where the key parameters of pixel values of the pixels are analyzed according to key values of the same pixels corresponding to each video frame region (for example, the key values of the same pixels and the key values of the corresponding pixels).
For example, in some embodiments, for the neural network analyzed by using a target video frame, a step of performing heat resistance performance identification processing on a monitored video frame region of a to-be-processed sheath material included in each set of monitored video frame regions of the target to-be-processed sheath material to form a region joint heat resistance performance identification result corresponding to each set of monitored video frame regions of the target to-be-processed sheath material is specifically implemented as follows:
respectively analyzing image information of the sheath material monitoring video frame area to be processed in each target sheath material monitoring video frame area set, determining a corresponding video frame pixel unit based on the image information analysis result analyzed by the image information, marking the video frame pixel unit to form video frame pixel content identification information corresponding to each target sheath material monitoring video frame area set (exemplarily, profile information between every two PVC sheath materials in the sheath material monitoring video frame area to be processed can be compared and analyzed to determine a difference degree, then, pixel points corresponding to PVC sheath materials with larger difference degrees or larger average difference degrees are screened out, if the difference degree is larger than a difference threshold value, and non-screened pixel points are combined to form a corresponding video frame pixel unit, wherein the video frame pixel content identification information can be used for identifying each pixel point belonging to the video frame pixel unit in the sheath material monitoring video frame area to be processed;
according to the corresponding key pixel point content identification information and the video frame pixel content identification information, carrying out heat resistance identification processing by using a target video frame analysis neural network to form a region joint heat resistance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set; the key pixel point content identification information is analyzed and formed according to the determined video frame area pixel key points after the pixel key point identification processing is performed on the to-be-processed sheath material surveillance video frame area included in each target to-be-processed sheath material surveillance video frame area set (for example, the corresponding video frame area pixel key points can be determined based on the key pixel point content identification information, then the video frame area pixel key points are further screened according to the key pixel point content identification information to select each video frame area pixel key point belonging to the video frame pixel unit, and then the heat resistance identification processing is performed by utilizing a target video frame analysis neural network according to the data mining result corresponding to the video frame area pixel key point characteristic information corresponding to the video frame area pixel key points, so that the corresponding area combined heat resistance identification result is obtained).
For example, in some embodiments, for step S130, the specific implementation process is:
determining video frame time sequence information corresponding to each frame of the to-be-processed sheathing material monitoring video frame, and analyzing and outputting a weight coefficient corresponding to each frame of the to-be-processed sheathing material monitoring video frame based on the video frame time sequence information corresponding to each frame of the to-be-processed sheathing material monitoring video frame, wherein the weight coefficient and the video frame time sequence information have a positive correlation corresponding relationship (that is, the earlier the time sequence is, the smaller the corresponding weight coefficient can be, and the later the time sequence is, the larger the corresponding weight coefficient can be);
and performing weighted fusion on the initial heat resistance identification result corresponding to each frame of the jacket material monitoring video frame to be processed (namely performing weighted summation on the initial heat resistance identification results to obtain a target heat resistance identification result) according to the weight coefficient corresponding to each frame of the jacket material monitoring video frame to be processed, so as to form a target heat resistance identification result corresponding to the plurality of PVC jacket materials.
With reference to fig. 3, an embodiment of the present invention further provides a performance detection apparatus for a PVC sheathing material based on image processing, which may be applied to the performance detection system for a PVC sheathing material based on image processing. Wherein, the performance detection device of the PVC sheathing material based on image processing may include:
the sheath material video monitoring module is used for carrying out video monitoring on the state of a plurality of PVC sheath materials produced in batch in a temperature control space so as to form to-be-processed sheath material monitoring videos corresponding to the plurality of PVC sheath materials, wherein the to-be-processed sheath material monitoring videos comprise a plurality of frames of to-be-processed sheath material monitoring video frames; the video analysis and identification module is used for analyzing a neural network by using a target video frame, analyzing and identifying each to-be-processed sheathing material monitoring video frame included in the to-be-processed sheathing material monitoring video to output an initial heat resistance identification result corresponding to each to-be-processed sheathing material monitoring video frame, wherein the initial heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC sheathing materials in corresponding time; and the identification result fusion module is used for fusing initial heat resistance identification results corresponding to each frame of the to-be-processed sheath material monitoring video frame to form target heat resistance identification results corresponding to the multiple PVC sheath materials, and the target heat resistance identification results are used for reflecting the heat resistance degrees of the multiple PVC sheath materials.
In summary, the performance detection method and system for PVC sheathing materials based on image processing provided by the present invention perform video monitoring on the state of a plurality of PVC sheathing materials produced in batch in a temperature control space to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheathing materials. And analyzing and identifying each to-be-processed sheath material monitoring video frame included in the to-be-processed sheath material monitoring video by using the target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each to-be-processed sheath material monitoring video frame. And fusing initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed to form a corresponding target heat resistance identification result, wherein the target heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC jacket materials. Based on the steps, the overall identification and analysis of the sheath material monitoring videos to be processed corresponding to the multiple PVC sheath materials are carried out through the target video frame analysis neural network, and compared with a conventional technical means for analyzing the heat resistance of each PVC sheath material individually, the method can provide the detection efficiency of the heat resistance to a certain extent, so that the problem of low detection efficiency in the prior art is solved.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A performance detection method of a PVC sheath material based on image processing is characterized by comprising the following steps:
performing video monitoring on the state of a plurality of PVC sheathing materials produced in batch in a temperature control space to form sheath material monitoring videos to be processed corresponding to the plurality of PVC sheathing materials, wherein the sheath material monitoring videos to be processed comprise a plurality of frames of sheath material monitoring video frames to be processed;
analyzing and identifying each frame of to-be-processed sheath material monitoring video frame included in the to-be-processed sheath material monitoring video by using a target video frame analysis neural network so as to output an initial heat resistance identification result corresponding to each frame of to-be-processed sheath material monitoring video frame, wherein the initial heat resistance identification result is used for reflecting the heat resistance degree of the plurality of PVC sheath materials in corresponding time;
and fusing initial heat resistance identification results corresponding to each frame of the to-be-processed sheath material monitoring video frame to form target heat resistance identification results corresponding to the multiple PVC sheath materials, wherein the target heat resistance identification results are used for reflecting the heat resistance degrees of the multiple PVC sheath materials.
2. The method for detecting the performance of the PVC sheathing material based on the image processing as claimed in claim 1, wherein the step of performing video monitoring on the state of a plurality of PVC sheathing materials produced in batch in a temperature control space to form a sheath material monitoring video to be processed corresponding to the plurality of PVC sheathing materials comprises:
after a plurality of PVC sheath materials produced in batch are placed in a temperature control space, temperature control is carried out on the temperature control space through temperature control equipment arranged in the temperature control space, so that the temperature of the temperature control space is maintained at a target temperature or is sequentially increased to the target temperature;
it is right in-process that temperature control is carried out in the temperature control space, to batch production many PVC sheath material carry out video monitoring in the state in the temperature control space, in order to form the pending sheath material surveillance video that many PVC sheath material correspond, each frame pending sheath material surveillance video frame that pending sheath material surveillance video includes has the appearance form information of many PVC sheath materials.
3. The method for detecting the performance of the PVC sheathing material based on the image processing as claimed in claim 1, wherein the step of analyzing the neural network by using the target video frame to analyze and identify each frame of sheathing material monitoring video frame to be processed included in the sheathing material monitoring video to be processed, so as to output the initial heat resistance identification result corresponding to each frame of sheathing material monitoring video frame to be processed, includes:
analyzing and outputting a plurality of initial sheath material monitoring video frame area sets to be processed corresponding to the sheath material monitoring video frames to be processed according to the video frame area matching degree between the sheath material monitoring video frame areas to be processed included in the sheath material monitoring video frames to be processed;
performing characteristic matching classification processing on the initial sheath material monitoring video frame region set to be processed according to the video frame region pixel key point characteristic information corresponding to each initial sheath material monitoring video frame region set to be processed, so as to form a plurality of target sheath material monitoring video frame region sets to be processed corresponding to the sheath material monitoring video frames to be processed;
utilizing a target video frame analysis neural network to respectively perform heat resistance identification processing on the to-be-processed sheath material monitoring video frame region included in each target to-be-processed sheath material monitoring video frame region set so as to form a region joint heat resistance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set;
and fusing the combined heat resistance identification results of the areas corresponding to the target to-be-processed sheath material monitoring video frame area set to output the initial heat resistance identification result corresponding to the to-be-processed sheath material monitoring video frame.
4. The method for detecting the performance of the PVC sheathing material based on the image processing as claimed in claim 3, wherein the sheathing material surveillance video frame area to be processed includes a supervised sheathing material surveillance video frame area having identification information of the video frame area and an unsupervised sheathing material surveillance video frame area to be processed not having the identification information of the video frame area;
the method comprises the following steps of analyzing and outputting a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to-be-processed sheath material monitoring video frames according to the video frame region matching degree between to-be-processed sheath material monitoring video frame regions included by the to-be-processed sheath material monitoring video frames, wherein the steps comprise:
constructing a plurality of corresponding first to-be-processed sheath material monitoring video frame region sets according to video frame region identification information of the surveillance to-be-processed sheath material monitoring video frame regions, wherein the number of the first to-be-processed sheath material monitoring video frame region sets is equal to the number of identification categories corresponding to the video frame region identification information;
according to the matching degree of video frame regions between the unsupervised to-be-processed sheath material monitoring video frame regions and the supervised to-be-processed sheath material monitoring video frame regions included in each first to-be-processed sheath material monitoring video frame region set, carrying out set distribution processing on the unsupervised to-be-processed sheath material monitoring video frame regions so as to realize adjustment on the plurality of first to-be-processed sheath material monitoring video frame region sets and form a plurality of initial to-be-processed sheath material monitoring video frame region sets corresponding to the to-be-processed sheath material monitoring video frames.
5. The method for detecting performance of a PVC sheathing material based on image processing as claimed in claim 4, wherein said step of performing set distribution processing on said unsupervised to-be-processed sheathing material surveillance video frame region according to the video frame region matching degree between said unsupervised to-be-processed sheathing material surveillance video frame region and the supervised to-be-processed sheathing material surveillance video frame region included in each of said first set of to-be-processed sheathing material surveillance video frame regions, so as to realize adjustment of said plurality of sets of first to-be-processed sheathing material surveillance video frame regions, and form a plurality of initial to-be-processed sheathing material surveillance video frame region sets corresponding to said to-be-processed sheathing material surveillance video frame includes:
calculating and outputting a video frame region matching degree mean value between the unsupervised to-be-processed sheath material surveillance video frame region and each first to-be-processed sheath material surveillance video frame region set according to the video frame region matching degree between the unsupervised to-be-processed sheath material surveillance video frame region and the surveillance to-be-processed sheath material surveillance video frame region included in each first to-be-processed sheath material surveillance video frame region set;
if the average value of the video frame area matching degrees is greater than or equal to a pre-configured reference value of the video frame area matching degrees, distributing the corresponding unsupervised to-be-processed sheathing material surveillance video frame area to a corresponding first to-be-processed sheathing material surveillance video frame area set to realize the adjustment of the first to-be-processed sheathing material surveillance video frame area set;
if the video frame area matching degree mean value is smaller than the video frame area matching degree reference value, determining a new first to-be-processed sheath material monitoring video frame area set according to the corresponding unsupervised to-be-processed sheath material monitoring video frame area;
and marking the adjusted first to-be-processed sheath material monitoring video frame region set to form a corresponding initial to-be-processed sheath material monitoring video frame region set.
6. The method for detecting performance of a PVC sheathing material based on image processing as claimed in claim 3, wherein said step of performing feature matching classification processing on the initial set of sheathing material surveillance video frame regions to be processed according to the feature information of the video frame region pixel key points corresponding to each set of sheathing material surveillance video frame regions to be processed to form a plurality of sets of sheathing material surveillance video frame regions to be processed corresponding to the sheathing material surveillance video frames to be processed includes:
respectively analyzing pixel key point characteristic information representative parameters corresponding to each initial sheath material monitoring video frame region set to be processed according to the video frame region pixel key point characteristic information corresponding to each initial sheath material monitoring video frame region set to be processed;
analyzing and outputting the video frame region matching degree between the initial sheath material monitoring video frame region sets to be processed according to the pixel key point characteristic information representative parameters corresponding to each initial sheath material monitoring video frame region set to be processed, and performing characteristic matching classification processing on the initial sheath material monitoring video frame region sets to be processed according to the video frame region matching degree between the initial sheath material monitoring video frame region sets to be processed so as to form a plurality of target sheath material monitoring video frame region sets to be processed corresponding to the sheath material monitoring video frames to be processed.
7. The method for detecting the performance of the PVC sheathing material based on image processing as claimed in claim 6, wherein the step of respectively analyzing the pixel key point characteristic information representative parameters corresponding to each initial sheath material monitoring video frame region set to be processed according to the video frame region pixel key point characteristic information corresponding to each initial sheath material monitoring video frame region set to be processed comprises:
analyzing and outputting the video frame region pixel key points of each initial to-be-processed sheathing material monitoring video frame region set according to the same pixel value quantity of each pixel point pixel value in each initial to-be-processed sheathing material monitoring video frame region set;
and performing result fusion processing on data mining results corresponding to the video frame region key point characteristic information corresponding to each initial to-be-processed sheathing material monitoring video frame region set according to corresponding key pixel value parameters of pixel points to output pixel key point characteristic information representative parameters corresponding to each initial to-be-processed sheathing material monitoring video frame region set, wherein the key pixel value parameters of the pixel points are obtained by analyzing the number of the same pixel values corresponding to the key pixel points of each video frame region.
8. The method for detecting the performance of the PVC sheathing material based on image processing as claimed in claim 3, wherein the step of performing the heat resistance performance identification processing on the sheathing material surveillance video frame region to be processed included in each set of sheath material surveillance video frame region to be processed by using the target video frame analysis neural network to form the region joint heat resistance performance identification result corresponding to each set of sheath material surveillance video frame region to be processed includes:
respectively analyzing image information of the sheath material monitoring video frame areas to be processed in each target sheath material monitoring video frame area set to determine corresponding video frame pixel units based on the image information analysis result analyzed by the image information, and then marking the video frame pixel units to form video frame pixel content identification information corresponding to each target sheath material monitoring video frame area set;
according to the corresponding key pixel point content identification information and the video frame pixel content identification information, carrying out heat resistance identification processing by using a target video frame analysis neural network to form a region joint heat resistance identification result corresponding to each target to-be-processed sheath material monitoring video frame region set; and analyzing and forming the key pixel point content identification information according to the determined video frame region pixel key points after pixel key point identification processing is carried out on the jacket material monitoring video frame region to be processed, which is included in each target jacket material monitoring video frame region set to be processed.
9. The method for detecting the performance of the PVC sheathing material based on the image processing as claimed in any one of claims 1 to 8, wherein the step of fusing the initial heat resistance recognition results corresponding to each frame of the monitoring video frame of the sheathing material to be processed to form the target heat resistance recognition results corresponding to the plurality of PVC sheathing materials comprises the steps of:
determining video frame time sequence information corresponding to each frame of the sheath material monitoring video frame to be processed, analyzing and outputting a weight coefficient corresponding to each frame of the sheath material monitoring video frame to be processed based on the video frame time sequence information corresponding to each frame of the sheath material monitoring video frame to be processed, wherein the weight coefficient and the video frame time sequence information have a positive correlation corresponding relation;
and weighting and fusing the initial heat resistance identification results corresponding to each frame of the jacket material monitoring video frame to be processed according to the weight coefficient corresponding to each frame of the jacket material monitoring video frame to be processed, so as to form target heat resistance identification results corresponding to the plurality of PVC jacket materials.
10. An image processing based performance detection system for PVC sheathing materials, which is characterized by comprising a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program to realize the method of any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211417756.7A CN115497031B (en) | 2022-11-14 | 2022-11-14 | Performance detection method and system for PVC sheath material based on image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211417756.7A CN115497031B (en) | 2022-11-14 | 2022-11-14 | Performance detection method and system for PVC sheath material based on image processing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115497031A true CN115497031A (en) | 2022-12-20 |
CN115497031B CN115497031B (en) | 2023-06-13 |
Family
ID=85115617
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211417756.7A Active CN115497031B (en) | 2022-11-14 | 2022-11-14 | Performance detection method and system for PVC sheath material based on image processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115497031B (en) |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010182287A (en) * | 2008-07-17 | 2010-08-19 | Steven C Kays | Intelligent adaptive design |
US20190043488A1 (en) * | 2018-09-18 | 2019-02-07 | Intel Corporation | Method and system of neural network keyphrase detection |
US20190273948A1 (en) * | 2019-01-08 | 2019-09-05 | Intel Corporation | Method and system of neural network loop filtering for video coding |
CN110797096A (en) * | 2019-10-30 | 2020-02-14 | 合肥通用机械研究院有限公司 | Deep learning-based high-temperature mechanical property prediction method for heat-resistant alloy |
WO2020125839A1 (en) * | 2018-12-18 | 2020-06-25 | GRID INVENT gGmbH | Electronic element and electrically controlled display element |
CN112785613A (en) * | 2020-11-30 | 2021-05-11 | 上海大学 | Intelligent hearth flame image identification method |
US20210158489A1 (en) * | 2017-08-11 | 2021-05-27 | Semiconductor Energy Laboratory Co., Ltd. | Display device |
US20210192175A1 (en) * | 2019-12-20 | 2021-06-24 | Volant Aerial, Inc. | System and method for the early visual detection of forest fires using a deep convolutional neural network |
US20210335008A1 (en) * | 2020-04-27 | 2021-10-28 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for processing video frame |
CN114442582A (en) * | 2022-04-07 | 2022-05-06 | 广东祥利科技有限公司 | Polyolefin modified cable pellet production monitoring control method and system |
CN114527721A (en) * | 2022-04-25 | 2022-05-24 | 广东祥利科技有限公司 | Method and system for monitoring online production of poly-PVC (polyvinyl chloride) modified cable granules and cloud platform |
CN114596144A (en) * | 2022-03-30 | 2022-06-07 | 罗莱生活科技股份有限公司 | Training method of quilt core recommendation model and quilt core recommendation method |
CN217180664U (en) * | 2022-01-12 | 2022-08-12 | 东莞市翔宇实业有限公司 | PVC sizing material heat resistance detection device |
CN115018840A (en) * | 2022-08-08 | 2022-09-06 | 珠海市南特金属科技股份有限公司 | Method, system and device for detecting cracks of precision casting |
CN115205765A (en) * | 2022-09-15 | 2022-10-18 | 成都中轨轨道设备有限公司 | FPGA-based video analysis method and system |
-
2022
- 2022-11-14 CN CN202211417756.7A patent/CN115497031B/en active Active
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010182287A (en) * | 2008-07-17 | 2010-08-19 | Steven C Kays | Intelligent adaptive design |
US20210158489A1 (en) * | 2017-08-11 | 2021-05-27 | Semiconductor Energy Laboratory Co., Ltd. | Display device |
US20190043488A1 (en) * | 2018-09-18 | 2019-02-07 | Intel Corporation | Method and system of neural network keyphrase detection |
WO2020125839A1 (en) * | 2018-12-18 | 2020-06-25 | GRID INVENT gGmbH | Electronic element and electrically controlled display element |
US20190273948A1 (en) * | 2019-01-08 | 2019-09-05 | Intel Corporation | Method and system of neural network loop filtering for video coding |
CN110797096A (en) * | 2019-10-30 | 2020-02-14 | 合肥通用机械研究院有限公司 | Deep learning-based high-temperature mechanical property prediction method for heat-resistant alloy |
US20210192175A1 (en) * | 2019-12-20 | 2021-06-24 | Volant Aerial, Inc. | System and method for the early visual detection of forest fires using a deep convolutional neural network |
US20210335008A1 (en) * | 2020-04-27 | 2021-10-28 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for processing video frame |
CN112785613A (en) * | 2020-11-30 | 2021-05-11 | 上海大学 | Intelligent hearth flame image identification method |
CN217180664U (en) * | 2022-01-12 | 2022-08-12 | 东莞市翔宇实业有限公司 | PVC sizing material heat resistance detection device |
CN114596144A (en) * | 2022-03-30 | 2022-06-07 | 罗莱生活科技股份有限公司 | Training method of quilt core recommendation model and quilt core recommendation method |
CN114442582A (en) * | 2022-04-07 | 2022-05-06 | 广东祥利科技有限公司 | Polyolefin modified cable pellet production monitoring control method and system |
CN114527721A (en) * | 2022-04-25 | 2022-05-24 | 广东祥利科技有限公司 | Method and system for monitoring online production of poly-PVC (polyvinyl chloride) modified cable granules and cloud platform |
CN115018840A (en) * | 2022-08-08 | 2022-09-06 | 珠海市南特金属科技股份有限公司 | Method, system and device for detecting cracks of precision casting |
CN115205765A (en) * | 2022-09-15 | 2022-10-18 | 成都中轨轨道设备有限公司 | FPGA-based video analysis method and system |
Non-Patent Citations (2)
Title |
---|
吴迪等: "基于对象的监控视频摘要生成优化方法", 《微处理机》 * |
赵峰: "PVC热稳定性质量缺陷的检测和控制", 《中国氯碱》 * |
Also Published As
Publication number | Publication date |
---|---|
CN115497031B (en) | 2023-06-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Sabokrou et al. | Real-time anomaly detection and localization in crowded scenes | |
US10769261B2 (en) | User image verification | |
US20120294535A1 (en) | Face detection method and apparatus | |
Sabokrou et al. | Fast and accurate detection and localization of abnormal behavior in crowded scenes | |
KR20190071079A (en) | Apparatus and method for recognizing image | |
CN111401202A (en) | Pedestrian mask wearing real-time detection method based on deep learning | |
CN111783665A (en) | Action recognition method and device, storage medium and electronic equipment | |
KR101854258B1 (en) | Memory optimization method for applying deep learning to pc | |
CN110866428A (en) | Target tracking method and device, electronic equipment and storage medium | |
TW202117667A (en) | Object localization system and method thereof | |
CN115497031B (en) | Performance detection method and system for PVC sheath material based on image processing | |
Sahoo et al. | Unusual event detection in surveillance video using transfer learning | |
CN115620243B (en) | Pollution source monitoring method and system based on artificial intelligence and cloud platform | |
CN115588157B (en) | Performance data processing method and system of cross-linked low-smoke low-halogen polyolefin material | |
CN115147134B (en) | Product anti-counterfeiting traceability method, system and cloud platform based on industrial Internet | |
CN115457652A (en) | Pedestrian re-identification method and device based on semi-supervised learning and storage medium | |
CN115620211A (en) | Performance data processing method and system of flame-retardant low-smoke halogen-free sheath | |
JP2022184761A (en) | Concept for detecting abnormality in input data | |
CN114095734A (en) | User data compression method and system based on data processing | |
Yousefi et al. | Energy aware multi-object detection method in visual sensor network | |
CN112381136B (en) | Target detection method and device | |
CN116437057B (en) | System optimization method and system for diborane production monitoring system | |
CN115631448B (en) | Audio and video quality inspection processing method and system | |
Pathade et al. | Recognition of crowd abnormal activities using fusion of handcrafted and deep features | |
CN115620210B (en) | Method and system for determining performance of electronic wire material based on image processing |
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 | ||
CP03 | Change of name, title or address | ||
CP03 | Change of name, title or address |
Address after: No. 9, Xincheng Second Road, Xincheng Industrial Park, Xincheng Town, Xinxing County, Yunfu City, Guangdong Province 527300 Patentee after: Guangdong Xiangli Technology Group Co.,Ltd. Address before: 527499 No. 9, Xincheng Second Road, Xincheng Industrial Park, Xincheng Town, Xinxing County, Yunfu City, Guangdong Province Patentee before: GUANGDONG XIANGLI TECHNOLOGY Co.,Ltd. |