CN109886934A - Be carbonized bamboo chip defect inspection method and system - Google Patents
Be carbonized bamboo chip defect inspection method and system Download PDFInfo
- Publication number
- CN109886934A CN109886934A CN201910079411.7A CN201910079411A CN109886934A CN 109886934 A CN109886934 A CN 109886934A CN 201910079411 A CN201910079411 A CN 201910079411A CN 109886934 A CN109886934 A CN 109886934A
- Authority
- CN
- China
- Prior art keywords
- bamboo chip
- carbonization bamboo
- carbonization
- image data
- defective
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of carbonization bamboo chip defect inspection method and systems.Wherein, the described method includes: acquisition has the image data sample of defective history carbonization bamboo chip, with the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition, establish the defects detection model of the image data based on carbonization bamboo chip, with the defects detection model of the image data according to the foundation based on carbonization bamboo chip, defects detection is carried out to current carbonization bamboo chip, and the testing result of defects detection is carried out to current carbonization bamboo chip according to this, it screens out and has defective carbonization bamboo chip in current carbonization bamboo chip.By the above-mentioned means, can be realized it is not necessary that manually defects detection can be carried out to carbonization bamboo chip automatically, detection efficiency and Detection accuracy are improved, the quality of the carbonization bamboo chip after defects detection can be protected, to improve the quality of bamboo product.
Description
Technical field
The present invention relates to carbonized bamboo technical field more particularly to a kind of carbonization bamboo chip defect inspection method and systems.
Background technique
Bamboo is branch gramineous, is perennial plant, many kinds of, is distributed in the torrid zone, subtropical zone to warm-temperature zone
Area, China are known as the reputation of " bamboo kingdom ", bamboo resource very abundant.According to statistics, whole world bamboo grove area is there are about 22,000,000 hectares,
And China's bamboo grove area accounts for about one third up to 7,200,000 hectares, occupies second place of the world, is only second to India.Global Sinobambusa plant
More than 70 belong to, and more than 1200;And China is there are about the category of Sinobambusa plant 35, nearly 400 kinds.16,000,000 tons of whole world bamboo grove annual output, China
Bamboo grove annual output reach 8,000,000 tons or more, almost account for the half of world's bamboo yield;Occupy No. 1 in the world.So at me
State, bamboo are otherwise known as " the second forest ".Moso bamboo area distributions are maximum in China's bamboo grove area, widest in area.Wherein, Fujian,
The bamboo grove area in Jiangxi, Zhejiang San Sheng accounts for national half.
With the improvement of living standards, bamboo product industry such as bamboo summer sleeping mat, bamboo chip art engineering, bamboo chip building are
As the hot spot of public attention, however develop bamboo product, it is necessary to " mouldy, rot, damage by worms " etc. of solution bamboo material first
Defect problem, i.e., mould proof, anti-corrosion, mothproof three proofings processing.The composition of bamboo material is similar with timber, form bamboo wood it is main at
Point there are cellulose, hemicellulose and lignin, containing protein-based, fats and various carbohydrates and a small amount of ash element.
Carbonization is also known as destructive distillation, charing, coking, refers to the reaction of solid or the organic matter heat resolve in the case where completely cutting off air conditions
Process heats solid matter to produce liquid or gas, it will usually become a kind of mode of solid product.Carbonisation is different
Surely it can be related to cracking or be pyrolyzed, collect product after condensation.Compared with usual distillation, carbonisation needs higher temperature, makes
The fuel of liquid can be extracted from charcoal or timber with carbonization.Carbonization can also decompose mineral salt by being pyrolyzed, such as right
Sulfate destructive distillation can produce sulfur dioxide and sulfur trioxide, can be obtained by sulfuric acid after being dissolved in water;To coal carbonization, can obtain burnt
Charcoal, coal tar, ammonia liquor, coal gas;It is carbonized to bamboo chip, can must be carbonized bamboo chip.
The principle of carbonization bamboo chip is with high temperature the sugar " burn-up " in bamboo chip, and bamboo chip processed so is just not easy by worm
Sub to bite, as color, true qualities are not different with dark, and only passing through for dark color paints.The bamboo chip of carbonization passes through
It will not be split after crossing carbonization dehydration, and bamboo chip, that is, raw bamboo of true qualities can be split.Carbonization bamboo chip equally exists identical with the bamboo chip of true qualities
The defects of defect problem, i.e. " mouldy, rot, damage by worms " problem.
Bamboo chip be carbonized before carrying out the bamboo product processing of next step, it is necessary to first examine to carbonization bamboo chip surface defect
It surveys, otherwise not can guarantee the quality of bamboo product, cause qualification rate low.
The existing scheme that defects detection is carried out to carbonization bamboo chip relies primarily on artificial eye and knows otherwise to carbonization bamboo chip
It is screened, large labor intensity, screening efficiency is low, and especially human eye carries out dull pan inspection for a long time, easily generation vision
Fatigue is easier to increase erroneous judgement and missing inspection.The quality of carbonization bamboo chip after artificial screening is difficult to be protected, and causes bamboo product matter
Amount can not stablize raising, greatly restrict the fast development of bamboo product industry.
But at least there are the following problems in the prior art for inventor's discovery:
The scheme of existing carbonization bamboo chip defects detection relies primarily on artificial eye knowledge and sieves otherwise to carbonization bamboo chip
Choosing, large labor intensity, screening efficiency is low, is easy erroneous judgement and missing inspection, the quality of the carbonization bamboo chip after artificial screening are difficult to obtain
It ensures, causes bamboo product quality that can not stablize raising.
Summary of the invention
In view of this, can be realized it is an object of the invention to propose a kind of carbonization bamboo chip defect inspection method and system
It is not necessary that manually defects detection can be carried out to carbonization bamboo chip automatically, detection efficiency and Detection accuracy are improved, after defects detection
The quality of carbonization bamboo chip can be protected, to improve the quality of bamboo product.
According to an aspect of the present invention, a kind of carbonization bamboo chip defect inspection method is provided, comprising:
Acquisition has the image data sample of defective history carbonization bamboo chip;Wherein, the defective history carbonization of the tool
It include multiple image informations for having defective history carbonization bamboo chip and corresponding defect type in the image data sample of bamboo chip
Label;
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, establish based on carbonization bamboo chip
The defects detection model of image data;
The defects detection model of image data according to the foundation based on carbonization bamboo chip, to current carbonization bamboo chip into
Row defects detection;
According to the testing result for carrying out defects detection to current carbonization bamboo chip, screen out in current carbonization bamboo chip
Has defective carbonization bamboo chip.
Wherein, the image data sample of the defective history carbonization bamboo chip of the tool according to the acquisition, foundation are based on
The defects detection model of the image data of carbonization bamboo chip, comprising:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, it is defective to obtain the tool
It the image information for multiple tools defective history carbonization bamboo chip that history is carbonized in the image data sample of bamboo chip and corresponding lacks
Fall into type label;
It will be multiple with defect in the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition
History carbonization bamboo chip image information and corresponding defect type label be divided into N sections;Wherein, the N is the nature greater than 1
Number;
By convolutional neural networks, the image of the defective history carbonization bamboo chip of multiple tools after N sections are divided into described in extraction
The time weight feature of information and corresponding defect type label;
The defective history carbonization of multiple tools according to the time weight feature of the extraction, after being divided into N sections described in acquisition
The Analysis On Multi-scale Features of the image information of bamboo chip and corresponding defect type label;
It merges multiple with defect in the image data sample of the defective history carbonization bamboo chip of N section tool of the acquisition
History carbonization bamboo chip image information and corresponding defect type label Analysis On Multi-scale Features, calculate prediction score;
According to the prediction score being calculated, the defective history carbonization of tool of the final association acquisition is obtained
The classification of the image data sample of bamboo chip;
According to point of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition
Class obtains the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
According to the instruction of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition
Practice feature and carry out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
Wherein, the defects detection model of the image data based on carbonization bamboo chip according to the foundation, to current
The bamboo chip that is carbonized carries out defects detection, comprising:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbon
Change the training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of bamboo chip, using institute
The mode that the training characteristics matched are trained current carbonized bamboo picture information is stated, to current carbonized bamboo picture
Information carries out defects detection, carries out defects detection to current carbonization bamboo chip.
Wherein, described according to the testing result for carrying out defects detection to current carbonization bamboo chip, it screens out current
Have defective carbonization bamboo chip in carbonization bamboo chip, comprising:
According to the testing result for carrying out defects detection to current carbonization bamboo chip, the inspection is prompted using type of alarm
It surveys as a result, screening out according to the testing result of the prompt and having defective carbonization bamboo chip in current carbonization bamboo chip.
Wherein, before the image data sample that the acquisition has defective history carbonization bamboo chip, further includes:
After the completion of having the detection process of defective history carbonization bamboo chip, the defective history carbonized bamboo of tool is obtained
The image information of piece, and the corresponding defect type of image information of the defective history carbonization bamboo chip of tool according to the acquisition,
Generate the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool.
According to another aspect of the present invention, a kind of carbonization bamboo chip defect detecting system is provided, comprising:
Acquisition unit establishes unit, detection unit and screening unit;
The acquisition unit, for acquiring the image data sample for having defective history carbonization bamboo chip;Wherein, the tool
It include the image information of the defective history carbonization bamboo chip of multiple tools in the image data sample of defective history carbonization bamboo chip
And corresponding defect type label;
It is described to establish unit, for the image data sample according to the tool defective history carbonization bamboo chip of the acquisition,
Establish the defects detection model of the image data based on carbonization bamboo chip;
The detection unit, for the defects detection model of the image data according to the foundation based on carbonization bamboo chip,
Defects detection is carried out to current carbonization bamboo chip;
The screening unit, for according to the testing result for carrying out defects detection to current carbonization bamboo chip, screening
Fall and has defective carbonization bamboo chip in current carbonization bamboo chip.
Wherein, described to establish unit, it is specifically used for:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, it is defective to obtain the tool
It the image information for multiple tools defective history carbonization bamboo chip that history is carbonized in the image data sample of bamboo chip and corresponding lacks
Fall into type label;
It will be multiple with defect in the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition
History carbonization bamboo chip image information and corresponding defect type label be divided into N sections;Wherein, the N be greater than natural number;
By convolutional neural networks, the image of the defective history carbonization bamboo chip of multiple tools after N sections are divided into described in extraction
The time weight feature of information and corresponding defect type label;
The defective history carbonization of multiple tools according to the time weight feature of the extraction, after being divided into N sections described in acquisition
The Analysis On Multi-scale Features of the image information of bamboo chip and corresponding defect type label;
It merges multiple with defect in the image data sample of the defective history carbonization bamboo chip of N section tool of the acquisition
History carbonization bamboo chip image information and corresponding defect type label Analysis On Multi-scale Features, calculate prediction score;
According to the prediction score being calculated, the defective history carbonization of tool of the final association acquisition is obtained
The classification of the image data sample of bamboo chip;
According to point of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition
Class obtains the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
According to the instruction of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition
Practice feature and carry out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
Wherein, the detection unit, is specifically used for:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbon
Change the training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of bamboo chip, using institute
The mode that the training characteristics matched are trained current carbonized bamboo picture information is stated, to current carbonized bamboo picture
Information carries out defects detection, carries out defects detection to current carbonization bamboo chip.
Wherein, the screening unit, is specifically used for:
According to the testing result for carrying out defects detection to current carbonization bamboo chip, the inspection is prompted using type of alarm
It surveys as a result, screening out according to the testing result of the prompt and having defective carbonization bamboo chip in current carbonization bamboo chip.
Wherein, the carbonization bamboo chip defect detecting system, further includes:
Generation unit, for having described in acquisition and lacking after the completion of having the detection process of defective history carbonization bamboo chip
The image information of sunken history carbonization bamboo chip, and the image information pair of the defective history carbonization bamboo chip of tool according to the acquisition
The defect type answered generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool.
It can be found that above scheme, can acquire the image data sample for having defective history carbonization bamboo chip, wherein
It include the image of the defective history carbonization bamboo chip of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool
Information and corresponding defect type label, and the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition,
Establish the defects detection model of the image data based on carbonization bamboo chip, and the image data based on carbonization bamboo chip according to the foundation
Defects detection model, defects detection is carried out to current carbonization bamboo chip, and carry out to current carbonization bamboo chip according to this scarce
The testing result for falling into detection, screens out and has defective carbonization bamboo chip in current carbonization bamboo chip, can be realized without artificial energy
Automatically defects detection is carried out to carbonization bamboo chip, improves detection efficiency and Detection accuracy, the carbonization bamboo chip after defects detection
Quality can be protected, to improve the quality of bamboo product.
Further, above scheme, can be according to the image data sample of the defective history carbonization bamboo chip of tool of the acquisition
This, obtains the defective history carbonization bamboo chip of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool
Image information and corresponding defect type label, and by the image data sample of the tool of the acquisition defective history carbonization bamboo chip
The image information of the defective history carbonization bamboo chip of multiple tools in this and corresponding defect type label are divided into N sections, wherein should
N is the natural number greater than 1, and by convolutional neural networks, extracts the defective history carbonized bamboo of multiple tools after this is divided into N sections
The time weight feature of the image information of piece and corresponding defect type label, and according to the time weight feature of the extraction, obtain
The defective history of multiple tools after being divided into N section is carbonized the image information of bamboo chip and more rulers of corresponding defect type label
Spend feature, and merge the acquisition N section have it is multiple with defect in the image data sample of defective history carbonization bamboo chip
History carbonization bamboo chip image information and corresponding defect type label Analysis On Multi-scale Features, calculate prediction score, and according to
The prediction score being calculated obtains the image data sample of the final defective history carbonization bamboo chip of the tool for being associated with the acquisition
This classification, with point of the image data sample of the defective history carbonization bamboo chip of the tool for being associated with the acquisition obtained according to this
Class obtains the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition, and according to this
The training characteristics of the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition carry out model training,
The defects detection model for establishing the image data based on carbonization bamboo chip can be realized to improve and establish the picture number based on carbonization bamboo chip
According to defects detection model modeling effect and accuracy.
Further, above scheme, can be according to the defects detection mould of the image data based on carbonization bamboo chip of the foundation
Type matches current carbonized bamboo picture letter from the defects detection model of the image data based on carbonization bamboo chip of the foundation
The training characteristics of breath are right in such a way that the training characteristics matched are trained current carbonized bamboo picture information
Current carbonized bamboo picture information carries out defects detection, carries out defects detection to current carbonization bamboo chip, can effectively improve
The detection efficiency and accuracy rate of the testing result of current carbonized bamboo picture information.
Further, above scheme can carry out the testing result of defects detection to current carbonization bamboo chip according to this, adopt
The testing result is prompted with type of alarm, according to the testing result of the prompt, screening out has defect in current carbonization bamboo chip
Carbonization bamboo chip, can be realized the omission factor for improving by type of alarm and carrying out defects detection to carbonization bamboo chip automatically, improve warp
The quality assurance of carbonization bamboo chip after defects detection, to improve the quality of bamboo product.
Further, above scheme, can be after the completion of having the detection process of defective history carbonization bamboo chip, and obtaining should
Has the image information of defective history carbonization bamboo chip, according to the image information of the defective history carbonization bamboo chip of the tool of the acquisition
Corresponding defect type generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool, can
Realize the building efficiency for effectively improving the defects detection model for establishing the image data based on carbonization bamboo chip.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram of present invention carbonization one embodiment of bamboo chip defect inspection method;
Fig. 2 is the flow diagram of present invention carbonization another embodiment of bamboo chip defect inspection method;
Fig. 3 is the structural schematic diagram of present invention carbonization one embodiment of bamboo chip defect detecting system;
Fig. 4 is the structural schematic diagram of present invention carbonization another embodiment of bamboo chip defect detecting system;
Fig. 5 is the structural schematic diagram of the present invention carbonization another embodiment of bamboo chip defect detecting system.
Specific embodiment
With reference to the accompanying drawings and examples, the present invention is described in further detail.It is emphasized that following implement
Example is merely to illustrate the present invention, but is not defined to the scope of the present invention.Likewise, following embodiment is only portion of the invention
Point embodiment and not all embodiments, institute obtained by those of ordinary skill in the art without making creative efforts
There are other embodiments, shall fall within the protection scope of the present invention.
The present invention provides a kind of carbonization bamboo chip defect inspection method, can be realized without manually can automatically to carbonization bamboo chip into
Row defects detection, improves detection efficiency and Detection accuracy, and the quality of the carbonization bamboo chip after defects detection can be protected,
To improve the quality of bamboo product.
Referring to Figure 1, Fig. 1 is the flow diagram of present invention carbonization one embodiment of bamboo chip defect inspection method.It should be noted
If having substantially the same as a result, method of the invention is not limited with process sequence shown in FIG. 1.As shown in Figure 1,
This method comprises the following steps:
S101: acquisition has the image data sample of defective history carbonization bamboo chip;Wherein, the defective history carbon of the tool
Changing includes multiple image informations and corresponding defect class for having defective history carbonization bamboo chip in the image data sample of bamboo chip
Type label.
Wherein, before the image data sample that the acquisition has defective history carbonization bamboo chip, can also include:
After the completion of having the detection process of defective history carbonization bamboo chip, the defective history carbonization bamboo chip of the tool is obtained
Image information;
According to the corresponding defect type of image information of the defective history carbonization bamboo chip of the tool of the acquisition, generating this has
The corresponding defect type label of image information of the history carbonization bamboo chip of defect.
In the present embodiment, the electronic equipment such as server of carbonization bamboo chip defect inspection method operation thereon can lead to
It crosses wired connection mode or radio connection and is acquired from tester using the terminal device that it is logged in defect
History carbonization bamboo chip image data sample.
In the present embodiment, which can be with camera and multiple sensors including but not limited to photosensitive,
Distance, gravity, acceleration, the various electric terminals of the sensors such as magnetic induction, including but not limited to smart phone, tablet computer,
Pocket computer on knee and desktop computer etc..
In the present embodiment, which can be to provide the server of various services, such as show on terminal device
The image data login interface of the carbonization bamboo chip shown provides the backstage login service device supported, which can be right
The data such as the image data of history carbonization bamboo chip and the image data of current carbonization bamboo chip carry out the processing such as analyzing, and processing is tied
Fruit, such as can will recommend the advisory information of tester for reference and feed back to terminal device.
In the present embodiment, tester can be used terminal device and be handed over by network and electronic equipment such as server
Mutually, to receive or send message etc..The various client applications for needing to verify tester's information can be installed on terminal device,
Such as carbonization bamboo chip class application, instant messaging tools, mailbox client, carbonization bamboo chip platform software etc..
S102: it according to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, establishes and is based on carbonized bamboo
The defects detection model of the image data of piece.
Wherein, the image data sample of the tool defective history carbonization bamboo chip according to the acquisition, is established based on carbonization
The defects detection model of the image data of bamboo chip may include:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the defective history of the tool is obtained
The image information and corresponding defect class of the defective history carbonization bamboo chip of the multiple tools being carbonized in the image data sample of bamboo chip
Type label;
It goes through multiple tools in the image data sample of the tool of the acquisition defective history carbonization bamboo chip are defective
The image information of history carbonization bamboo chip and corresponding defect type label are divided into N sections;Wherein, which is the natural number greater than 1;
By convolutional neural networks, the image letter of the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is extracted
The time weight feature of breath and corresponding defect type label;
According to the time weight feature of the extraction, the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is obtained
Image information and corresponding defect type label Analysis On Multi-scale Features;
The multiple tools merged in the image data sample of the defective history carbonization bamboo chip of N section tool of the acquisition are defective
The image information of history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score;
The prediction score being calculated according to this obtains the defective history carbonization bamboo chip of the final tool for being associated with the acquisition
Image data sample classification;
According to the classification of the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition, obtain
To the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
Training according to the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition is special
Sign carries out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
In the present embodiment, convolutional neural networks are a kind of comprising convolutional calculation and with the Feedforward Neural Networks of depth structure
Network is one of representative algorithm of deep learning.
In the present embodiment, the convolutional neural networks may include: at least one Three dimensional convolution layer, at least one three-dimensional
Pond layer and at least one full articulamentum etc..
S103: according to the defects detection model of the image data based on carbonization bamboo chip of the foundation, to current carbonized bamboo
Piece carries out defects detection.
Wherein, the defects detection model of the image data according to the foundation based on carbonization bamboo chip, to current carbonization
Bamboo chip carries out defects detection, may include:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbonized bamboo
The training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of piece, using the matching
The mode that training characteristics out are trained current carbonized bamboo picture information, to current carbonized bamboo picture information into
Row defects detection carries out defects detection to current carbonization bamboo chip.
In the present embodiment, the current carbonized bamboo picture information can be the image letter of current goal carbonization bamboo chip
Breath etc., the present invention is not limited.
S104: the testing result of defects detection is carried out to current carbonization bamboo chip according to this, screens out current carbonized bamboo
Has defective carbonization bamboo chip in piece.
Wherein, the testing result for carrying out defects detection to current carbonization bamboo chip according to this, screens out current carbonization
Have defective carbonization bamboo chip in bamboo chip, may include:
The testing result for carrying out defects detection to current carbonization bamboo chip according to this, prompts the detection knot using type of alarm
Fruit screens out according to the testing result of the prompt and has defective carbonization bamboo chip in current carbonization bamboo chip.
It can be found that in the present embodiment, the image data sample for having defective history carbonization bamboo chip can be acquired,
In, it include the figure of the defective history carbonization bamboo chip of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool
As information and corresponding defect type label, and the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition
This, establishes the defects detection model of the image data based on carbonization bamboo chip, and the image based on carbonization bamboo chip according to the foundation
The defects detection model of data carries out defects detection to current carbonization bamboo chip, and according to this to current carbonization bamboo chip into
The testing result of row defects detection screens out and has defective carbonization bamboo chip in current carbonization bamboo chip, can be realized without people
Work can carry out defects detection to carbonization bamboo chip automatically, improve detection efficiency and Detection accuracy, the carbonization after defects detection
The quality of bamboo chip can be protected, to improve the quality of bamboo product.
It further, in the present embodiment, can be according to the picture number of the defective history carbonization bamboo chip of tool of the acquisition
According to sample, the defective history carbonized bamboo of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool is obtained
The image information of piece and corresponding defect type label, and by the picture number of the tool of the acquisition defective history carbonization bamboo chip
It is divided into N sections according to the image information and corresponding defect type label of the defective history carbonization bamboo chip of multiple tools in sample,
In, which is the natural number greater than 1, and by convolutional neural networks, extracts the defective history of multiple tools after this is divided into N sections
The image information of bamboo chip that is carbonized and the time weight feature of corresponding defect type label, and it is special according to the time weight of the extraction
Sign, obtain the defective history of multiple tools after this is divided into N section be carbonized bamboo chip image information and corresponding defect type label
Analysis On Multi-scale Features, and the N section for merging the acquisition has multiple tools in the image data sample of defective history carbonization bamboo chip
The image information of defective history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score,
With the prediction score being calculated according to this, the image of the final defective history carbonization bamboo chip of the tool for being associated with the acquisition is obtained
The classification of data sample, with the image data sample of the defective history carbonization bamboo chip of the tool for being associated with the acquisition obtained according to this
Classification, obtain being associated with the training characteristics of the image data sample of the defective history carbonization bamboo chip of tool of the acquisition, Yi Jigen
The training characteristics of the image data sample of the defective history carbonization bamboo chip of the tool for being associated with the acquisition obtained according to this carry out model
The defects detection model of the image data based on carbonization bamboo chip is established in training, be can be realized to improve and be established based on carbonization bamboo chip
The modeling effect and accuracy of the defects detection model of image data.
Further, in the present embodiment, it can be examined according to the defect of the image data based on carbonization bamboo chip of the foundation
Model is surveyed, matches current carbonization bamboo chip figure from the defects detection model of the image data based on carbonization bamboo chip of the foundation
As the training characteristics of information, the side that current carbonized bamboo picture information is trained using the training characteristics matched
Formula, carries out defects detection to current carbonized bamboo picture information, carries out defects detection to current carbonization bamboo chip, can be effective
Improve the detection efficiency and accuracy rate of the testing result of current carbonized bamboo picture information.
Further, in the present embodiment, the detection knot of defects detection can be carried out to current carbonization bamboo chip according to this
Fruit prompts the testing result using type of alarm, and according to the testing result of the prompt, screening out has in current carbonization bamboo chip
The carbonization bamboo chip of defect can be realized the omission factor for improving by type of alarm and carrying out defects detection to carbonization bamboo chip automatically, mention
The quality assurance of carbonization bamboo chip of the height after defects detection, to improve the quality of bamboo product.
Fig. 2 is referred to, Fig. 2 is the flow diagram of present invention carbonization another embodiment of bamboo chip defect inspection method.This reality
It applies in example, method includes the following steps:
S201: after the completion of having the detection process of defective history carbonization bamboo chip, the defective history carbon of the tool is obtained
The image information for changing bamboo chip, according to the corresponding defect type of image information of the tool defective history carbonization bamboo chip of the acquisition,
Generate the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool.
S202: acquisition has the image data sample of defective history carbonization bamboo chip;Wherein, the defective history carbon of the tool
Changing includes multiple image informations and corresponding defect class for having defective history carbonization bamboo chip in the image data sample of bamboo chip
Type label.
Can be as above described in S101, therefore not to repeat here.
S203: it according to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, establishes and is based on carbonized bamboo
The defects detection model of the image data of piece.
Can be as above described in S102, therefore not to repeat here.
S204: according to the defects detection model of the image data based on carbonization bamboo chip of the foundation, to current carbonized bamboo
Piece carries out defects detection.
Can be as above described in S103, therefore not to repeat here.
S205: the testing result of defects detection is carried out to current carbonization bamboo chip according to this, screens out current carbonized bamboo
Has defective carbonization bamboo chip in piece.
Can be as above described in S104, therefore not to repeat here.
It can be found that in the present embodiment, can be obtained after the completion of having the detection process of defective history carbonization bamboo chip
The image information for taking the defective history carbonization bamboo chip of the tool, according to the image of the defective history carbonization bamboo chip of the tool of the acquisition
The corresponding defect type of information generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool,
It can be realized the building efficiency for effectively improving the defects detection model for establishing the image data based on carbonization bamboo chip.
The present invention also provides a kind of carbonization bamboo chip defect detecting system, can be realized without manually can automatically to carbonization bamboo chip
Defects detection is carried out, improves detection efficiency and Detection accuracy, the quality of the carbonization bamboo chip after defects detection can be protected
Barrier, to improve the quality of bamboo product.
Fig. 3 is referred to, Fig. 3 is the structural schematic diagram of present invention carbonization one embodiment of bamboo chip defect detecting system.This implementation
In example, which includes acquisition unit 31, establishes unit 32, detection unit 33 and screening unit
34。
The acquisition unit 31, for acquiring the image data sample for having defective history carbonization bamboo chip;Wherein, this has
Include in the image data sample of the history carbonization bamboo chip of defect multiple tools defective history carbonization bamboo chip image informations and
Corresponding defect type label.
This establishes unit 32, for the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition, builds
Be based on be carbonized bamboo chip image data defects detection model.
The detection unit 33 is right for the defects detection model of the image data according to the foundation based on carbonization bamboo chip
Current carbonization bamboo chip carries out defects detection.
The screening unit 34 is screened out for carrying out the testing result of defects detection to current carbonization bamboo chip according to this
Has defective carbonization bamboo chip in current carbonization bamboo chip.
Optionally, this establishes unit 32, can be specifically used for:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the defective history of the tool is obtained
The image information and corresponding defect class of the defective history carbonization bamboo chip of the multiple tools being carbonized in the image data sample of bamboo chip
Type label;
It goes through multiple tools in the image data sample of the tool of the acquisition defective history carbonization bamboo chip are defective
The image information of history carbonization bamboo chip and corresponding defect type label are divided into N sections;Wherein, which is the natural number greater than 1;
By convolutional neural networks, the image letter of the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is extracted
The time weight feature of breath and corresponding defect type label;
According to the time weight feature of the extraction, the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is obtained
Image information and corresponding defect type label Analysis On Multi-scale Features;
The multiple tools merged in the image data sample of the defective history carbonization bamboo chip of N section tool of the acquisition are defective
The image information of history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score;
The prediction score being calculated according to this obtains the defective history carbonization bamboo chip of the final tool for being associated with the acquisition
Image data sample classification;
According to the classification of the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition, obtain
To the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
Training according to the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition is special
Sign carries out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
Optionally, the detection unit 33, can be specifically used for:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbonized bamboo
The training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of piece, using the matching
The mode that training characteristics out are trained current carbonized bamboo picture information, to current carbonized bamboo picture information into
Row defects detection carries out defects detection to current carbonization bamboo chip.
Optionally, the screening unit 34, can be specifically used for:
The testing result for carrying out defects detection to current carbonization bamboo chip according to this, prompts the detection knot using type of alarm
Fruit screens out according to the testing result of the prompt and has defective carbonization bamboo chip in current carbonization bamboo chip.
Fig. 4 is referred to, Fig. 4 is the structural schematic diagram of present invention carbonization another embodiment of bamboo chip defect detecting system.Difference
Yu Shangyi embodiment, carbonization bamboo chip defect detecting system 40 described in the present embodiment further include: generation unit 41.
The generation unit 41, for after the completion of having the detection process of defective history carbonization bamboo chip, obtaining this to have
The image information of the history carbonization bamboo chip of defect, and the image information pair of the defective history carbonization bamboo chip of tool according to the acquisition
The defect type answered generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool.
Fig. 5 is referred to, Fig. 5 is the structural schematic diagram of the present invention carbonization another embodiment of bamboo chip defect detecting system.The carbon
Each unit module for changing bamboo chip defect detecting system can execute respectively corresponds to step in above method embodiment.Related content
The detailed description in the above method is referred to, it is no longer superfluous herein to chat.
The storage that in the present embodiment, which includes: processor 51, is coupled with the processor 51
Device 52, detector 53, screening washer 54.
The processor 51, for after the completion of having the detection process of defective history carbonization bamboo chip, obtain this have it is scarce
The image information of sunken history carbonization bamboo chip, and it is corresponding according to the image information of the defective history carbonization bamboo chip of tool of the acquisition
Defect type, generate the corresponding defect type label of image information of the tool defective history carbonization bamboo chip, and acquisition tool
The image data sample of defective history carbonization bamboo chip, wherein the image data sample of the defective history carbonization bamboo chip of the tool
Include multiple image informations for having defective history carbonization bamboo chip and corresponding defect type label in this, and is adopted according to this
The image data sample of the defective history carbonization bamboo chip of the tool of collection, establishes the defects detection of the image data based on carbonization bamboo chip
Model.
The memory 52, the instruction etc. executed for storage program area, the processor 51.
The detector 53, for the defects detection model of the image data according to the foundation based on carbonization bamboo chip, to working as
Preceding carbonization bamboo chip carries out defects detection.
The screening washer 54 screens out and works as carrying out the testing result of defects detection to current carbonization bamboo chip according to this
Has defective carbonization bamboo chip in preceding carbonization bamboo chip.
Optionally, the processor 51, can be specifically used for:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the defective history of the tool is obtained
The image information and corresponding defect class of the defective history carbonization bamboo chip of the multiple tools being carbonized in the image data sample of bamboo chip
Type label;
It goes through multiple tools in the image data sample of the tool of the acquisition defective history carbonization bamboo chip are defective
The image information of history carbonization bamboo chip and corresponding defect type label are divided into N sections;Wherein, which is the natural number greater than 1;
By convolutional neural networks, the image letter of the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is extracted
The time weight feature of breath and corresponding defect type label;
According to the time weight feature of the extraction, the defective history carbonization bamboo chip of multiple tools after this is divided into N sections is obtained
Image information and corresponding defect type label Analysis On Multi-scale Features;
The multiple tools merged in the image data sample of the defective history carbonization bamboo chip of N section tool of the acquisition are defective
The image information of history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score;
The prediction score being calculated according to this obtains the defective history carbonization bamboo chip of the final tool for being associated with the acquisition
Image data sample classification;
According to the classification of the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition, obtain
To the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
Training according to the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition is special
Sign carries out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
Optionally, the detector 53, can be specifically used for:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbonized bamboo
The training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of piece, using the matching
The mode that training characteristics out are trained current carbonized bamboo picture information, to current carbonized bamboo picture information into
Row defects detection carries out defects detection to current carbonization bamboo chip.
Optionally, the screening washer 54, can be specifically used for:
The testing result for carrying out defects detection to current carbonization bamboo chip according to this, prompts the detection knot using type of alarm
Fruit screens out according to the testing result of the prompt and has defective carbonization bamboo chip in current carbonization bamboo chip.
It can be found that above scheme, can acquire the image data sample for having defective history carbonization bamboo chip, wherein
It include the image of the defective history carbonization bamboo chip of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool
Information and corresponding defect type label, and the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition,
Establish the defects detection model of the image data based on carbonization bamboo chip, and the image data based on carbonization bamboo chip according to the foundation
Defects detection model, defects detection is carried out to current carbonization bamboo chip, and carry out to current carbonization bamboo chip according to this scarce
The testing result for falling into detection, screens out and has defective carbonization bamboo chip in current carbonization bamboo chip, can be realized without artificial energy
Automatically defects detection is carried out to carbonization bamboo chip, improves detection efficiency and Detection accuracy, the carbonization bamboo chip after defects detection
Quality can be protected, to improve the quality of bamboo product.
Further, above scheme, can be according to the image data sample of the defective history carbonization bamboo chip of tool of the acquisition
This, obtains the defective history carbonization bamboo chip of multiple tools in the image data sample of the defective history carbonization bamboo chip of the tool
Image information and corresponding defect type label, and by the image data sample of the tool of the acquisition defective history carbonization bamboo chip
The image information of the defective history carbonization bamboo chip of multiple tools in this and corresponding defect type label are divided into N sections, wherein should
N is the natural number greater than 1, and by convolutional neural networks, extracts the defective history carbonized bamboo of multiple tools after this is divided into N sections
The time weight feature of the image information of piece and corresponding defect type label, and according to the time weight feature of the extraction, obtain
The defective history of multiple tools after being divided into N section is carbonized the image information of bamboo chip and more rulers of corresponding defect type label
Spend feature, and merge the acquisition N section have it is multiple with defect in the image data sample of defective history carbonization bamboo chip
History carbonization bamboo chip image information and corresponding defect type label Analysis On Multi-scale Features, calculate prediction score, and according to
The prediction score being calculated obtains the image data sample of the final defective history carbonization bamboo chip of the tool for being associated with the acquisition
This classification, with point of the image data sample of the defective history carbonization bamboo chip of the tool for being associated with the acquisition obtained according to this
Class obtains the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition, and according to this
The training characteristics of the image data sample of the obtained defective history carbonization bamboo chip of the tool for being associated with the acquisition carry out model training,
The defects detection model for establishing the image data based on carbonization bamboo chip can be realized to improve and establish the picture number based on carbonization bamboo chip
According to defects detection model modeling effect and accuracy.
Further, above scheme, can be according to the defects detection mould of the image data based on carbonization bamboo chip of the foundation
Type matches current carbonized bamboo picture letter from the defects detection model of the image data based on carbonization bamboo chip of the foundation
The training characteristics of breath are right in such a way that the training characteristics matched are trained current carbonized bamboo picture information
Current carbonized bamboo picture information carries out defects detection, carries out defects detection to current carbonization bamboo chip, can effectively improve
The detection efficiency and accuracy rate of the testing result of current carbonized bamboo picture information.
Further, above scheme can carry out the testing result of defects detection to current carbonization bamboo chip according to this, adopt
The testing result is prompted with type of alarm, according to the testing result of the prompt, screening out has defect in current carbonization bamboo chip
Carbonization bamboo chip, can be realized the omission factor for improving by type of alarm and carrying out defects detection to carbonization bamboo chip automatically, improve warp
The quality assurance of carbonization bamboo chip after defects detection, to improve the quality of bamboo product.
Further, above scheme, can be after the completion of having the detection process of defective history carbonization bamboo chip, and obtaining should
Has the image information of defective history carbonization bamboo chip, according to the image information of the defective history carbonization bamboo chip of the tool of the acquisition
Corresponding defect type generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool, can
Realize the building efficiency for effectively improving the defects detection model for establishing the image data based on carbonization bamboo chip.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can
To realize by another way.For example, device embodiments described above are only schematical, for example, module or
The division of unit, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units
Or component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, institute
Display or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit
Indirect coupling or communication connection can be electrical property, mechanical or other forms.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.It can select some or all of unit therein according to the actual needs to realize the mesh of present embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) or processor (processor) execute each implementation of the present invention
The all or part of the steps of methods.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The foregoing is merely section Examples of the invention, are not intended to limit protection scope of the present invention, all utilizations
Equivalent device made by description of the invention and accompanying drawing content or equivalent process transformation are applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of carbonization bamboo chip defect inspection method characterized by comprising
Acquisition has the image data sample of defective history carbonization bamboo chip;Wherein, the defective history carbonization bamboo chip of the tool
Image data sample in include multiple tools defective history carbonization bamboo chip image informations and corresponding defect type label;
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the image based on carbonization bamboo chip is established
The defects detection model of data;
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, current carbonization bamboo chip is carried out scarce
Fall into detection;
According to the testing result for carrying out defects detection to current carbonization bamboo chip, screening out has in current carbonization bamboo chip
The carbonization bamboo chip of defect.
2. carbonization bamboo chip defect inspection method as described in claim 1, which is characterized in that the having according to the acquisition
The image data sample of the history carbonization bamboo chip of defect, establishes the defects detection model of the image data based on carbonization bamboo chip, packet
It includes:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the defective history of tool is obtained
The image information and corresponding defect class of the defective history carbonization bamboo chip of the multiple tools being carbonized in the image data sample of bamboo chip
Type label;
It goes through multiple tools in the image data sample of the tool of the acquisition defective history carbonization bamboo chip are defective
The image information of history carbonization bamboo chip and corresponding defect type label are divided into N sections;Wherein, the N is the natural number greater than 1;
By convolutional neural networks, the image information of the defective history carbonization bamboo chip of multiple tools after N sections are divided into described in extraction
And the time weight feature of corresponding defect type label;
The defective history carbonization bamboo chip of multiple tools according to the time weight feature of the extraction, after being divided into N sections described in acquisition
Image information and corresponding defect type label Analysis On Multi-scale Features;
The N section for merging the acquisition has that multiple tools in the image data sample of defective history carbonization bamboo chip are defective to be gone through
The image information of history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score;
According to the prediction score being calculated, the defective history carbonization bamboo chip of tool of the final association acquisition is obtained
Image data sample classification;
According to the classification of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition, obtain
To the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
Training according to the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition is special
Sign carries out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
3. carbonization bamboo chip defect inspection method as described in claim 1, which is characterized in that it is described according to the foundation based on
The defects detection model of the image data of carbonization bamboo chip carries out defects detection to current carbonization bamboo chip, comprising:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbonized bamboo
The training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of piece, using described
The mode that the training characteristics allotted are trained current carbonized bamboo picture information, to current carbonized bamboo picture information
Defects detection is carried out, defects detection is carried out to current carbonization bamboo chip.
4. as described in claim 1 carbonization bamboo chip defect inspection method, which is characterized in that it is described according to described to current carbon
Change the testing result that bamboo chip carries out defects detection, screen out and have defective carbonization bamboo chip in current carbonization bamboo chip, comprising:
According to the testing result for carrying out defects detection to current carbonization bamboo chip, the detection is prompted to tie using type of alarm
Fruit screens out according to the testing result of the prompt and has defective carbonization bamboo chip in current carbonization bamboo chip.
5. carbonization bamboo chip defect inspection method as described in claim 1, which is characterized in that have defective go through in the acquisition
History is carbonized before the image data sample of bamboo chip, further includes:
After the completion of having the detection process of defective history carbonization bamboo chip, the defective history carbonization bamboo chip of the tool is obtained
Image information, and the corresponding defect type of image information of the defective history carbonization bamboo chip of tool according to the acquisition, generate
The corresponding defect type label of image information of the defective history carbonization bamboo chip of tool.
6. a kind of carbonization bamboo chip defect detecting system characterized by comprising
Acquisition unit establishes unit, detection unit and screening unit;
The acquisition unit, for acquiring the image data sample for having defective history carbonization bamboo chip;Wherein, described with scarce
It include the image informations of the defective history carbonization bamboo chip of multiple tools in the image data sample of sunken history carbonization bamboo chip and right
The defect type label answered;
It is described to establish unit, for the image data sample of the defective history carbonization bamboo chip of tool according to the acquisition, establish
The defects detection model of image data based on carbonization bamboo chip;
The detection unit, for the defects detection model of the image data according to the foundation based on carbonization bamboo chip, to working as
Preceding carbonization bamboo chip carries out defects detection;
The screening unit, for screening out and working as according to the testing result for carrying out defects detection to current carbonization bamboo chip
Has defective carbonization bamboo chip in preceding carbonization bamboo chip.
7. carbonization bamboo chip defect detecting system as claimed in claim 6, which is characterized in that it is described to establish unit, it is specifically used for:
According to the image data sample of the defective history carbonization bamboo chip of the tool of the acquisition, the defective history of tool is obtained
The image information and corresponding defect class of the defective history carbonization bamboo chip of the multiple tools being carbonized in the image data sample of bamboo chip
Type label;
It goes through multiple tools in the image data sample of the tool of the acquisition defective history carbonization bamboo chip are defective
The image information of history carbonization bamboo chip and corresponding defect type label are divided into N sections;Wherein, the N be greater than natural number;
By convolutional neural networks, the image information of the defective history carbonization bamboo chip of multiple tools after N sections are divided into described in extraction
And the time weight feature of corresponding defect type label;
The defective history carbonization bamboo chip of multiple tools according to the time weight feature of the extraction, after being divided into N sections described in acquisition
Image information and corresponding defect type label Analysis On Multi-scale Features;
The N section for merging the acquisition has that multiple tools in the image data sample of defective history carbonization bamboo chip are defective to be gone through
The image information of history carbonization bamboo chip and the Analysis On Multi-scale Features of corresponding defect type label, calculate prediction score;
According to the prediction score being calculated, the defective history carbonization bamboo chip of tool of the final association acquisition is obtained
Image data sample classification;
According to the classification of the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition, obtain
To the training characteristics of the image data sample for the defective history carbonization bamboo chip of tool for being associated with the acquisition;
Training according to the image data sample of the defective history carbonization bamboo chip of tool of the obtained association acquisition is special
Sign carries out model training, establishes the defects detection model of the image data based on carbonization bamboo chip.
8. carbonization bamboo chip defect detecting system as claimed in claim 6, which is characterized in that the detection unit is specifically used for:
According to the defects detection model of the image data based on carbonization bamboo chip of the foundation, from the foundation based on carbonized bamboo
The training characteristics that current carbonized bamboo picture information is matched in the defects detection model of the image data of piece, using described
The mode that the training characteristics allotted are trained current carbonized bamboo picture information, to current carbonized bamboo picture information
Defects detection is carried out, defects detection is carried out to current carbonization bamboo chip.
9. carbonization bamboo chip defect detecting system as claimed in claim 6, which is characterized in that the screening unit is specifically used for:
According to the testing result for carrying out defects detection to current carbonization bamboo chip, the detection is prompted to tie using type of alarm
Fruit screens out according to the testing result of the prompt and has defective carbonization bamboo chip in current carbonization bamboo chip.
10. carbonization bamboo chip defect detecting system as claimed in claim 6, which is characterized in that the carbonization bamboo chip defects detection
System, further includes:
Generation unit, for it is defective to obtain the tool after the completion of having the detection process of defective history carbonization bamboo chip
The image information of history carbonization bamboo chip, and it is corresponding according to the image information of the defective history carbonization bamboo chip of tool of the acquisition
Defect type generates the corresponding defect type label of image information of the defective history carbonization bamboo chip of the tool.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910079411.7A CN109886934B (en) | 2019-01-28 | 2019-01-28 | Carbonized bamboo chip defect detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910079411.7A CN109886934B (en) | 2019-01-28 | 2019-01-28 | Carbonized bamboo chip defect detection method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109886934A true CN109886934A (en) | 2019-06-14 |
CN109886934B CN109886934B (en) | 2020-12-18 |
Family
ID=66926945
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910079411.7A Active CN109886934B (en) | 2019-01-28 | 2019-01-28 | Carbonized bamboo chip defect detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109886934B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112241805A (en) * | 2019-07-19 | 2021-01-19 | 因斯派克托里奥股份有限公司 | Defect prediction using historical inspection data |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103954994A (en) * | 2014-04-17 | 2014-07-30 | 中国石油天然气集团公司 | Seismic signal enhancement method and device based on continuous wavelet transformation |
CN107256549A (en) * | 2017-06-06 | 2017-10-17 | 滁州市天达汽车部件有限公司 | A kind of bamboo strip defect detection method based on machine vision |
CN107564002A (en) * | 2017-09-14 | 2018-01-09 | 广东工业大学 | Plastic tube detection method of surface flaw, system and computer-readable recording medium |
CN108562589A (en) * | 2018-03-30 | 2018-09-21 | 慧泉智能科技(苏州)有限公司 | A method of magnetic circuit material surface defect is detected |
CN108664989A (en) * | 2018-03-27 | 2018-10-16 | 北京达佳互联信息技术有限公司 | Image tag determines method, apparatus and terminal |
CN109064454A (en) * | 2018-07-12 | 2018-12-21 | 上海蝶鱼智能科技有限公司 | Product defects detection method and system |
CN109242825A (en) * | 2018-07-26 | 2019-01-18 | 北京首钢自动化信息技术有限公司 | A kind of steel surface defect identification method and device based on depth learning technology |
CN109871455A (en) * | 2019-01-28 | 2019-06-11 | 厦门理工学院 | Be carbonized bamboo chip color separation method and system |
-
2019
- 2019-01-28 CN CN201910079411.7A patent/CN109886934B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103954994A (en) * | 2014-04-17 | 2014-07-30 | 中国石油天然气集团公司 | Seismic signal enhancement method and device based on continuous wavelet transformation |
CN107256549A (en) * | 2017-06-06 | 2017-10-17 | 滁州市天达汽车部件有限公司 | A kind of bamboo strip defect detection method based on machine vision |
CN107564002A (en) * | 2017-09-14 | 2018-01-09 | 广东工业大学 | Plastic tube detection method of surface flaw, system and computer-readable recording medium |
CN108664989A (en) * | 2018-03-27 | 2018-10-16 | 北京达佳互联信息技术有限公司 | Image tag determines method, apparatus and terminal |
CN108562589A (en) * | 2018-03-30 | 2018-09-21 | 慧泉智能科技(苏州)有限公司 | A method of magnetic circuit material surface defect is detected |
CN109064454A (en) * | 2018-07-12 | 2018-12-21 | 上海蝶鱼智能科技有限公司 | Product defects detection method and system |
CN109242825A (en) * | 2018-07-26 | 2019-01-18 | 北京首钢自动化信息技术有限公司 | A kind of steel surface defect identification method and device based on depth learning technology |
CN109871455A (en) * | 2019-01-28 | 2019-06-11 | 厦门理工学院 | Be carbonized bamboo chip color separation method and system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112241805A (en) * | 2019-07-19 | 2021-01-19 | 因斯派克托里奥股份有限公司 | Defect prediction using historical inspection data |
Also Published As
Publication number | Publication date |
---|---|
CN109886934B (en) | 2020-12-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106447046B (en) | A kind of Size Dwelling Design scheme evaluating method based on machine learning | |
CN108108736A (en) | A kind of solar energy photovoltaic panel spot identification method | |
CN105334224B (en) | Automate quality testing cloud platform | |
CN111127448B (en) | Method for detecting air spring fault based on isolated forest | |
Liao et al. | Using drones for thermal imaging photography and building 3D images to analyze the defects of solar modules | |
CN106206356A (en) | The method improving Yield lmproved defect inspection efficiency | |
CN103927400A (en) | Web site product detailed information classification crawling and product information base establishing method | |
CN112651966A (en) | Printed circuit board micro-defect detection method based on ACYOLOV4_ CSP | |
CN102998350B (en) | Method for distinguishing edible oil from swill-cooked dirty oil by electrochemical fingerprints | |
CN107132266A (en) | A kind of Classification of water Qualities method and system based on random forest | |
CN110930470A (en) | Cloth defect image generation method based on deep learning | |
CN113820017A (en) | Generator carbon brush temperature monitoring system based on infrared image and temperature prediction method | |
Puno et al. | Quality Assessment of Mangoes using Convolutional Neural Network | |
CN109886934A (en) | Be carbonized bamboo chip defect inspection method and system | |
CN104933365B (en) | A kind of malicious code based on calling custom automates homologous decision method and system | |
CN109871455A (en) | Be carbonized bamboo chip color separation method and system | |
CN116485802B (en) | Insulator flashover defect detection method, device, equipment and storage medium | |
CN108765391A (en) | A kind of plate glass foreign matter image analysis methods based on deep learning | |
CN115809795B (en) | Method and device for evaluating bearing capacity of production team based on digitalization | |
CN115937555A (en) | Industrial defect detection algorithm based on standardized flow model | |
CN104091480A (en) | Transmission line condition-based maintenance simulation training method | |
CN114565581A (en) | Detection method, recording medium and system for low-value insulator of distribution line | |
CN116843605A (en) | Fruit and vegetable defect detection method and system based on AI algorithm | |
CN112686843B (en) | Board defect detection method and system based on neural network | |
CN111898314B (en) | Lake water parameter inspection method and device, electronic equipment 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |