CN109724993A - Detection method, device and the storage medium of the degree of image recognition apparatus - Google Patents
Detection method, device and the storage medium of the degree of image recognition apparatus Download PDFInfo
- Publication number
- CN109724993A CN109724993A CN201811614985.1A CN201811614985A CN109724993A CN 109724993 A CN109724993 A CN 109724993A CN 201811614985 A CN201811614985 A CN 201811614985A CN 109724993 A CN109724993 A CN 109724993A
- Authority
- CN
- China
- Prior art keywords
- target
- degree
- image
- recognition apparatus
- image recognition
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of detection method of the degree of image recognition apparatus, device and storage mediums.This method comprises: obtaining the target image that image recognition apparatus shoots target object；Obtain the target sharpness of target image；In the case where target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus is goal gradient.Through the invention, the efficiency for improving and being detected to the degree of image recognition apparatus has been reached.
Description
Technical field
The present invention relates to field of image processings, in particular to a kind of detection of the degree of image recognition apparatus
Method, apparatus and storage medium.
Background technique
Currently, image recognition apparatus using more and more common.After image recognition apparatus is using a period of time, identification
Effect will receive the influence of indoor and outdoor surroundings, for example, under air-tight moist, the environment that dust stratification is serious, greasy dirt is dense, image recognition
The picture that equipment is identified is unintelligible, image recognition accuracy declines, it is also possible to occur causing due to the blocking of equipment cooling mouth
Equipment damage situations such as.If not carrying out maintenance and repair, the dirt of image recognition apparatus to image recognition apparatus in time
Damage degree will be aggravated constantly.
The measurement of the degree of image recognition apparatus is completely dependent on artificial subjective judgement, generallys use the side of regular visit
Formula sets the period of artificial regular visit, lacks the quantizating index of objective standard as reference, this demand cost to manpower compared with
Greatly, it is likely to result in response lag, for example, just fed back when the degree of image recognition apparatus is quite serious,
Cause the extent of damage of equipment to increase, it is also possible to lead to overreact, for example, in the case where degree is slight frequently into
Row maintenance, to generate meaningless human cost and fund cost.
The technical issues of for the low efficiency detected in the prior art to the degree of image recognition apparatus, at present
Not yet put forward effective solutions.
Summary of the invention
The main purpose of the present invention is to provide a kind of detection methods of the degree of image recognition apparatus, at least to solve
The certainly detection method of the degree of image recognition apparatus, device and storage medium.
To achieve the goals above, according to an aspect of the invention, there is provided a kind of image recognition apparatus is stained journey
The detection method of degree characterized by comprising obtain the target figure that image recognition apparatus shoots target object
Picture；Obtain the target sharpness of target image；In the case where target sharpness is within the scope of targets threshold, image recognition is determined
The degree of equipment is goal gradient.
Optionally, in the case where target sharpness is within the scope of targets threshold, determine image recognition apparatus is stained journey
Degree is that goal gradient includes: to determine the degree of image recognition apparatus in the case where target sharpness is greater than first threshold
For the first sub- grade；In the case where target sharpness is greater than second threshold and is less than or equal to first threshold, image recognition is determined
The degree of equipment is the second sub- grade, wherein the degree of the second sub- grade is greater than the degree of the first sub- grade；
In the case where target sharpness is less than or equal to second threshold, determine that the degree of image recognition apparatus is the sub- grade of third,
Wherein, the degree of the sub- grade of third is greater than the degree of the second sub- grade.
Optionally, after the degree for determining image recognition apparatus is the second sub- grade, this method further include: output
First prompt information, wherein the degree that the first prompt information is used to indicate image recognition apparatus is the second sub- grade, and/
Or image recognition apparatus needs to arrange maintenance；After the degree for determining image recognition apparatus is the sub- grade of third, method
Further include: the second prompt information of output, wherein the degree that the second prompt information is used to indicate image recognition apparatus is third
Sub- grade and/or image recognition apparatus need to safeguard immediately.
Optionally, the first prompt information of output includes: to export the first prompt information to target terminal；The second prompt of output
Information includes: to export the second prompt information to target terminal.
Optionally, it when exporting the first prompt information to target terminal, or exports to target by the second prompt information
When terminal, this method further include: the image data of target image is sent to target terminal.
Optionally, obtaining the target image that image recognition apparatus shoots target object includes: in the first mesh
The mark moment obtains the first object image that image recognition apparatus shoots target object；Obtain the target of target image
Clarity includes: to obtain the first object clarity of first object image；In feelings of the target sharpness within the scope of targets threshold
Under condition, it includes: in first object clarity in targets threshold range that the degree for determining image recognition apparatus, which is goal gradient,
In the case where interior, determine that the degree of image recognition apparatus is first object grade；Image recognition apparatus is obtained to target pair
As the target image shot further include: step 1 obtains image recognition apparatus to target pair in the second object time
As the second target image shot, wherein the second object time and first object moment are separated by the first object time
Section；Obtain the target sharpness of target image further include: step 2 obtains the second target sharpness of the second target image；?
In the case that target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus includes: for goal gradient
Step 3 determines that the degree of image recognition apparatus is in the case where the second target sharpness is within the scope of targets threshold
Second goal gradient；Wherein, method further include: the first mesh will be separated by after the second object time and with the second object time
The third object time of mark period is determined as the second object time, return step one, step 2 and step 3.
Optionally, after the degree for determining image recognition apparatus is the second goal gradient, this method further include: extremely
It is few that first object grade and the second goal gradient are handled, obtain objective result, wherein objective result is used to indicate image
Identify the variation tendency of the degree of equipment.
Optionally, obtaining the target image that image recognition apparatus shoots target object includes: in the second mesh
The different moments in the period are marked, multiple target images that image recognition apparatus shoots target object are obtained；It obtains
The target sharpness for taking target image includes: the target sharpness for obtaining every target image, obtains multiple target sharpness；?
In the case that target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus includes: for goal gradient
In the case where being greater than the target sharpness of destination number within the scope of targets threshold, the degree of image recognition apparatus is determined
For goal gradient.
To achieve the goals above, according to another aspect of the present invention, a kind of being stained for image recognition apparatus is additionally provided
The detection device of degree characterized by comprising first acquisition unit, for obtain image recognition apparatus to target object into
The target image that row shooting obtains；Second acquisition unit, for obtaining the target sharpness of target image；Determination unit is used for
In the case where target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus is goal gradient.
To achieve the goals above, according to another aspect of the present invention, a kind of storage medium is additionally provided.The storage medium
Program including storage, wherein the image that equipment where control storage medium executes the embodiment of the present invention in program operation is known
The detection method of the degree of other equipment.
To achieve the goals above, according to another aspect of the present invention, a kind of processor is additionally provided.The processor is used for
Run program, wherein the detection method of the degree of the image recognition apparatus of the embodiment of the present invention is executed when program is run.
Through the invention, the target image that image recognition apparatus shoots target object is obtained；Obtain target
The target sharpness of image；In the case where target sharpness is within the scope of targets threshold, being stained for image recognition apparatus is determined
Degree is goal gradient.That is, the mesh of the target image shot by image recognition apparatus to target object
Clarity is marked, the degree of image recognition apparatus is determined based on quantifiable index, is avoided to image recognition apparatus
The judgement of degree is completely dependent on artificial subjective judgement, solves the effect detected to the degree of image recognition apparatus
The low technical problem of rate has reached the technical effect for improving the efficiency detected to the degree of image recognition apparatus.
Detailed description of the invention
The attached drawing constituted part of this application is used to provide further understanding of the present invention, schematic reality of the invention
It applies example and its explanation is used to explain the present invention, do not constitute improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is a kind of detection method of the degree of image recognition apparatus according to an embodiment of the present invention；
Fig. 2 is a kind of schematic diagram of the detection system of the degree of video monitoring equipment according to an embodiment of the present invention；
And
Fig. 3 is a kind of schematic diagram of the detection device of the degree of image recognition apparatus according to an embodiment of the present invention.
Specific embodiment
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The present invention will be described in detail below with reference to the accompanying drawings and embodiments.
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection
It encloses.
It should be noted that the description and claims of this application and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to embodiments herein described herein.In addition, term " includes " and " tool
Have " and their any deformation, it is intended that cover it is non-exclusive include, for example, containing a series of steps or units
Process, method, system, product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include without clear
Other step or units listing to Chu or intrinsic for these process, methods, product or equipment.
Embodiment 1
The embodiment of the invention provides a kind of detection methods of the degree of image recognition apparatus.
Fig. 1 is a kind of detection method of the degree of image recognition apparatus according to an embodiment of the present invention.Such as Fig. 1 institute
Show, method includes the following steps:
Step S102 obtains the target image that image recognition apparatus shoots target object.
In the technical solution that above-mentioned steps S102 of the present invention is provided, image recognition apparatus is setting for degree to be detected
It is standby, there is image identification function, for example, the image recognition apparatus can be used for for video monitoring equipment, video camera apparatus etc.
The equipment of image taking.Target object can be any object that image recognition apparatus can take in a certain region, be used for
The degree of detection image identification equipment.The target image that image recognition apparatus shoots target object is obtained,
The image data of available target image, the target image are that image recognition apparatus takes under normal operating conditions
Picture material, can be picture.After obtaining the target image that image recognition apparatus shoots target object,
The target image can be saved, for example, target image is saved in local.
Step S104 obtains the target sharpness of target image.
In the technical solution that above-mentioned steps S104 of the present invention is provided, target object is carried out obtaining image recognition apparatus
After shooting obtained target image, the target sharpness of target image is obtained.
The target sharpness of the embodiment is used to indicate the readability of target image, is measure objective image quality one
A index.Edge is the most basic feature of target image, and when target image obscures, blur margin is clear, then can pass through detection
The edge of target image determines the clarity of target image, and when target image is fuzzyyer, the gradient value of target image is also smaller.
Optionally, which can be detected when obtaining the target sharpness of target image based on the clarity of gradient
Algorithm detects the target sharpness of target image.For example, detecting mesh using based on the clarity detection method of Sobel operator
The target sharpness of logo image.Optionally, field convolutional calculation is carried out to pixel each in target image, can use 3 × 3
Direction template and target image in each pixel carry out field convolutional calculation, extract each pixel on four direction
Marginal element calculate for example, extract marginal element of each pixel respectively on 0 °, 45 °, 90 °, 135 ° of four directions
The gradient value of each pixel in target image, and the gradient value of each pixel is compared with threshold value, gradient value is big
It is determined as the marginal point of target image in the pixel of threshold value, to filter out the edge pixel point of target image.It can be by side
The gradient value of edge pixel is added, and using edge gradient energy and as the evaluation of estimate that the clarity of target image detects, can be incited somebody to action
The evaluation of estimate is determined as the target sharpness N of target image.
Optionally, which is also based on squared gradient algorithm, fast algorithm of detecting to detect the mesh of target image
Mark clarity, wherein for squared gradient algorithm by differential value square, fast detection method is that the gray scale of image is now sought mean value.Point
Not Ji Suan gray value larger and smaller than image grayscale mean value pixel mean value H and L, pass through evaluation points F=(H-L)/(H+
L) clarity of target image is detected.
When obtaining the target sharpness of target image, the evaluation effect of the clarity detection method of Sobel operator is than flat
The evaluation effect of square gradient algorithm and fast algorithm of detecting is good, and the calculating time longest of squared gradient algorithm, real-time is poor, fastly
Fast detection algorithm computation rate is high, but low to fuzzy sensitivity.
It should be noted that the algorithm of the clarity of above-mentioned detection target image is only illustrating for the embodiment of the present invention
Bright, the definition algorithm for not representing the detection target image of the embodiment of the present invention is only the clarity inspection of above-mentioned Sobel operator
The method of survey method, squared gradient algorithm and fast algorithm of detecting, any clarity that detection target image may be implemented all exists
Within the scope of the embodiment of the present invention, no longer illustrate one by one herein.
Step S106 determines being stained for image recognition apparatus in the case where target sharpness is within the scope of targets threshold
Degree is goal gradient.
In the technical solution that above-mentioned steps S106 of the present invention is provided, after the target sharpness for obtaining target image,
Judge threshold range locating for target sharpness, in the case where target sharpness is within the scope of targets threshold, determines that image is known
The degree of other equipment is goal gradient.
The targets threshold range of the embodiment is pre-set for defining the threshold of the degree of image recognition apparatus
It is worth range, goal gradient is corresponding with targets threshold range, can serve to indicate that the degree of image recognition apparatus is relatively light, dirty
Damage degree is heavier, degree is very serious etc., so that the quantizating index of objective standard is provided, to determine image recognition apparatus
Degree, the subjective judgement of the people that avoids placing one's entire reliance upon are improved and are detected to the degree of image recognition apparatus
Efficiency, and then avoid the problem that caused by response lag or overreact to the low efficiency of image recognition apparatus maintenance.
S102 to step S104 through the above steps obtains the mesh that image recognition apparatus shoots target object
Logo image；Obtain the target sharpness of target image；In the case where target sharpness is within the scope of targets threshold, image is determined
The degree for identifying equipment is goal gradient.That is, being shot to obtain to target object by image recognition apparatus
Target image target sharpness, the degree of image recognition apparatus is determined based on quantifiable index, is avoided to figure
As identifying that the judgement of the degree of equipment is completely dependent on artificial subjective judgement, solves the degree to image recognition apparatus
The technical issues of low efficiency detected, has reached and has improved the efficiency detected to the degree of image recognition apparatus
Technical effect.
As an alternative embodiment, step S106, in the case where target sharpness is within the scope of targets threshold,
It includes: in the case where target sharpness is greater than first threshold, really that the degree for determining image recognition apparatus, which is goal gradient,
The degree for determining image recognition apparatus is the first sub- grade；It is greater than second threshold in target sharpness and is less than or equal to the first threshold
In the case where value, determine that the degree of image recognition apparatus is the second sub- grade, wherein the degree of the second sub- grade is big
In the degree of the first sub- grade；In the case where target sharpness is less than or equal to second threshold, image recognition apparatus is determined
Degree be the sub- grade of third, wherein the degree of the sub- grade of third be greater than the second sub- grade degree.
In this embodiment, when determining targets threshold range, acquisition image recognition apparatus is when starting to work to target
The image that object is shot, the image can be defaulted as being taken by image recognition apparatus in entire working stage clear
The clear highest image of degree, is determined as N1 for its clarity.With the extension of image recognition apparatus working time, image recognition apparatus
The clarity of the image taken declines, and first threshold N2 and second threshold N3 can be determined based on N1, wherein first threshold is big
In second threshold, first threshold is not need to safeguard image recognition apparatus and need to safeguard image recognition apparatus arrangement
Image critical definition, second threshold N3 be need to image recognition apparatus arrangement safeguard and need immediately to image recognition
The critical definition for the image that equipment is safeguarded.Optionally, which is first threshold N2 by the first ratio-dependent of N1,
For example, the first ratio is 70%, it is second threshold N3 by the second ratio-dependent of N1, for example, the second ratio is 20%.
After the target sharpness N for obtaining target image, judge whether target sharpness N is greater than first threshold N2, such as
Fruit judges that target sharpness N is greater than first threshold N2, it is determined that the degree of image recognition apparatus is the first sub- grade, can
Degree to determine image recognition apparatus is lighter, and image recognition apparatus can work normally, and does not need specially to indicate user
It is maintained.
If it is judged that target sharpness N is not more than first threshold N2, then judge whether target sharpness is greater than the second threshold
Value N3, if it is judged that target sharpness is greater than second threshold N3, it is determined that the degree of image recognition apparatus is the second son
Grade can determine that the degree of image recognition apparatus is more serious, and at this moment image recognition apparatus needs to arrange maintenance, can be with
Instruction user is maintained.Wherein, if do not safeguarded to image recognition apparatus, image recognition apparatus can also work
For a period of time.
If it is judged that image recognition apparatus is not more than second threshold N3, it is determined that the degree of image recognition apparatus is
The sub- grade of third, at this moment the degree of image recognition apparatus is very serious, needs immediately to carry out image recognition apparatus
Maintenance, otherwise image recognition apparatus will be unable to work normally.
As an alternative embodiment, after the degree for determining image recognition apparatus is the second sub- grade,
This method further include: the first prompt information of output, wherein the first prompt information is used to indicate the degree of image recognition apparatus
Maintenance is needed to arrange for the second sub- grade and/or image recognition apparatus；It is third in the degree for determining image recognition apparatus
After sub- grade, this method further include: the second prompt information of output, wherein the second prompt information is used to indicate image recognition and sets
Standby degree is that the sub- grade of third and/or image recognition apparatus need to safeguard immediately.
In this embodiment, after the degree for determining image recognition apparatus is the second sub- grade, the can be exported
One prompt information, first prompt information can be speech prompt information, text prompt information, optical information etc., not appoint herein
What is limited.The degree that first prompt information is used to indicate image recognition apparatus is the second sub- grade and/or image recognition
Equipment needs to arrange maintenance, for example, the first prompt information is the suggestion voice of " needing to arrange maintenance ".Optionally, second son etc.
Grade needs to arrange that maintenance is corresponding with image recognition apparatus, is only used for indicating being stained for image recognition apparatus in the first prompt information
When degree is the second sub- grade, user can be confirmed that image recognition apparatus needs to be serviced.
After the degree for determining image recognition apparatus is the sub- grade of third, the second prompt information is exported, this second
Prompt information can be speech prompt information, text prompt information, optical information etc., and no limitations are hereby intended.Second prompt
The degree that information is used to indicate image recognition apparatus is that the sub- grade of third and/or image recognition apparatus need to safeguard immediately,
That is, the degree of image recognition apparatus is very serious, need immediately to safeguard image recognition apparatus, otherwise image is known
Other equipment will be unable to work normally, for example, the second prompt information is the signal language of " needing to safeguard immediately, otherwise cannot work "
Sound.Optionally, the sub- grade of third needs to safeguard immediately corresponding with image recognition apparatus, is only used for indicating in the second prompt information
When the degree of image recognition apparatus is third sub- grade, user can be confirmed that image recognition apparatus is to need to be tieed up immediately
Shield, otherwise can not work normally.
Optionally, after the degree for determining image recognition apparatus is the first sub- grade, third prompt can be exported
Information, the third prompt information can be speech prompt information, text prompt information, optical information etc., and no limitations are hereby intended.
The third prompt information be used to indicate image recognition apparatus degree be the first sub- grade and/or image recognition apparatus not
It needs to safeguard immediately.
Optionally, after the degree for determining image recognition apparatus is the first sub- grade, any mention can not be exported
Show information, the degree of default image identification equipment can make image recognition apparatus work normally, without exclusively carrying out dimension
Shield.
As an alternative embodiment, the first prompt information of output includes: to export the first prompt information to target
Terminal；Exporting the second prompt information includes: to export the second prompt information to target terminal.
In this embodiment, when exporting the first prompt information, the first prompt information can be exported to the mesh of user side
Mark terminal can export the second prompt information to the target terminal of user side when exporting the second prompt information, and the target is whole
End can be smart phone (such as Android phone, iOS mobile phone), tablet computer, palm PC and mobile internet device
The user-friendly terminal device such as (Mobile Internet Devices, referred to as MID), PAD, so that user
The degree that can know image recognition apparatus immediately has reached raising and has detected to the degree of image recognition apparatus
Efficiency effect, and then realize effective maintenance to image recognition apparatus.
As an alternative embodiment, being mentioned when exporting the first prompt information to target terminal, or by second
When showing that information is exported to target terminal, this method further include: the image data of target image is sent to target terminal.
In this embodiment, it when exporting the first prompt information to target terminal, or exports by the second prompt information
When to target terminal, in order to prevent to the erroneous judgement of the degree of image recognition apparatus, for example, due to the interference of light, it can be right
The clarity of image is influenced, and the image data of target image can be sent to target terminal, that is, by target image
Image data is back to target terminal, carries out secondary judgement for user, whether user can need in conjunction with actual conditions comprehensive descision
Related personnel is arranged to safeguard image recognition apparatus, to improve the efficiency safeguarded to image recognition apparatus machine.
As an alternative embodiment, step S102, obtains image recognition apparatus and shoot to target object
To target image include: to obtain the first mesh for being shot to target object of image recognition apparatus at the first object moment
Logo image；Step S104, obtain target image target sharpness include: obtain first object image first object it is clear
Degree；Step S106 determines that the degree of image recognition apparatus is in the case where target sharpness is within the scope of targets threshold
Goal gradient includes: to determine being stained for image recognition apparatus in the case where first object clarity is within the scope of targets threshold
Degree is first object grade；Step S102 obtains the target image that image recognition apparatus shoots target object
Further include: step 1 obtains the second target that image recognition apparatus shoots target object in the second object time
Image, wherein the second object time and first object moment are separated by the first object period；Step S104 obtains target image
Target sharpness further include: step 2, obtain the second target image the second target sharpness；Step S106 is clear in target
In the case that clear degree is within the scope of targets threshold, it includes: step 3 that the degree for determining image recognition apparatus, which is goal gradient,
In the case where the second target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus is the second target
Grade；Wherein, this method further include: the first object time will be separated by after the second object time and with the second object time
The third object time of section is determined as the second object time, return step one, step 2 and step 3.
In this embodiment it is possible to be shot in different moments to target image, to obtain image recognition apparatus not
Degree in the same time, and then determine the situation of change of the degree of image recognition apparatus.The embodiment can be first
Object time obtains the first object image that image recognition apparatus shoots target object, which can
At the time of thinking in the day of image recognition apparatus start-up operation, the first object clarity of first object image is obtained, first
In the case that target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus is at the first object moment
First object grade.Second object time of the embodiment and first object moment are separated by the first object period, first mesh
Marking the period can be two weeks, and no limitations are hereby intended.
This method further includes step 1, obtains image recognition apparatus in the second object time and shoot to target object
The second target image arrived, for example, the image recognition apparatus of synchronization in one day is taken to clap target object every two weeks
The target image taken the photograph, step 2 obtain the second target sharpness of the second target image, and step 3 is clear in the second target
In the case that clear degree is within the scope of targets threshold, determine the degree of image recognition apparatus the second object time be second
Goal gradient.
The embodiment can be subsequent every the first object period, and detection image identifies the degree of equipment, can incite somebody to action
The third object time for being separated by the first object period after the second object time and with the second object time is determined as second
Object time executes above-mentioned steps one, step 2 and step 3 again, and so on, realize every the first object period from
The purpose of the dynamic degree for obtaining image recognition apparatus.
As an alternative embodiment, the degree for determining image recognition apparatus be the second goal gradient it
Afterwards, this method further include: at least first object grade and the second goal gradient are handled, obtain objective result, wherein mesh
Mark result is used to indicate the variation tendency of the degree of image recognition apparatus.
In this embodiment, after the degree for determining image recognition apparatus is the second goal gradient, at least to the
One goal gradient and the second goal gradient are handled, and objective result is obtained, which can be image recognition apparatus
The data report of degree, form can be textual form, form, represented as histograms etc., not do any limit herein
System.By the objective result, user will be seen that the variation tendency of the degree of image recognition apparatus, to understand image in time
Identify the state of equipment.Optionally, every the first object period, the degree for determining image recognition apparatus grade it
Afterwards, processing result is updated, that is, the figure determined by the grade and history of the newest degree of image recognition apparatus
As identifying that the grade of the degree of equipment regenerates objective result, to indicate the variation of the degree of image recognition apparatus
Trend, and then take effective measures and safeguard to image recognition apparatus is improved and is safeguarded to image recognition apparatus
Efficiency.
As an alternative embodiment, step S102, obtains image recognition apparatus and shoot to target object
To target image include: the different moments in the second target time section, obtain image recognition apparatus to target object carry out
Shoot multiple obtained target images；Step S104, the target sharpness for obtaining target image include: to obtain every target image
Target sharpness, obtain multiple target sharpness；Step S106, in situation of the target sharpness within the scope of targets threshold
Under, it includes: to be greater than the target sharpness of destination number in target that the degree for determining image recognition apparatus, which is goal gradient,
In the case where in threshold range, determine that the degree of image recognition apparatus is goal gradient.
In this embodiment, for the accuracy that the degree improved to image recognition apparatus is detected, Ke Yi
Different moments in second target time section obtain multiple target figures that image recognition apparatus shoots target image
Picture, second target time section can be image recognition apparatus when detecting every time to the degree of image recognition apparatus
To the period that target image is shot, for example, being 1 minute, no limitations are hereby intended, wherein multiple target images
Shooting interval can be identical.Obtain multiple target images that image recognition apparatus shoots target object it
Afterwards, the target sharpness of every target image is obtained, so that multiple target sharpness are obtained, it is clear in the target for being greater than destination number
In the case that clear degree is all within the scope of targets threshold, it is determined that the degree of image recognition apparatus is goal gradient, to arrange
Except the influence that degree of the accidentalia to image recognition apparatus is detected, improves and journey is stained to image recognition apparatus
Spend the accuracy detected.
The target sharpness for the target image that the embodiment shoots target object by image recognition apparatus,
The degree that image recognition apparatus is determined based on quantifiable index improves the degree progress to image recognition apparatus
The effect of the efficiency of detection avoids and is completely dependent on artificial subjective judgement to the judgement of the degree of image recognition apparatus, has
It avoids to effect and deficiency is safeguarded to image recognition apparatus caused by the deviation of artificial subjective judgement or safeguards excessive ask
Topic, improves the maintenance efficiency of image recognition apparatus, reduces the generation of meaningless cost, image recognition apparatus is allowed to remain good
The performance got well, it is ensured that the work quality of image recognition apparatus, and the use longevity of image recognition apparatus is extended to a certain extent
Life.
It should be noted that step shown in the flowchart of the accompanying drawings can be in such as a group of computer-executable instructions
It is executed in computer system, although also, logical order is shown in flow charts, and it in some cases, can be with not
The sequence being same as herein executes shown or described step.
Embodiment 2
Technical solution of the present invention is illustrated below with reference to preferred embodiment.Specifically it is with image recognition apparatus
Video monitoring equipment is illustrated.
The embodiment passes through image definition evaluation algorithm, automatic identification and the degree for determining video monitoring equipment,
And system early warning is carried out according to preset quantizating index in advance, the degree of video monitoring equipment is commented to realize
Estimate, can objective, accurately measure the degree of video monitoring equipment, and timely and effectively arrange subsequent to video monitoring equipment
Maintenance maintenance work, it is ensured that the good monitoring effect of video monitoring equipment, while promoting the maintenance efficiency of video monitoring equipment,
Extend the service life of video monitoring equipment.
Fig. 2 is a kind of schematic diagram of the detection system of the degree of video monitoring equipment according to an embodiment of the present invention.
As shown in Fig. 2, the system includes: video monitoring equipment 1, detection device 2 and terminal 3.Wherein, detection device 2 includes memory
21 and processor 22.
In this embodiment, video monitoring equipment 1 shoots target object, obtains for executing data collection steps
To the image data of target image, the image data of the target image is input in detection device 2, detection device 2 is to the mesh
The image data of logo image is stored, and is detected in processor 22 using the clarity detection method based on Sobel operator
The target sharpness of target image.Optionally, field convolutional calculation is carried out to pixel each in target image, can use 3
Each pixel in × 3 direction template and target image carries out field convolutional calculation, extract each pixel respectively 0 °,
45 °, 90 °, the marginal element on 135 ° of four directions, calculate the gradient value of each pixel in target image, and by each picture
The gradient value of vegetarian refreshments is compared with threshold value, and the pixel that gradient value is greater than threshold value is determined as to the marginal point of target image, from
And filter out the edge pixel point of target image.The gradient value of edge pixel point can be added, with edge gradient energy and work
For the evaluation of estimate that the clarity of target image detects, which can be determined as to the target sharpness N of target image.
Optionally, the clarity evaluation method of the image of the embodiment further includes squared gradient algorithm and fast algorithm of detecting
Deng.Evaluation effect of the evaluation effect of the clarity detection method of Sobel operator than squared gradient algorithm and fast algorithm of detecting
Good, the calculating time longest of squared gradient algorithm, real-time is poor, and fast algorithm of detecting computation rate is high, but to fuzzy spirit
Sensitivity is low, can select alternative solution depending on operation time and intelligibility evaluation precision.
In this embodiment, video monitoring equipment is clearest in the image data default for the acquisition started to work first day
, if its clarity is N1, with the extension of video monitoring equipment working time, greasy dirt situation is aggravated, the image data of acquisition
Clarity decline, the definition values of the image for the maintenance that determines visit are benchmark threshold value N2, video monitoring equipment without
The image clarity values of method work are benchmark threshold value N3.Optionally, which is benchmark threshold value by the first ratio-dependent of N1
N2, for example, the first ratio is 70%, it is benchmark threshold value N3 by the second ratio-dependent of N1, for example, the second ratio is 20%.
The processor 22 of the embodiment by the clarity N of the target image currently acquired and said reference threshold value N2, N3 into
Row compares, and as N > N2, video monitoring equipment is worked normally, it may not be necessary to issue instruction；It, can be to end as N3 < N≤N2
End 3 issues warning-" needing to arrange maintenance " instruction；As N≤N3, can be guarded against to the sending of terminal 3-" it needs to safeguard immediately, it is no
Cannot then work " instruction.Wherein, terminal can be smart phone (such as Android phone, iOS mobile phone), tablet computer, the palm
The user-friendly terminal device such as upper computer and mobile internet device, PAD, so that user can be known immediately
The degree of video monitoring equipment has reached the effect for improving the efficiency detected to the degree of video monitoring equipment
Fruit, and then realize effective maintenance to video monitoring equipment.
Optionally, the video monitoring equipment 1 with image identification function of the embodiment is taken the photograph in the state of normal work
It makes and saves instant picture material.From can starting to work from video monitoring equipment, every two weeks, take same in one day
The image data at moment, is saved in local, as being compared with baseline threshold N2, baseline threshold N3 to determine that video monitoring is set
The data of standby degree.
Processor 22 in the detection device 2 of the embodiment can be used for executing image analysis, can be to memory 21
The picture of newest preservation executes definition values and calculates, and comparison basis threshold value N2 and N3.It guards against, " needs when comparison result issues
Arrange maintenance " when, then processor 22 can be by the data back of picture to terminal 3, and for artificial secondary judgement, final decision is
No needs send someone to carry out the on-site maintenance of video monitoring equipment.
The present invention using intelligence system and can quantitative criteria, be effectively prevented from caused by artificial subjective judgement deviation
Or problem that maintenance excessive insufficient to plant maintenance, so that equipment is maintained superperformance, it is ensured that quality monitoring is horizontal, improves equipment
Maintenance efficiency, reduces the generation of meaningless cost, and extends service life of equipment to a certain extent.
The embodiment application video monitoring equipment per se with camera function, carry out the assessment of the degree of image,
The daily clean and maintenance of equipment is instructed to work.The target image that target object is shot by image recognition apparatus
Target sharpness is determined the degree of image recognition apparatus based on quantifiable index, improved to image recognition apparatus
The effect for the efficiency that degree is detected is avoided and is completely dependent on artificially to the judgement of the degree of image recognition apparatus
Subjective judgement efficiently avoids safeguarding insufficient or dimension to image recognition apparatus caused by the deviation of artificial subjective judgement
Excessive problem is protected, the maintenance efficiency of image recognition apparatus is improved, reduces the generation of meaningless cost, so that image recognition apparatus
It can maintain good performance, it is ensured that the work quality of image recognition apparatus, and extend image recognition to a certain extent and set
Standby service life.
Embodiment 3
The embodiment of the invention also provides a kind of detection devices of the degree of image recognition apparatus.It needs to illustrate
It is that the image that the detection device of the degree of the image recognition apparatus of the embodiment can be used for executing the embodiment of the present invention is known
The detection method of the degree of other equipment.
Fig. 3 is a kind of schematic diagram of the detection device of the degree of image recognition apparatus according to an embodiment of the present invention.
As shown in figure 3, the device includes: first acquisition unit 10, second acquisition unit 20 and determination unit 30.
First acquisition unit 10, the target image that target object is shot for obtaining image recognition apparatus.
Second acquisition unit 20, for obtaining the target sharpness of target image.
Determination unit 30, for determining image recognition apparatus in the case where target sharpness is within the scope of targets threshold
Degree be goal gradient.
Optionally it is determined that unit 30 includes: the first determining module, for the case where target sharpness is greater than first threshold
Under, determine that the degree of image recognition apparatus is the first sub- grade；Second determining module, for being greater than the in target sharpness
Two threshold values and be less than or equal to first threshold in the case where, determine image recognition apparatus degree be the second sub- grade, wherein
The degree of second sub- grade is greater than the degree of the first sub- grade；Third determining module, for small in target sharpness
In the case where being equal to second threshold, determine that the degree of image recognition apparatus is the sub- grade of third, wherein the sub- grade of third
Degree be greater than the second sub- grade degree.
Optionally, the device further include: the first output unit, for being the in the degree for determining image recognition apparatus
After two sub- grades, the first prompt information is exported, wherein the first prompt information is used to indicate the degree of image recognition apparatus
Maintenance is needed to arrange for the second sub- grade and/or image recognition apparatus；Second output unit, for determining that image recognition sets
Standby degree is the second prompt information of output, wherein the second prompt information is used to indicate image knowledge after the sub- grade of third
The degree of other equipment is that the sub- grade of third and/or image recognition apparatus need to safeguard immediately.
Optionally, the first output unit includes: the first output module, for exporting the first prompt information to target end
End；Second output unit includes: the second output module, for exporting the second prompt information to target terminal.
Optionally, the device further include: transmission unit, for when exporting the first prompt information to target terminal, or
When exporting the second prompt information to target terminal, the image data of target image is sent to target terminal.
Optionally, first acquisition unit 10 includes: the first acquisition module, for obtaining image recognition at the first object moment
The first object image that equipment shoots target object；Second acquisition unit 20 includes: the second acquisition module, is used for
Obtain the first object clarity of first object image；Determination unit 30 includes: the 4th determining module, for clear in first object
In the case that clear degree is within the scope of targets threshold, determine that the degree of image recognition apparatus is first object grade；First obtains
It takes unit 10 to be also used to execute step 1, obtains image recognition apparatus in the second object time and target object is shot to obtain
The second target image, wherein the second object time and first object moment are separated by the first object period；Second acquisition unit
20 are also used to execute step 2, obtain the second target sharpness of the second target image；Determination unit 30 is also used to execute step
Three, in the case where the second target sharpness is within the scope of targets threshold, determine that the degree of image recognition apparatus is second
Goal gradient；Wherein, be also used to will be after the second object time and when being separated by first object with the second object time for the device
Between the third object time of section be determined as the second object time, return step one, step 2 and step 3.
Optionally, the device further include: processing unit, for being the second mesh in the degree for determining image recognition apparatus
After marking grade, at least first object grade and the second goal gradient are handled, obtain objective result, wherein target knot
Fruit is used to indicate the variation tendency of the degree of image recognition apparatus.
Optionally, first acquisition unit 10 includes: that third obtains module, when for difference in the second target time section
It carves, obtains multiple target images that image recognition apparatus shoots target object；Second acquisition unit 20 includes:
Four acquisition modules obtain multiple target sharpness for obtaining the target sharpness of every target image；Determination unit 30 is wrapped
It includes: the 5th determining module, for determining figure in the case where being greater than the target sharpness of destination number within the scope of targets threshold
As the degree of identification equipment is goal gradient.
The embodiment obtains the mesh that image recognition apparatus shoots target object by first acquisition unit 10
Logo image obtains the target sharpness of target image by second acquisition unit 20, by determination unit 30 in target sharpness
In the case where within the scope of targets threshold, determine that the degree of image recognition apparatus is goal gradient.That is, passing through figure
As the target sharpness for the target image that identification equipment shoots target object, being stained for image recognition apparatus is determined
Degree avoids and is completely dependent on artificial subjective judgement to the judgement of the degree of image recognition apparatus, solves and knows to image
The technical issues of low efficiency that the degree of other equipment is detected, has reached the degree improved to image recognition apparatus
The technical effect of the efficiency detected.
Embodiment 4
The embodiment of the invention also provides a kind of storage mediums.The storage medium includes the program of storage, wherein in program
Equipment executes the detection side of the degree of the image recognition apparatus in the embodiment of the present invention where controlling storage medium when operation
Method.
Embodiment 5
The embodiment of the invention also provides a kind of processors.The processor is for running program, wherein program is held when running
The detection method of the degree of image recognition apparatus in the row embodiment of the present invention.
Obviously, those skilled in the art should be understood that each module of the above invention or each step can be with general
Computing device realize that they can be concentrated on a single computing device, or be distributed in multiple computing devices and formed
Network on, optionally, they can be realized with the program code that computing device can perform, it is thus possible to which they are stored
Be performed by computing device in the storage device, perhaps they are fabricated to each integrated circuit modules or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.In this way, the present invention is not limited to any specific
Hardware and software combines.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (11)
1. a kind of detection method of the degree of image recognition apparatus characterized by comprising
Obtain the target image that image recognition apparatus shoots target object；
Obtain the target sharpness of the target image；
In the case where the target sharpness is within the scope of targets threshold, determine that the degree of described image identification equipment is
Goal gradient.
2. the method according to claim 1, wherein in the target sharpness within the scope of the targets threshold
In the case where, determine that the degree of described image identification equipment is that the goal gradient includes:
In the case where the target sharpness is greater than first threshold, determine that the degree of described image identification equipment is first
Sub- grade；
In the case where the target sharpness is greater than second threshold and is less than or equal to the first threshold, determine that described image is known
The degree of other equipment is the second sub- grade, wherein the degree of the second sub- grade is greater than the described first sub- grade
Degree；
In the case where the target sharpness is less than or equal to the second threshold, determine described image identification equipment is stained journey
Degree is the sub- grade of third, wherein the degree of the sub- grade of third is greater than the degree of the described second sub- grade.
3. according to the method described in claim 2, it is characterized in that,
After determining that the degree of described image identification equipment is the described second sub- grade, the method also includes: output
First prompt information, wherein the degree that first prompt information is used to indicate described image identification equipment is described the
Two sub- grades and/or described image identification equipment need to arrange maintenance；
After determining that the degree of described image identification equipment is the sub- grade of the third, the method also includes: output
Second prompt information, wherein the degree that second prompt information is used to indicate described image identification equipment is described the
Three sub- grades and/or described image identification equipment need to safeguard immediately.
4. according to the method described in claim 3, it is characterized in that,
Exporting first prompt information includes: to export first prompt information to target terminal；
Exporting second prompt information includes: to export second prompt information to the target terminal.
5. according to the method described in claim 4, it is characterized in that, exporting by first prompt information to target terminal
When, or when exporting second prompt information to the target terminal, the method also includes:
The image data of the target image is sent to the target terminal.
6. method as claimed in any of claims 1 to 5, which is characterized in that
Obtaining the target image that image recognition apparatus shoots target object includes: to obtain institute at the first object moment
State the first object image that image recognition apparatus shoots the target object；
The target sharpness for obtaining the target image includes: the first object clarity for obtaining the first object image；
In the case where the target sharpness is within the scope of targets threshold, determine that the degree of described image identification equipment is
Goal gradient includes: to determine that described image is known in the case where the first object clarity is within the scope of the targets threshold
The degree of other equipment is first object grade；
Obtain the target image that image recognition apparatus shoots target object further include: step 1, in the second target
Moment obtains the second target image for being shot to the target object of described image identification equipment, wherein described the
Two object times and the first object moment are separated by the first object period；
Obtain the target sharpness of the target image further include: step 2 obtains the second target of second target image
Clarity；
In the case where the target sharpness is within the scope of targets threshold, determine that the degree of described image identification equipment is
Goal gradient includes: step 3, in the case where second target sharpness is within the scope of the targets threshold, described in determination
The degree of image recognition apparatus is the second goal gradient；
Wherein, the method also includes: will be separated by after second object time and with second object time described
The third object time of first object period is determined as second object time, returns described Step 1: the step 2
With the step 3.
7. according to the method described in claim 6, it is characterized in that, determining that the degree of described image identification equipment is the
After two goal gradients, the method also includes:
At least the first object grade and second goal gradient are handled, obtain objective result, wherein the mesh
Mark result is used to indicate the variation tendency of the degree of described image identification equipment.
8. method as claimed in any of claims 1 to 5, which is characterized in that
Obtaining the target image that image recognition apparatus shoots target object includes: in the second target time section
Different moments obtain multiple described target images that described image identification equipment shoots the target object；
The target sharpness for obtaining the target image includes: the target sharpness for obtaining every target image, is obtained
To multiple target sharpness；
In the case where the target sharpness is within the scope of targets threshold, determine that the degree of described image identification equipment is
Goal gradient includes: to determine in the case where being greater than the target sharpness of destination number within the scope of the targets threshold
Described image identifies that the degree of equipment is the goal gradient.
9. a kind of detection device of the degree of image recognition apparatus characterized by comprising
First acquisition unit, the target image that target object is shot for obtaining image recognition apparatus；
Second acquisition unit, for obtaining the target sharpness of the target image；
Determination unit, in the case where the target sharpness is within the scope of targets threshold, determining that described image identification is set
Standby degree is goal gradient.
10. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program
When control the storage medium where equipment perform claim require any one of 1 to 8 described in method.
11. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run
Benefit require any one of 1 to 8 described in method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811614985.1A CN109724993A (en) | 2018-12-27 | 2018-12-27 | Detection method, device and the storage medium of the degree of image recognition apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811614985.1A CN109724993A (en) | 2018-12-27 | 2018-12-27 | Detection method, device and the storage medium of the degree of image recognition apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109724993A true CN109724993A (en) | 2019-05-07 |
Family
ID=66296720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811614985.1A Pending CN109724993A (en) | 2018-12-27 | 2018-12-27 | Detection method, device and the storage medium of the degree of image recognition apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109724993A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111241953A (en) * | 2020-01-03 | 2020-06-05 | 秒针信息技术有限公司 | Decontamination verification method and device, storage medium and electronic device |
| CN111524125A (en) * | 2020-04-28 | 2020-08-11 | 北京海益同展信息科技有限公司 | Equipment cleaning method, device and system, electronic equipment and storage medium |
| CN111988215A (en) * | 2020-08-11 | 2020-11-24 | 上海连尚网络科技有限公司 | Method and equipment for pushing user |
| CN112188190A (en) * | 2020-09-30 | 2021-01-05 | 广东美的厨房电器制造有限公司 | Stain detection method, cooking appliance, server, and storage medium |
| CN112959328A (en) * | 2021-04-02 | 2021-06-15 | 北京云迹科技有限公司 | Robot control method, robot control device, robot, and storage medium |
| CN113325752A (en) * | 2021-05-12 | 2021-08-31 | 北京戴纳实验科技有限公司 | Equipment management system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104822568A (en) * | 2012-12-04 | 2015-08-05 | 罗伯特·博世有限公司 | Method and device for controlling cleaning device for vehicle camera, method for cleaning vehicle camera, and camera system for vehicle |
| CN105827914A (en) * | 2016-04-20 | 2016-08-03 | 北京小米移动软件有限公司 | Lens cleaning prompting method and device |
| CN108668080A (en) * | 2018-06-22 | 2018-10-16 | 北京小米移动软件有限公司 | Prompt method and device, the electronic equipment of camera lens degree of fouling |
| CN108956095A (en) * | 2018-05-17 | 2018-12-07 | 北京风云天地信息科技有限公司 | A kind of optical lens pollution level measurement method and device |
-
2018
- 2018-12-27 CN CN201811614985.1A patent/CN109724993A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104822568A (en) * | 2012-12-04 | 2015-08-05 | 罗伯特·博世有限公司 | Method and device for controlling cleaning device for vehicle camera, method for cleaning vehicle camera, and camera system for vehicle |
| CN105827914A (en) * | 2016-04-20 | 2016-08-03 | 北京小米移动软件有限公司 | Lens cleaning prompting method and device |
| CN108956095A (en) * | 2018-05-17 | 2018-12-07 | 北京风云天地信息科技有限公司 | A kind of optical lens pollution level measurement method and device |
| CN108668080A (en) * | 2018-06-22 | 2018-10-16 | 北京小米移动软件有限公司 | Prompt method and device, the electronic equipment of camera lens degree of fouling |
Non-Patent Citations (2)
| Title |
|---|
| 张金玉等: "《红外热波检测及其图像序列处理技术》", 30 June 2015 * |
| 陈华等: "《数字共焦显微技术》", 1 August 2018 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111241953A (en) * | 2020-01-03 | 2020-06-05 | 秒针信息技术有限公司 | Decontamination verification method and device, storage medium and electronic device |
| CN111241953B (en) * | 2020-01-03 | 2023-11-10 | 河北太行农牧供应链有限公司 | Decontamination verification method and device, storage medium and electronic device |
| CN111524125A (en) * | 2020-04-28 | 2020-08-11 | 北京海益同展信息科技有限公司 | Equipment cleaning method, device and system, electronic equipment and storage medium |
| CN111524125B (en) * | 2020-04-28 | 2023-09-01 | 京东科技信息技术有限公司 | Equipment cleaning method, device, system, electronic equipment and storage medium |
| CN111988215A (en) * | 2020-08-11 | 2020-11-24 | 上海连尚网络科技有限公司 | Method and equipment for pushing user |
| CN111988215B (en) * | 2020-08-11 | 2022-07-12 | 上海连尚网络科技有限公司 | Method, equipment and computer readable medium for pushing user |
| CN112188190A (en) * | 2020-09-30 | 2021-01-05 | 广东美的厨房电器制造有限公司 | Stain detection method, cooking appliance, server, and storage medium |
| CN112959328A (en) * | 2021-04-02 | 2021-06-15 | 北京云迹科技有限公司 | Robot control method, robot control device, robot, and storage medium |
| CN113325752A (en) * | 2021-05-12 | 2021-08-31 | 北京戴纳实验科技有限公司 | Equipment management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109724993A (en) | Detection method, device and the storage medium of the degree of image recognition apparatus | |
| US11488294B2 (en) | Method for detecting display screen quality, apparatus, electronic device and storage medium | |
| US11380232B2 (en) | Display screen quality detection method, apparatus, electronic device and storage medium | |
| US20200380899A1 (en) | Method and apparatus for detecting peripheral circuit of display screen, electronic device, and storage medium | |
| CN110866515B (en) | Method and device for identifying behaviors of objects in factory building and electronic equipment | |
| CN105631439A (en) | Human face image collection method and device | |
| CN108921840A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| CN105426828A (en) | Face detection method, face detection device and face detection system | |
| CN107240092A (en) | A kind of image blur detection method and device | |
| CN111310826B (en) | Method and device for detecting labeling abnormality of sample set and electronic equipment | |
| CN105821538A (en) | Detection method and system for breaking of spun yarns | |
| CN113642474A (en) | Hazardous area personnel monitoring method based on YOLOV5 | |
| CN109087281A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| CN109389105A (en) | A kind of iris detection and viewpoint classification method based on multitask | |
| CN111476160A (en) | Loss function optimization method, model training method, target detection method, and medium | |
| CN113780342A (en) | Intelligent detection method and device based on self-supervision pre-training and robot | |
| CN115620192A (en) | Method and device for detecting wearing of safety rope in aerial work | |
| CN102799861A (en) | Method for rapidly identifying reading of instrument by using color | |
| CN108268903A (en) | Article control method, device, readable storage medium storing program for executing and control terminal | |
| CN111582183A (en) | Mask identification method and system in public place | |
| CN113727022B (en) | Method and device for collecting inspection image, electronic equipment and storage medium | |
| CN110087063A (en) | A kind of image processing method, device and electronic equipment | |
| CN109975307A (en) | Bearing surface defect detection system and detection method based on statistics projection training | |
| CN107403192A (en) | A kind of fast target detection method and system based on multi-categorizer | |
| CN115672778A (en) | Intelligent visual recognition system |
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: 20190507 |
|
| RJ01 | Rejection of invention patent application after publication |