CN109978833A - Picture quality automatic testing method, system, equipment and storage medium - Google Patents
Picture quality automatic testing method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN109978833A CN109978833A CN201910162123.8A CN201910162123A CN109978833A CN 109978833 A CN109978833 A CN 109978833A CN 201910162123 A CN201910162123 A CN 201910162123A CN 109978833 A CN109978833 A CN 109978833A
- Authority
- CN
- China
- Prior art keywords
- region
- target image
- commodity
- fuzzy
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of picture quality automatic testing method, system, equipment and storage mediums, include the following steps: to obtain target image, determine the fuzzy region in target image, when occupied area ratio is less than the first proportion threshold value pre-seted to fuzzy region in the target image, determine that target image is qualified images；Identify each commodity region of target image, judge whether commodity region is overexposure region or under-exposure region, it has served as exposure area and/or when under-exposure region accounts for all commodity regions and is less than the second proportion threshold value pre-seted, has determined that target image is qualified images；Up-to-standard image is filtered out by the picture material disaggregated model pre-seted is passed through according to the qualified images selected.The present invention can screen target image according to fuzzy region, exposure area and picture material in target image, filter out up-to-standard image, so as to convenient for the commodity in target image into accurately identifying.
Description
Technical field
The present invention relates to image procossings, and in particular, to a kind of picture quality automatic testing method, system, equipment and deposits
Storage media.
Background technique
With the computing capability enhancing of hardware components in computer and smart phone, in recent years, machine learning and calculating
Machine vision technique achieves development at full speed, and many research staff, which have done a large amount of work, can assist people's everyday tasks to research and develop
Machine learning and computer vision algorithms make.
It in new retail domain, needs periodically to be acquired shelf or refrigerator image by mobile phone, and by the figure after acquisition
As being pooled in background system.Background system using image recognition technology obtain image in each commodity type, number and
Location information etc..The data of generation are pooled into report, and after regularly updating commodity data report.It just can be according to commodity data
Including that can be controlled completely to the condition of merchandise on shelf or refrigerator.
It, usually can shake because of mobile phone or camera but when being acquired by image of the mobile phone to shelf or refrigerator
Exposure problems lead to the unqualified of Image Acquisition, lead to not to image carry out identification or image recognition accuracy rate it is lower,
Therefore need to provide a kind of method that can be realized picture quality detection.
Summary of the invention
For the defects in the prior art, the object of the present invention is to provide a kind of picture quality automatic testing method, system,
Equipment and storage medium.
The picture quality automatic testing method provided according to the present invention, includes the following steps:
Step S1: target image is obtained, the fuzzy region in the target image is determined, when the fuzzy region is in institute
When stating occupied area ratio in target image and being less than the first proportion threshold value pre-seted, determine the target image for qualification figure
Picture；
Step S2: identifying each commodity region of the target image, judges whether the commodity region is overexposure
Region or under-exposure region are less than the pre-seted when the overexposure region and/or under-exposure region account for all commodity regions
When two proportion threshold values, determine the target image for qualified images；
Step S3: it will successively pass through in step S1, step S2 or successively by the qualification that step S2, step S1 are filtered out and scheme
As the picture material disaggregated model by pre-seting filters out up-to-standard image.
Preferably, the step S1 includes the following steps:
Step S101: obtaining the target image, carries out fuzzy region detection to the target image, determines the target
Fuzzy region in image, and then determine the area of the fuzzy region；
Step S102: the area of all fuzzy regions is gone out in the target image according to the areal calculation of the fuzzy region
Middle occupied area ratio；
Step S103: determining whether the occupied area ratio is less than first proportion threshold value pre-seted, and is working as
When the occupied area ratio is less than the first proportion threshold value pre-seted, determine the target image for qualified images.
Preferably, the step S2 includes the following steps:
Step S201: identifying each commodity region of the target image, and generates the ash in each commodity region
Spend histogram；
Step S202: by the grey level histogram in each commodity region benchmark corresponding with the commodity region pre-seted
Grey level histogram compares, and the commodity region is divided into overexposure region, owes to expose region and qualified region；
Step S203: it when the exposure area in the target image and/or owes in quantity and the target image for exposing region
When the total quantity ratio in commodity region is less than the second proportion threshold value pre-seted, determine the target image for qualified images.
Preferably, further include following steps before step S1 and step S2:
Step M1: obtaining multiple training images for scene classification, schemes to each training for scene classification
Classification mark as carrying out application scenarios；
Step M2: the scene classification model is established using the training image after mark application scenarios；
Step M3: the target image is inputted into the scene classification model and carries out scene classification, determines each mesh
The scene type of logo image, and then step S1 to step S3 is executed to the target image in each scene type.
Preferably, when the target image is refrigerator image, described image classifying content model is used for will be following any
Kind appoints an a variety of optical sievings to fall:
The image of refrigerator excalation；
The image that refrigerator doors are closed；
There are two the images of refrigerator for tool；
Refrigerator region occupied area ratio in the target image is less than the image of the third proportion threshold value pre-seted.
Preferably, the step S101 includes the following steps:
Step S1011: carrying out fuzzy region detection to the target image by the fuzzy region detection model pre-seted,
Fuzzy region out of focus in the target image is identified with operation fuzzy region；
Step S1012: fuzzy region out of focus in the target image, motion blur region and normal region are mapped
FUZZY MAPPING figure is generated, and fuzzy region out of focus is indicated using the first color region, the second color region indicates motion blur area
Domain, third color region indicate normal region；
Step S1013: cumulative first color region and the second color region generate the area of the fuzzy region.
Preferably, the step S202 includes the following steps
Step S2021: by the grey level histogram in each commodity region base corresponding with the commodity region pre-seted
Quasi- grey level histogram compares；
Step S2022: the grey level histogram and the benchmark grey level histogram are divided into according to gray value sequence respectively
First gray areas and the second gray areas, and pixel accounts for the total pixel number in the commodity region in determining first gray areas
The first ratio and the second gray areas in pixel account for the second ratio of the total pixel number；
Step S2023: it compares by first ratio and second ratio and with the ratio section pre-seted, works as institute
State the first ratio and when second ratio is located in the ratio section, then assert the commodity region for qualified region, when
When first ratio is lower than in the ratio interval limit, then assert that the commodity region is overexposure region, when described the
When two ratios are greater than in the upper limit of the ratio section, then assert that the commodity region is to owe to expose region.
Picture quality automatic checkout system provided by the invention, for realizing figure described in any one of claims 1 to 7
As quality automatic detection method characterized by comprising
Fuzzy detection module determines the fuzzy region in the target image, when the mould for obtaining target image
Paste region in the target image occupied area ratio be less than pre-set the first proportion threshold value when, determine the target image
For qualified images；
Exposure tests module, each commodity region of the target image out, judges that the commodity region is for identification
No overexposure region or under-exposure region, when the overexposure region and/or under-exposure region account for all commodity regions and be less than in advance
When the second proportion threshold value being arranged, determine the target image for qualified images；
Content, classification module will successively pass through fuzzy detection module, in exposure tests module or successively by exposure tests
The qualified images that module, fuzzy detection module filter out pass through the picture material disaggregated model pre-seted and filter out up-to-standard figure
Picture.
Picture quality automatic checkout equipment provided by the invention, comprising:
Processor；
Memory, wherein being stored with the executable instruction of the processor；
Wherein, the processor is configured to detect automatically via the execution executable instruction to execute described image quality
The step of method.
Computer readable storage medium provided by the invention, for storing program, described program is performed described in realization
The step of picture quality automatic testing method.
Compared with prior art, the present invention have it is following the utility model has the advantages that
The present invention can sieve target image according to fuzzy region, exposure area and picture material in target image
Choosing, filters out up-to-standard image, so as to convenient for into accurately identifying, and may remind the user that the commodity in target image
Retake is carried out to object off quality；The present invention can be collected when user carries out and patrols shop in photo filter out it is up-to-standard
After image, and then accurately identify and generating commodity data report for commodity in up-to-standard image can be carried out, realized to quotient
The effective monitoring of commodity situation in.
Detailed description of the invention
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon:
Fig. 1 is the step flow chart of picture quality automatic testing method in the present invention；
Fig. 2 is the step flow chart for carrying out scene classification in the present invention to target object；
Fig. 3 is the step flow chart that the fuzzy region in target image is detected and screened in the present invention；
Fig. 4 is the step flow chart detected in the present invention to the fuzzy region in target image；
Fig. 5 is the step flow chart that the exposure area in target image is detected and screened in the present invention；
Fig. 6 is the step flow chart detected in the present invention to the exposure area in target image；
Fig. 7 is picture quality automatic detection module schematic diagram in the present invention；
Fig. 8 is the structural schematic diagram of picture quality automatic checkout equipment in the present invention；And
Fig. 9 is the structural schematic diagram of computer readable storage medium in the present invention.
Specific embodiment
The present invention is described in detail combined with specific embodiments below.Following embodiment will be helpful to the technology of this field
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill of this field
For personnel, without departing from the inventive concept of the premise, various modifications and improvements can be made.These belong to the present invention
Protection scope.
The specific application scenarios of the present invention can be, when user passes through intelligent terminal, such as smart phone, to shelf or refrigerator
On end article be acquired, when carrying out commodity data report making, can using the present invention to the target image of shooting only
Detection, implementing the feedback target image, whether fuzzy region area is excessive, if there are overexposure region and/or under-exposure areas
Domain is excessive and whether picture material meets default rule, so as to remind user to carry out underproof target image
Even if retake, so as to guarantee that the image of acquisition is up-to-standard image, can prepare to identify target in image recognition
Commodity in image generate more perfect data sheet.
Fig. 1 is the step flow chart of picture quality automatic testing method in the present invention, as shown in Figure 1, provided by the invention
Picture quality automatic testing method, includes the following steps:
Step S1: target image is obtained, the fuzzy region in the target image is determined, when the fuzzy region is in institute
When stating occupied area ratio in target image and being less than the first proportion threshold value pre-seted, determine the target image for qualification figure
Picture；
Step S2: identifying each commodity region of the target image, judges whether the commodity region is overexposure
Region or under-exposure region are less than the pre-seted when the overexposure region and/or under-exposure region account for all commodity regions
When two proportion threshold values, determine the target image for qualified images；
Step S3: it will successively pass through in step S1, step S2 or successively by the qualification that step S2, step S1 are filtered out and scheme
As the picture material disaggregated model by pre-seting filters out up-to-standard image.
In the present embodiment, first proportion threshold value is set as 0.55, and first proportion threshold value is set as 0.6.
In the present embodiment, the present invention can be according to fuzzy region, exposure area and picture material in target image to mesh
Logo image is screened, and up-to-standard image is filtered out, so as to convenient for the commodity in target image into accurately identifying, and
It may remind the user that and retake is carried out to object off quality；The present invention can be collected in photo when user carries out and patrols shop and be sieved
After selecting up-to-standard image, and then accurately identify and generating commodity data report for commodity in up-to-standard image can be carried out
Table realizes the effective monitoring to commodity situation in market.
When the target image is refrigerator image, described image classifying content model is used to be following any or appoint more
Kind optical sieving falls:
The image of refrigerator excalation；
The image that refrigerator doors are closed；
There are two the images of refrigerator for tool；
Refrigerator region occupied area ratio in the target image is less than the image of the third proportion threshold value pre-seted.
Described image classifying content model uses convolutional neural networks structure, is trained under deep learning frame
It arrives.In the present embodiment, the third proportion threshold value is 60%.
Fig. 2 is the step flow chart for carrying out scene classification in the present invention to target object, as shown in Fig. 2, in step S1 and
Further include following steps before step S2:
Step M1: obtaining multiple training images for scene classification, schemes to each training for scene classification
Classification mark as carrying out application scenarios；
Step M2: the scene classification model is established using the training image after mark application scenarios；
Step M3: the target image is inputted into the scene classification model and carries out scene classification, determines each mesh
The scene type of logo image, and then step S1 to step S3 is executed to the target image in each scene type.
In the present embodiment, the present invention first classifies to the target image, consequently facilitating passing through picture material
Disaggregated model carries out the screening of picture material, filters out up-to-standard image.
In the present embodiment, the classification of the application scenarios includes following any or appoints plurality of application scenes:
Refrigerator；
Shelf；
Ground pushes away commodity；
It sets a table；
Pendant frame.
Fig. 3 is the step flow chart that the fuzzy region in target image is detected and screened in the present invention, such as Fig. 3 institute
Show, the step S1 includes the following steps:
Step S101: obtaining the target image, carries out fuzzy region detection to the target image, determines the target
Fuzzy region in image, and then determine the area of the fuzzy region；
Step S102: the area of all fuzzy regions is gone out in the target image according to the areal calculation of the fuzzy region
Middle occupied area ratio；
Step S103: determining whether the occupied area ratio is less than first proportion threshold value pre-seted, and is working as
When the occupied area ratio is less than the first proportion threshold value pre-seted, determine the target image for qualified images.
In the present embodiment, to the target image carry out fuzzy region detection when, using fuzzy region detection model into
Row detection determines that the paste region detection model uses convolutional neural networks structure, is trained under deep learning frame
It arrives.
Fig. 4 is the step flow chart detected in the present invention to the fuzzy region in target image, as shown in figure 4, institute
Step S101 is stated to include the following steps:
Step S1011: carrying out fuzzy region detection to the target image by the fuzzy region detection model pre-seted,
Fuzzy region out of focus in the target image is identified with operation fuzzy region；
Step S1012: fuzzy region out of focus in the target image, motion blur region and normal region are mapped
FUZZY MAPPING figure is generated, and fuzzy region out of focus is indicated using the first color region, the second color region indicates motion blur area
Domain, third color region indicate normal region；
Step S1013: cumulative first color region and the second color region generate the area of the fuzzy region.
In the present embodiment, fuzzy region is divided into fuzzy region out of focus in the present invention and operation fuzzy region is distinguished
Identification, and it is converted to different color regions, to improve the computational efficiency of fuzzy region area.
Fig. 5 is the step flow chart that the exposure area in target image is detected and screened in the present invention, such as Fig. 5 institute
Show, the step S2 includes the following steps:
Step S201: identifying each commodity region of the target image, and generates the ash in each commodity region
Spend histogram；
Step S202: by the grey level histogram in each commodity region benchmark corresponding with the commodity region pre-seted
Grey level histogram compares, and the commodity region is divided into overexposure region, owes to expose region and qualified region；
Step S203: it when the exposure area in the target image and/or owes in quantity and the target image for exposing region
When the total quantity ratio in commodity region is less than the second proportion threshold value pre-seted, determine the target image for qualified images.
In the present embodiment, the present invention is by each commodity region according to grey level histogram and benchmark grey level histogram
Comparison commodity region is divided into overexposure region, owes to expose region and qualified region, consequently facilitating the system of unqualified image-region
Meter, convenient for not conforming to the exclusion of table images.
Fig. 6 is the step flow chart detected in the present invention to the exposure area in target image, as shown in fig. 6, institute
Step S202 is stated to include the following steps:
Step S2021: by the grey level histogram in each commodity region base corresponding with the commodity region pre-seted
Quasi- grey level histogram compares；
Step S2022: the grey level histogram and the benchmark grey level histogram are divided into according to gray value sequence respectively
First gray areas and the second gray areas, and pixel accounts for the total pixel number in the commodity region in determining first gray areas
The first ratio and the second gray areas in pixel account for the second ratio of the total pixel number；
In the present embodiment, first gray areas is 0 to 127 region of gray value, and the second gray areas is gray value
128 to 255 regions
In variation, can the first ratio to grey level histogram and the second ratio optimize, as described in can set
First ratio of grey level histogram is a, and the second ratio is b, and first ratio of benchmark grey level histogram is A, and the second ratio is
B；
First suboptimization is carried out to a and generates m, m=a+0.5b, the first suboptimization is carried out to b and generates n, n=a+0.5b, to a
It carries out the second suboptimization and generates x,Second suboptimization is carried out to b and generates y,
Step S2023: it compares by first ratio and second ratio and with the ratio section pre-seted, works as institute
State the first ratio and when second ratio is located in the ratio section, then assert the commodity region for qualified region, when
When first ratio is lower than in the ratio interval limit, then assert that the commodity region is overexposure region, when described the
When two ratios are greater than in the upper limit of the ratio section, then assert that the commodity region is to owe to expose region.Other types regard as it
His region.
In the present embodiment, the ratio section is [0.5,0.7].
It can be using x and having compared with the ratio section pre-seted, as the x and y described in variation
When in the ratio section, then the commodity region is assert for qualified region, when the x is lower than in the ratio interval limit
When, then assert that the commodity region is overexposure region, when the y is greater than in the upper limit of the ratio section, then described in identification
Commodity region is to owe to expose region.
Fig. 7 is picture quality automatic detection module schematic diagram in the present invention, as shown in fig. 7, image matter provided by the invention
Automatic checkout system is measured, for realizing the picture quality automatic testing method, comprising:
Fuzzy detection module determines the fuzzy region in the target image, when the mould for obtaining target image
Paste region in the target image occupied area ratio be less than pre-set the first proportion threshold value when, determine the target image
For qualified images；
Exposure tests module, each commodity region of the target image out, judges that the commodity region is for identification
No overexposure region or under-exposure region, when the overexposure region and/or under-exposure region account for all commodity regions and be less than in advance
When the second proportion threshold value being arranged, determine the target image for qualified images；
Content, classification module will successively pass through fuzzy detection module, in exposure tests module or successively by exposure tests
The qualified images that module, fuzzy detection module filter out pass through the picture material disaggregated model pre-seted and filter out up-to-standard figure
Picture.
A kind of picture quality automatic checkout equipment, including processor are also provided in the embodiment of the present invention.Memory, wherein depositing
Contain the executable instruction of processor.Wherein, processor is configured to automatic to execute picture quality via executable instruction is executed
The step of detection.
It as above, can be according to fuzzy region, exposure area and picture material in target image to target figure in the embodiment
As being screened, filter out up-to-standard image, so as to convenient for the commodity in target image into accurately identifying, and can be with
User is reminded to carry out retake to object off quality.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, it may be assumed that complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as " circuit ", " module " or " platform ".
Fig. 8 is the structural schematic diagram of the online shopping ancillary equipment in the present invention based on augmented reality.It is described referring to Fig. 8
The electronic equipment 600 of this embodiment according to the present invention.The electronic equipment 600 that Fig. 8 is shown is only an example, is not answered
Any restrictions are brought to the function and use scope of the embodiment of the present invention.
As shown in figure 8, electronic equipment 600 is showed in the form of universal computing device.The component of electronic equipment 600 can wrap
Include but be not limited to: at least one processing unit 610, at least one storage unit 620, connection different platform component (including storage
Unit 620 and processing unit 610) bus 630, display unit 640 etc..
Wherein, storage unit is stored with program code, and program code can be executed with unit 610 processed, so that processing is single
Member 610 executes various exemplary implementations according to the present invention described in this specification above-mentioned electronic prescription circulation processing method part
The step of mode.For example, processing unit 610 can execute step as shown in fig. 1.
Storage unit 620 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 6201 and/or cache memory unit 6202, it can further include read-only memory unit (ROM) 6203.
Storage unit 620 can also include program/utility with one group of (at least one) program module 6205
6204, such program module 6205 includes but is not limited to: operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.
Bus 630 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 600 can also be with one or more external equipments 700 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 600 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 600 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 650.Also, electronic equipment 600 can be with
By network adapter 660 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.Network adapter 660 can be communicated by bus 630 with other modules of electronic equipment 600.It should
Understand, although being not shown in Fig. 8, other hardware and/or software module can be used in conjunction with electronic equipment 600, including unlimited
In: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and number
According to backup storage platform etc..
A kind of computer readable storage medium is also provided in the embodiment of the present invention, for storing program, program is performed
The step of image split-joint method of realization.In some possible embodiments, various aspects of the invention are also implemented as
A kind of form of program product comprising program code, when program product is run on the terminal device, program code is for making
Terminal device executes various exemplary according to the present invention described in this specification above-mentioned electronic prescription circulation processing method part
The step of embodiment.
As it appears from the above, the program of the computer readable storage medium of the embodiment is when being executed, it can be according to target image
Middle fuzzy region, exposure area and picture material screen target image, up-to-standard image are filtered out, so as to just
In into accurately identifying, and may remind the user that the commodity in target image and carry out retake to object off quality.
Fig. 9 is the structural schematic diagram of computer readable storage medium of the invention.Refering to what is shown in Fig. 9, describing according to this
The program product 800 for realizing the above method of the embodiment of invention can use the read-only storage of portable compact disc
Device (CD-ROM) and including program code, and can be run on terminal device, such as PC.However, journey of the invention
Sequence product is without being limited thereto, and in this document, readable storage medium storing program for executing can be any tangible medium for including or store program, the journey
Sequence can be commanded execution system, device or device use or in connection.
Program product can be using any combination of one or more readable mediums.Readable medium can be readable signal Jie
Matter or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or partly lead
System, device or the device of body, or any above combination.More specific example (the non exhaustive column of readable storage medium storing program for executing
Table) it include: the electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only storage
Device (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer readable storage medium may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable storage medium storing program for executing can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.The program code for including on readable storage medium storing program for executing can transmit with any suitable medium, including but not
It is limited to wireless, wired, optical cable, RF etc. or above-mentioned any appropriate combination.
The program for executing operation of the present invention can be write with any combination of one or more programming languages
Code, programming language include object oriented program language-Java, C++ etc., further include conventional process
Formula programming language-such as " C " language or similar programming language.Program code can be calculated fully in user
It executes in equipment, partly execute on a user device, executing, as an independent software package partially in user calculating equipment
Upper part executes on a remote computing or executes in remote computing device or server completely.It is being related to remotely counting
In the situation for calculating equipment, remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
In the present embodiment, the present invention can be according to fuzzy region, exposure area and picture material in target image to mesh
Logo image is screened, and up-to-standard image is filtered out, so as to convenient for the commodity in target image into accurately identifying, and
It may remind the user that and retake is carried out to object off quality；The present invention can be collected in photo when user carries out and patrols shop and be sieved
After selecting up-to-standard image, and then accurately identify and generating commodity data report for commodity in up-to-standard image can be carried out
Table realizes the effective monitoring to commodity situation in market.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited to above-mentioned
Particular implementation, those skilled in the art can make various deformations or amendments within the scope of the claims, this not shadow
Ring substantive content of the invention.
Claims (10)
1. a kind of picture quality automatic testing method, which comprises the steps of:
Step S1: target image is obtained, the fuzzy region in the target image is determined, when the fuzzy region is in the mesh
When occupied area ratio is less than the first proportion threshold value pre-seted in logo image, determine the target image for qualified images；
Step S2: identifying each commodity region of the target image, judges whether the commodity region is overexposure region
Or under-exposure region, when the overexposure region and/or under-exposure region account for all commodity regions and be less than the second ratio pre-seted
When example threshold value, determine the target image for qualified images；
Step S3: lead to successively passing through in step S1, step S2 or successively by the qualified images that step S2, step S1 are filtered out
It crosses the picture material disaggregated model pre-seted and filters out up-to-standard image.
2. picture quality automatic testing method according to claim 1, which is characterized in that the step S1 includes following step
It is rapid:
Step S101: obtaining the target image, carries out fuzzy region detection to the target image, determines the target image
In fuzzy region, and then determine the area of the fuzzy region；
Step S102: the areas of all fuzzy regions institute in the target image is gone out according to the areal calculation of the fuzzy region
Account for area ratio；
Step S103: determining whether the occupied area ratio is less than first proportion threshold value pre-seted, and when described
When occupied area ratio is less than the first proportion threshold value pre-seted, determine the target image for qualified images.
3. picture quality automatic testing method according to claim 1, which is characterized in that the step S2 includes following step
It is rapid:
Step S201: identifying each commodity region of the target image, and the gray scale for generating each commodity region is straight
Fang Tu；
Step S202: by the grey level histogram in each commodity region benchmark gray scale corresponding with the commodity region pre-seted
Histogram compares, and the commodity region is divided into overexposure region, owes to expose region and qualified region；
Step S203: when the exposure area in the target image and/or commodity in the quantity and the target image that expose region are owed
When the total quantity ratio in region is less than the second proportion threshold value pre-seted, determine the target image for qualified images.
4. picture quality automatic testing method according to claim 1, which is characterized in that before step S1 and step S2
Further include following steps:
Step M1: obtaining multiple training images for scene classification, to each training image for scene classification into
The classification of row application scenarios marks；
Step M2: the scene classification model is established using the training image after mark application scenarios；
Step M3: the target image is inputted into the scene classification model and carries out scene classification, determines each target figure
The scene type of picture, and then step S1 to step S3 is executed to the target image in each scene type.
5. picture quality automatic testing method according to claim 1, which is characterized in that when the target image is refrigerator
When image, described image classifying content model is used to fall following any or a variety of optical sievings:
The image of refrigerator excalation；
The image that refrigerator doors are closed；
There are two the images of refrigerator for tool；
Refrigerator region occupied area ratio in the target image is less than the image of the third proportion threshold value pre-seted.
6. picture quality automatic testing method according to claim 2, which is characterized in that the step S101 includes as follows
Step:
Step S1011: fuzzy region detection is carried out to the target image by the fuzzy region detection model pre-seted, by institute
It states the fuzzy region out of focus in target image and operation fuzzy region identifies；
Step S1012: fuzzy region out of focus in the target image, motion blur region and normal region are subjected to mapping generation
FUZZY MAPPING figure, and fuzzy region out of focus is indicated using the first color region, the second color region indicates motion blur region, the
Three color regions indicate normal region；
Step S1013: cumulative first color region and the second color region generate the area of the fuzzy region.
7. picture quality automatic testing method according to claim 3, which is characterized in that the step S202 includes as follows
Step
Step S2021: by the grey level histogram in each commodity region benchmark ash corresponding with the commodity region pre-seted
Degree histogram compares；
Step S2022: the grey level histogram and the benchmark grey level histogram are divided into first according to gray value sequence respectively
Gray areas and the second gray areas, and determine that pixel in the first gray areas accounts for the of the total pixel number in the commodity region
Pixel accounts for the second ratio of the total pixel number in one ratio and the second gray areas；
Step S2023: comparing by first ratio and second ratio and with the ratio section pre-seted, when described
When one ratio and second ratio are located in the ratio section, then the commodity region is assert for qualified region, when described
When first ratio is lower than in the ratio interval limit, then assert that the commodity region is overexposure region, when second ratio
When example is greater than in the upper limit of the ratio section, then assert that the commodity region is to owe to expose region.
8. a kind of picture quality automatic checkout system, automatic for realizing picture quality described in any one of claims 1 to 7
Detection method characterized by comprising
Fuzzy detection module determines the fuzzy region in the target image, when the confusion region for obtaining target image
Domain in the target image occupied area ratio be less than pre-set the first proportion threshold value when, determine the target image for close
Table images；
Exposure tests module, each commodity region of the target image out for identification, judge the commodity region whether mistake
Exposure area or under-exposure region are pre-seted when the overexposure region and/or under-exposure region account for all commodity regions and be less than
The second proportion threshold value when, determine the target image for qualified images；
Content, classification module, will successively pass through fuzzy detection module, in exposure tests module or successively by exposure tests module,
The qualified images that fuzzy detection module filters out pass through the picture material disaggregated model pre-seted and filter out up-to-standard image.
9. a kind of picture quality automatic checkout equipment characterized by comprising
Processor；
Memory, wherein being stored with the executable instruction of the processor；
Wherein, the processor is configured to come any one of perform claim requirement 1 to 7 institute via the execution executable instruction
The step of stating picture quality automatic testing method.
10. a kind of computer readable storage medium, for storing program, which is characterized in that described program is performed realization power
Benefit requires the step of any one of 1 to 7 described image quality automatic detection method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910162123.8A CN109978833A (en) | 2019-03-05 | 2019-03-05 | Picture quality automatic testing method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910162123.8A CN109978833A (en) | 2019-03-05 | 2019-03-05 | Picture quality automatic testing method, system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109978833A true CN109978833A (en) | 2019-07-05 |
Family
ID=67077924
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910162123.8A Pending CN109978833A (en) | 2019-03-05 | 2019-03-05 | Picture quality automatic testing method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109978833A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110415225A (en) * | 2019-07-22 | 2019-11-05 | 南充折衍智能光电科技有限公司 | A kind of finger vein image quality evaluation method based on multi-information fusion |
| CN111191054A (en) * | 2019-12-18 | 2020-05-22 | 腾讯科技（深圳）有限公司 | Recommendation method and device for media data |
| CN111539962A (en) * | 2020-01-10 | 2020-08-14 | 济南浪潮高新科技投资发展有限公司 | Target image classification method, device and medium |
| CN112329522A (en) * | 2020-09-24 | 2021-02-05 | 上海品览数据科技有限公司 | Goods shelf goods fuzzy detection method based on deep learning and image processing |
| CN112954229A (en) * | 2021-02-08 | 2021-06-11 | 青岛海尔电冰箱有限公司 | Method and equipment for adjusting light intensity of light supplementing lamp based on gray value and refrigerator |
| CN114264657A (en) * | 2020-09-16 | 2022-04-01 | 南亚科技股份有限公司 | Wafer inspection method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101399924A (en) * | 2007-09-25 | 2009-04-01 | 展讯通信（上海）有限公司 | Automatic exposure method and device based on brightness histogram |
| CN101448174A (en) * | 2008-12-26 | 2009-06-03 | 深圳华为通信技术有限公司 | Image quality evaluation device and method thereof |
| CN101685542A (en) * | 2008-09-24 | 2010-03-31 | 索尼株式会社 | Electronic device, fuzzy image sorting method and program |
| JP2012124652A (en) * | 2010-12-07 | 2012-06-28 | Ricoh Co Ltd | Imaging apparatus and image processing method |
| CN103218778A (en) * | 2013-03-22 | 2013-07-24 | 华为技术有限公司 | Image and video processing method and device |
| CN107945156A (en) * | 2017-11-14 | 2018-04-20 | 宁波江丰生物信息技术有限公司 | A kind of method of automatic Evaluation numeral pathology scan image image quality |
| CN108734162A (en) * | 2018-04-12 | 2018-11-02 | 上海扩博智能技术有限公司 | Target identification method, system, equipment and storage medium in commodity image |
| CN108986075A (en) * | 2018-06-13 | 2018-12-11 | 浙江大华技术股份有限公司 | A kind of judgment method and device of preferred image |
| TWI645373B (en) * | 2017-11-13 | 2018-12-21 | 瑞昱半導體股份有限公司 | Auto exposure control method and electronic device using the same |
-
2019
- 2019-03-05 CN CN201910162123.8A patent/CN109978833A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101399924A (en) * | 2007-09-25 | 2009-04-01 | 展讯通信（上海）有限公司 | Automatic exposure method and device based on brightness histogram |
| CN101685542A (en) * | 2008-09-24 | 2010-03-31 | 索尼株式会社 | Electronic device, fuzzy image sorting method and program |
| CN101448174A (en) * | 2008-12-26 | 2009-06-03 | 深圳华为通信技术有限公司 | Image quality evaluation device and method thereof |
| JP2012124652A (en) * | 2010-12-07 | 2012-06-28 | Ricoh Co Ltd | Imaging apparatus and image processing method |
| CN103218778A (en) * | 2013-03-22 | 2013-07-24 | 华为技术有限公司 | Image and video processing method and device |
| TWI645373B (en) * | 2017-11-13 | 2018-12-21 | 瑞昱半導體股份有限公司 | Auto exposure control method and electronic device using the same |
| CN107945156A (en) * | 2017-11-14 | 2018-04-20 | 宁波江丰生物信息技术有限公司 | A kind of method of automatic Evaluation numeral pathology scan image image quality |
| CN108734162A (en) * | 2018-04-12 | 2018-11-02 | 上海扩博智能技术有限公司 | Target identification method, system, equipment and storage medium in commodity image |
| CN108986075A (en) * | 2018-06-13 | 2018-12-11 | 浙江大华技术股份有限公司 | A kind of judgment method and device of preferred image |
Non-Patent Citations (5)
| Title |
|---|
| HOJATOLLAH YEGANEH 等，: "Objective Quality Assessment of Tone-Mapped Images", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| 刘望明，: "基于图像质量评价的降低航片冗余度研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 方临明，: "《公关魅影 职场和生活必知的摄影秘决》", 31 January 2018, 浙江摄影出版社 * |
| 朱学芳 等，: "《计算机图像处理导论》", 30 June 2003, 科学技术文献出版社 * |
| 王艳红 等，: "《影视传播实验教学理论探索与实践创新》", 30 June 2011, 上海三联书店 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110415225A (en) * | 2019-07-22 | 2019-11-05 | 南充折衍智能光电科技有限公司 | A kind of finger vein image quality evaluation method based on multi-information fusion |
| CN111191054A (en) * | 2019-12-18 | 2020-05-22 | 腾讯科技（深圳）有限公司 | Recommendation method and device for media data |
| CN111191054B (en) * | 2019-12-18 | 2024-02-13 | 腾讯科技（深圳）有限公司 | Media data recommendation method and device |
| CN111539962A (en) * | 2020-01-10 | 2020-08-14 | 济南浪潮高新科技投资发展有限公司 | Target image classification method, device and medium |
| CN114264657A (en) * | 2020-09-16 | 2022-04-01 | 南亚科技股份有限公司 | Wafer inspection method and system |
| CN112329522A (en) * | 2020-09-24 | 2021-02-05 | 上海品览数据科技有限公司 | Goods shelf goods fuzzy detection method based on deep learning and image processing |
| CN112954229A (en) * | 2021-02-08 | 2021-06-11 | 青岛海尔电冰箱有限公司 | Method and equipment for adjusting light intensity of light supplementing lamp based on gray value and refrigerator |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109978833A (en) | Picture quality automatic testing method, system, equipment and storage medium | |
| CN108734162B (en) | Method, system, equipment and storage medium for identifying target in commodity image | |
| CN106650662B (en) | Target object shielding detection method and device | |
| CN111723786A (en) | Method and device for detecting wearing of safety helmet based on single model prediction | |
| CN111325258A (en) | Characteristic information acquisition method, device, equipment and storage medium | |
| CN106355367A (en) | Warehouse monitoring management device | |
| Cheng et al. | Smoke detection and trend prediction method based on Deeplabv3+ and generative adversarial network | |
| CN108280455A (en) | Human body critical point detection method and apparatus, electronic equipment, program and medium | |
| CN116579616B (en) | Risk identification method based on deep learning | |
| CN108198130A (en) | Image processing method, device, storage medium and electronic equipment | |
| CN108205680A (en) | Image characteristics extraction integrated circuit, method, terminal | |
| CN108389230A (en) | Refrigerator capacity automatic testing method, system, equipment and storage medium | |
| CN109618149A (en) | Test method, device and the smart machine of equipment image quality | |
| CN110443317A (en) | A kind of method, apparatus and electronic equipment of paper shelves electronic data processing | |
| CN103946887A (en) | Gaze position estimation system, control method for gaze position estimation system, gaze position estimation device, control method for gaze position estimation device, program, and information storage medium | |
| CN114255377A (en) | Differential commodity detection and classification method for intelligent container | |
| CN110505397A (en) | The method, apparatus and computer storage medium of camera selection | |
| CN106529455A (en) | Fast human posture recognition method based on SoC FPGA | |
| Rana et al. | Learning-based adaptive tone mapping for keypoint detection | |
| CN111797867A (en) | System resource optimization method and device, storage medium and electronic equipment | |
| CN106874835B (en) | A kind of image processing method and device | |
| Wang et al. | Smoking behavior detection algorithm based on YOLOv8-MNC | |
| CN108776959A (en) | Image processing method, device and terminal device | |
| CN108446693A (en) | Convenient labels method, system, equipment and the storage medium of target to be identified | |
| CN114360053A (en) | Action recognition method, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190705 |
|
| RJ01 | Rejection of invention patent application after publication |