CN108634241A - Select halogen method and its brine-adding system - Google Patents
Select halogen method and its brine-adding system Download PDFInfo
- Publication number
- CN108634241A CN108634241A CN201810439169.5A CN201810439169A CN108634241A CN 108634241 A CN108634241 A CN 108634241A CN 201810439169 A CN201810439169 A CN 201810439169A CN 108634241 A CN108634241 A CN 108634241A
- Authority
- CN
- China
- Prior art keywords
- betel nut
- halogen
- outer profile
- module
- nut outer
- 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.)
- Withdrawn
Links
Classifications
-
- A—HUMAN NECESSITIES
- A23—FOODS OR FOODSTUFFS; TREATMENT THEREOF, NOT COVERED BY OTHER CLASSES
- A23L—FOODS, FOODSTUFFS, OR NON-ALCOHOLIC BEVERAGES, NOT COVERED BY SUBCLASSES A21D OR A23B-A23J; THEIR PREPARATION OR TREATMENT, e.g. COOKING, MODIFICATION OF NUTRITIVE QUALITIES, PHYSICAL TREATMENT; PRESERVATION OF FOODS OR FOODSTUFFS, IN GENERAL
- A23L19/00—Products from fruits or vegetables; Preparation or treatment thereof
- A23L19/03—Products from fruits or vegetables; Preparation or treatment thereof consisting of whole pieces or fragments without mashing the original pieces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Nutrition Science (AREA)
- Chemical & Material Sciences (AREA)
- Food Science & Technology (AREA)
- Polymers & Plastics (AREA)
- Image Processing (AREA)
Abstract
A kind of halogen method of ordering includes the following steps：Present image is acquired, target area selection is carried out to present image, obtains target image.Parameter threshold processing is carried out to target image, obtains two-value picture.The identification operation of betel nut outer profile is carried out to two-value picture, obtains betel nut outer profile distribution map.Piecemeal cutting process is carried out to betel nut outer profile distribution map, obtains betel nut outer profile figure.Betel nut outer profile figure is calculated respectively, obtains each fruit caving Internal periphery.Calculate separately the target size of each fruit caving Internal periphery.Preset halogen path and preset halogen flow are obtained according to the target size of the fruit caving Internal periphery in each betel nut outer profile figure.The halogen method of ordering of the present invention uses image processing method, so as to obtain selecting brine amount and ordering halogen path for each betel nut nucleome on charging tray, thus it can be accurately controlled to selecting brine amount and order halogen path in actual betel nut orders halogen production so that betel nut orders the quality that halogen processes and effectively improves.
Description
Technical field
The present invention relates to betel nut processing technologies, more particularly to a kind of halogen method and its brine-adding system of ordering.
Background technology
The food custom of betel nut is varied, some users are directly edible fresh fruits, this is related to the processing of betel nut
It is less, still, the user of edible processing betel nut is liked for some, then complicated process operation can be carried out to betel nut.
Traditional way is complete using being manually processed, but with industrialized development and betel nut market
Growth requirement, this requires the artificial lifes that the process operation of betel nut needs that industrialization is relied on to replace production efficiency more low
Production.
Currently, in the integral processing of edible areca-nut, need each betel nut in an orderly manner and according to the side of rectangular array
Formula is placed in charging tray, after charging tray arrangement places betel nut, needs order halogen processing technology to the betel nut on charging tray.
However, existing betel nut orders halogen method during practice, it may appear that halogen amount of ordering problem not easy to control, from
And cause the processing quality of areca nut food to reduce so that the brine amount of different betel nuts differs in actual betel nut orders halogen production,
Meanwhile existing betel nut orders halogen method during practice, ordering halogen path is not easy to control, to which saltwater brine is fallen when making to order halogen
Enter betel nut nucleome edge or fall on charging tray, areca nut food shell surface is caused to be infected with by brine, does not meet the mark of production requirement
Thus standard reduces the processing quality of areca nut food.
Invention content
The purpose of the present invention is overcoming shortcoming in the prior art, providing one kind can order to betel nut halogen amount and halogen of ordering
What path was controlled order halogen method and its brine-adding system.
The purpose of the present invention is achieved through the following technical solutions：
A kind of halogen method of ordering includes the following steps：
Acquisition is placed with the present image of the charging tray of multiple betel nuts, carries out target area selection to the present image, obtains
To target image；
The parameter model of the target image is established, for carrying out parameter threshold processing to the target image, is obtained
Two-value picture；
Condition is identified according to default outer profile, the identification operation of betel nut outer profile is carried out to the two-value picture, to obtain Bin
Bulky outer profile distribution map；
Respectively according to the boundary rectangle of each betel nut outer profile, piecemeal cut place is carried out to the betel nut outer profile distribution map
Reason, to obtain multiple independent betel nut outer profile figures；
According to default computation model, the betel nut outer profile figure is calculated respectively, obtains each betel nut outer profile
Fruit caving Internal periphery in figure；
Calculate separately the target size of the fruit caving Internal periphery in each betel nut outer profile figure；
Preset halogen path is obtained according to the target size of the fruit caving Internal periphery in each betel nut outer profile figure and is preset
Select halogen flow；
Controlled motion module is moved according to preset halogen path, and corresponding control is located in the motion module
Halogen module of ordering order halogen operation according to the preset halogen flow.
Acquisition is placed with multiple present images of the charging tray of multiple betel nuts in one of the embodiments, according to betel nut material
Mark point on disk carries out the predeterminable area selection to each present image, obtains multiple initial pictures, will be each described first
Beginning image carries out image mosaic operation, obtains the target image.
The default computation model obtains as follows in one of the embodiments,：
Acquisition is placed with the training image of the charging tray of multiple betel nuts；
Fruit caving Internal periphery marking operation is carried out to each training image, and is normalized, multiple processing are obtained
Image；
Convolutional neural networks training operation is carried out according to each processing image, obtains the default computation model；Wherein,
The convolutional neural networks are based on residual error neural network and full convolutional neural networks framework, including 60 convolutional layers.
Further include following steps in one of the embodiments, before calculating the betel nut outer profile figure：
Picture superposition processing is carried out to the betel nut outer profile figure.
The parameter model of the target image is saturation parameters model in one of the embodiments,.
In one of the embodiments, in the betel nut outer profile distribution map, according to the external square of each betel nut outer profile
Shape, at the boundary position of the boundary rectangle outward amplification 5~10 pixels, then to the betel nut outer profile distribution map into
Row piecemeal cutting process, to obtain multiple independent betel nut outer profile figures.
The target size of the fruit caving Internal periphery includes fruit caving Internal periphery area, in fruit caving in one of the embodiments,
Profile width and fruit caving Internal periphery length, according to the fruit caving Internal periphery area in each betel nut outer profile figure, the fruit
Intracavitary profile width and the fruit caving Internal periphery length, to obtain the preset halogen flow.
A kind of brine-adding system includes：
Image capture module, described image acquisition module are used to acquire the present image for the charging tray for being placed with multiple betel nuts；
Image processing module, described image processing module are used to carry out target area selection to the present image, obtain
Target image；Described image processing module is additionally operable to establish the parameter model of the target image, for the target image
Parameter threshold processing is carried out, two-value picture is obtained；Described image processing module is additionally operable to identify condition according to default outer profile,
The identification operation of betel nut outer profile is carried out to the two-value picture, to obtain betel nut outer profile distribution map；Described image processing module
It is additionally operable to, respectively according to the boundary rectangle of each betel nut outer profile, piecemeal cutting process be carried out to the betel nut outer profile distribution map,
To obtain multiple independent betel nut outer profile figures；
Computing module prestores the default computation model in the computing module, and the computing module is for right respectively
The betel nut outer profile figure is calculated, and the fruit caving Internal periphery in each betel nut outer profile figure is obtained；The computing module is also
Target size for calculating separately the fruit caving Internal periphery in each betel nut outer profile figure；The computing module is additionally operable to basis
The target size of fruit caving Internal periphery in each betel nut outer profile figure obtains preset halogen path and preset halogen flow；
Motion module, the motion module are moved according to preset halogen path；
Halogen module is selected, the halogen module of ordering is located in the motion module, and the halogen module of ordering is according to the preset halogen
Flow order halogen operation；
Control module, the control module is respectively used to control the motion module and the halogen module of ordering executes excercises
Work and halogen operation of ordering.
The control module is XYZ three-axis moving modules in one of the embodiments, for driving the halogen module of ordering
Carry out the reciprocating type displacement operation of tri- axis directions of XYZ.
The brine-adding system further includes brine supply module in one of the embodiments, the brine supply module with
The halogen module of ordering is connected to, and the brine supply module is used to input brine to the halogen module of ordering using circuit cycle mode.
Compared with prior art, the present invention has at least the following advantages：
The present invention order halogen method use image processing method, by charging tray betel nut nucleome carry out Image Acquisition,
Analyzing processing and calculating, so as to obtain selecting brine amount and ordering halogen path for each betel nut nucleome on charging tray, thus in reality
Betel nut order and can be accurately controlled to selecting brine amount and order halogen path in halogen production so that betel nut orders the matter that halogen processes
Amount effectively improves.
Description of the drawings
Fig. 1 is the flow chart for ordering halogen method of the present invention；
Fig. 2 is the structural schematic diagram of the brine-adding system of one embodiment of the invention.
Specific implementation mode
To facilitate the understanding of the present invention, below with reference to relevant drawings to invention is more fully described.In attached drawing
Give the better embodiment of the present invention.But the present invention can realize in many different forms, however it is not limited to herein
Described embodiment.On the contrary, the purpose of providing these embodiments is that making to understand more the disclosure
Add thorough and comprehensive.
It should be noted that when element is referred to as " being fixed on " another element, it can be directly on another element
Or there may also be elements placed in the middle.When an element is considered as " connection " another element, it can be directly connected to
To another element or it may be simultaneously present centering elements.Term as used herein " vertical ", " horizontal ", " left side ",
" right side " and similar statement for illustrative purposes only, are not offered as being unique embodiment.
Unless otherwise defined, all of technologies and scientific terms used here by the article and belong to the technical field of the present invention
The normally understood meaning of technical staff is identical.Used term is intended merely to description tool in the description of the invention herein
The purpose of the embodiment of body, it is not intended that in the limitation present invention.Term " and or " used herein includes one or more
Any and all combinations of relevant Listed Items.
It is a kind of to order halogen method in one embodiment, include the following steps：Acquisition is placed with the current of the charging tray of multiple betel nuts
Image carries out target area selection to the present image, obtains target image；The parameter model of the target image is established,
For carrying out parameter threshold processing to the target image, two-value picture is obtained；Condition is identified according to default outer profile, to institute
It states two-value picture and carries out the identification operation of betel nut outer profile, to obtain betel nut outer profile distribution map；Respectively according to each betel nut outer profile
Boundary rectangle, to the betel nut outer profile distribution map carry out piecemeal cutting process, to obtain multiple independent betel nut outer profiles
Figure；According to default computation model, the betel nut outer profile figure is calculated respectively, is obtained in each betel nut outer profile figure
Fruit caving Internal periphery；Calculate separately the target size of the fruit caving Internal periphery in each betel nut outer profile figure；According to each betel nut
The target size of fruit caving Internal periphery in outer profile figure obtains preset halogen path and preset halogen flow；Controlled motion module root
It is moved according to preset halogen path, and corresponding control is located at the halogen module of ordering in the motion module and is preset according to described
It selects halogen flow and order halogen operation.A kind of brine-adding system includes：Image capture module, described image acquisition module are put for acquiring
It is placed with the present image of the charging tray of multiple betel nuts；Image processing module, described image processing module are used for the present image
Target area selection is carried out, target image is obtained；Described image processing module is additionally operable to establish the parameter mould of the target image
Type obtains two-value picture for carrying out parameter threshold processing to the target image；Described image processing module is additionally operable to root
Condition is identified according to default outer profile, the identification operation of betel nut outer profile is carried out to the two-value picture, to obtain betel nut outer profile point
Butut；Described image processing module is additionally operable to respectively according to the boundary rectangle of each betel nut outer profile, to the betel nut outer profile point
Butut carries out piecemeal cutting process, to obtain multiple independent betel nut outer profile figures；Computing module prestores in the computing module
There are the pre-designed calculation model, the computing module to be obtained each described for calculating respectively the betel nut outer profile figure
Fruit caving Internal periphery in betel nut outer profile figure；The computing module is additionally operable to calculate separately the fruit in each betel nut outer profile figure
The target size of intracavitary profile；The computing module is additionally operable to the mesh according to the fruit caving Internal periphery in each betel nut outer profile figure
Dimensioning obtains preset halogen path and preset halogen flow；Motion module, the motion module is according to preset halogen road
Diameter is moved；Halogen module is selected, the halogen module of ordering is located in the motion module, and the halogen module of ordering is according to the preset
Halogen flow order halogen operation；Control module, the control module are respectively used to control the motion module and the halogen mould of ordering
Block executes motor performance and halogen of ordering and operates.The halogen method of ordering of the present invention uses image processing method, by the betel nut on charging tray
Nucleome carries out Image Acquisition, analyzing processing and calculating, so as to obtain the point brine amount and point of each betel nut nucleome on charging tray
Thus halogen path can accurately control so that Bin in actual betel nut orders halogen production to selecting brine amount and ordering halogen path
The bulky quality for ordering halogen processing effectively improves.
In order to preferably be illustrated to above-mentioned halogen method of ordering, to more fully understand the above-mentioned design for ordering halogen method.It please join
Read Fig. 1, it is a kind of to order halogen method, include the following steps：
Step S110：Acquisition is placed with the present image of the charging tray of multiple betel nuts, and target area is carried out to the present image
Domain is chosen, and target image is obtained.
In one embodiment, acquisition is placed with multiple present images of the charging tray of multiple betel nuts, according to the mark on betel nut charging tray
Note point carries out the predeterminable area selection to each present image, obtains multiple initial pictures, by each initial pictures into
Row image mosaic operates, and obtains the target image.
It should be noted that the charging tray for being placed with multiple betel nuts in actual production process is transmitted by transport mechanism
Image taking is carried out to image taking mechanism, for example, transport mechanism can be the transport mechanisms such as conveyer belt or transmission chain, figure
As photographic unit can be the image takings such as industrial camera or CCD imaging mechanisms mechanism.Since betel nut charging tray has certain body
Product size so that image taking mechanism needs the betel nut repeatedly passed through to transmission when carrying out image taking acquisition to betel nut charging tray
Charging tray is shot, be previously provided on betel nut charging tray location hole or locating slot etc. be used as mark point, image taking mechanism according to
Mark point is continuously shot as the center of betel nut charging tray on betel nut charging tray, thus obtains the multiple initial of betel nut charging tray
Then image carries out coincidence splicing according to the mark point of betel nut charging tray to multiple initial pictures, thus obtain having complete Bin
The image of bulky charging tray, the i.e. present image of betel nut charging tray.
Step S210：The parameter model of the target image is established, for carrying out parameter threshold to the target image
Processing, obtains two-value picture.
In one embodiment, the parameter model of the target image is saturation parameters model.In another embodiment, the mesh
The parameter model of logo image can also be color parameter model or luminance parameter model etc..
It should be noted that after obtaining the present image of betel nut charging tray, by using the parameter model pair of target image
The target image carries out parameter threshold processing, to obtain the two-value picture of the target image.For example, using saturation degree
The target image is carried out picture saturation degree extraction, is handled by saturation analysis, the target figure at this time by parameter model
As being the clearly demarcated two-value picture of saturation degree, for example, only retaining the two-value picture of Black-White, enabling simpler accurately to exist
Required betel nut outer profile pictorial information is extracted in the two-value picture.
Step S310：Condition is identified according to default outer profile, and the identification operation of betel nut outer profile is carried out to the two-value picture,
To obtain betel nut outer profile distribution map；
In one embodiment, the default outer profile identification condition is area, length, the width etc. of preset betel nut outer profile
Data characteristic information, by regarding the data characteristic informations such as the area of betel nut outer profile, length, width as default outer profile identification
Thus condition obtains betel nut outer profile distribution map so as to carry out the identification operation of betel nut outer profile to the two-value picture.
It should be noted that after obtaining the two-value picture, by comparing each object pattern in the two-value picture
The features such as area, length, width, filter smaller interfering object feature (for example, in picture the outer profile feature of equipment or
Outer profile feature of betel nut charging tray etc.), to obtain outside each betel nut in regime values range in the target image
Profile diagram can so, it is possible to obtain betel nut foreign steamer clear from each betel nut outer profile in the position of the two-value picture
Wide distribution map.
Step S410：Respectively according to the boundary rectangle of each betel nut outer profile, the betel nut outer profile distribution map is divided
Block cutting process, to obtain multiple independent betel nut outer profile figures.
In one embodiment, in the betel nut outer profile distribution map, according to the boundary rectangle of each betel nut outer profile, described
Amplify 5~10 pixels at the boundary position of boundary rectangle outward, then piecemeal is carried out to the betel nut outer profile distribution map and is cut
Processing is cut, to obtain multiple independent betel nut outer profile figures.
It should be noted that the boundary rectangle of betel nut outer profile refers to the rectangle abutted with betel nut outer profile, outside betel nut
Profile be included in boundary rectangle in, for example, betel nut outer profile be ellipse when, boundary rectangle be respectively with betel nut outer profile both ends
And both sides abut and are formed by rectangle.After obtaining betel nut outer profile distribution map by step 3, the betel nut outer profile is distributed
Figure carries out piecemeal cutting process, to obtain multiple independent betel nut outer profile figures.In order to make outside each independent betel nut
Profile diagram can be when being split, it is ensured that the edge of the betel nut outer profile in each betel nut outer profile figure will not be divided
Fall, at the boundary position of the boundary rectangle outward amplification 5~10 pixels, then to the betel nut outer profile distribution map into
Row piecemeal cutting process, to obtain multiple independent betel nut outer profile figures.Specific dividing method is calculated outside betel nut
The minimum enclosed rectangle of profile carries out after then amplifying 5~10 pixels outward at the boundary position of the boundary rectangle
Picture is cut, it includes a betel nut thus to make the every pictures being cut into all.
Step S510：According to default computation model, the betel nut outer profile figure is calculated respectively, obtains each Bin
Fruit caving Internal periphery in bulky outer profile figure.
Before calculating the betel nut outer profile figure, picture superposition is carried out to the betel nut outer profile figure
Processing.
In one embodiment, enhance the picture contrast of the betel nut outer profile figure using Retinex algorithm for image enhancement,
So that the characteristic details on the betel nut outer profile figure are more distinct, thus, it is possible to preferably by the fruit in betel nut outer profile figure
Intracavitary profile calculates.
In another embodiment, the default computation model obtains as follows：
Acquisition is placed with the training image of the charging tray of multiple betel nuts；Fruit caving Internal periphery label is carried out to each training image
Operation, and be normalized, obtain multiple processing images；Convolutional neural networks training is carried out according to each processing image
Operation, obtains the default computation model；Wherein, the convolutional neural networks are based on residual error neural network and full convolutional Neural net
Network framework, including 60 convolutional layers.
It should be noted that the training image of the charging tray of multiple betel nuts, batch are placed with by industrial camera acquisition first
After the training image for obtaining the charging tray of these betel nuts, by manually to the betel nut intracavitary profile in these training images into rower
Note, and is normalized, and thus obtains multiple processing images to get to the data information of betel nut intracavitary profile, for example,
Area information, length information, width information, saturation infromation and chroma-luminance information of betel nut intracavitary profile etc..Completion is returned
After one change processing, convolutional neural networks training operation is carried out according to each processing image, obtains the default computation model；Also
That is, by the area information of the betel nut intracavitary profile in each processing image, length information, width information, saturation infromation
And the data informations such as chroma-luminance information are sorted out and are integrated into data associated with betel nut intracavitary outline data information
Network, i.e. convolutional neural networks, wherein the convolutional neural networks are based on residual error neural network and full convolutional neural networks frame
Structure, including 60 convolutional layers.By the default computation model that convolutional neural networks training operation obtains after optimization, it is fixed to have
Position is accurate, and calculation amount is small, and speed is fast, the advantages that being easy to train.So after obtaining the betel nut outer profile figure, by default
Computation model carries out calculation processing to the betel nut outer profile figure, to obtain the fruit caving lubrication groove in each betel nut outer profile figure
It is wide.
Step S610：Calculate separately the target size of the fruit caving Internal periphery in each betel nut outer profile figure.
Step S710：Preset halogen road is obtained according to the target size of the fruit caving Internal periphery in each betel nut outer profile figure
Diameter and preset halogen flow；
In one embodiment, the target size of the fruit caving Internal periphery includes fruit caving Internal periphery area, fruit caving Internal periphery width
With fruit caving Internal periphery length, according to the fruit caving Internal periphery area, the fruit caving Internal periphery in each betel nut outer profile figure
Width and the fruit caving Internal periphery length, to obtain the preset halogen flow.
It should be noted that by calculate fruit caving Internal periphery figure in fruit caving Internal periphery area, fruit caving Internal periphery width and
Fruit caving Internal periphery length, and the fruit caving Internal periphery area data being calculated is calculated by a certain proportion of scaling, to count
Calculation obtains preset halogen flow.Pass through the fruit caving Internal periphery area, fruit caving Internal periphery width and fruit caving Internal periphery length etc. of acquisition
Data information, and the fruit caving Internal periphery figure obtained is combined to calculate the central point of betel nut intracavitary profile, and by central point and Bin
Bulky fruit caving Internal periphery two-end-point is attached, to obtain preset halogen path.
Step S810：Controlled motion module is moved according to preset halogen path, and corresponding control is positioned at described
Halogen module of ordering in motion module order halogen operation according to the preset halogen flow.
It should be noted that after calculating the preset halogen flow and preset halogen flow that obtain each betel nut, controlled motion
Module is moved according to preset halogen path, to make to order halogen module to be moved above corresponding betel nut fruit caving,
While movement, controlled motion module order halogen control according to obtained preset halogen data on flows information to ordering halogen module
Operation so that halogen module of ordering clicks and enters brine in betel nut intracavitary, thus reaches and is accurately controlled to a brine amount so that betel nut is given birth to
Yield and quality is improved, meanwhile, it is precisely controlled to order halogen path, the case where preventing to order halogen position offset.
As shown in Fig. 2, a kind of brine-adding system 10 includes：Image capture module 200, image processing module 300, computing module
400, motion module 500, order halogen module 600 and control module 700.
Image capture module is used to acquire the present image for the charging tray for being placed with multiple betel nuts.Image processing module for pair
Present image carries out target area selection, obtains target image；Image processing module is additionally operable to establish the parameter mould of target image
Type obtains two-value picture for carrying out parameter threshold processing to target image；Image processing module is additionally operable to according to default outer
Outline identification condition carries out the identification operation of betel nut outer profile, to obtain betel nut outer profile distribution map to two-value picture；Image procossing
Module is additionally operable to, respectively according to the boundary rectangle of each betel nut outer profile, piecemeal cutting process be carried out to betel nut outer profile distribution map,
To obtain multiple independent betel nut outer profile figures.
It should be noted that image capture module acquisition is placed with multiple present images of the charging tray of multiple betel nuts, image
Processing module carries out the predeterminable area selection according to the mark point on betel nut charging tray to each present image, obtains multiple first
Each initial pictures are carried out image mosaic operation, obtain the target image by beginning image.It is placed with the charging tray of multiple betel nuts
It is that progress image taking at image capture module is sent to by transport mechanism in actual production process, for example, conveyer
Structure can be that transport mechanisms, the image capture modules such as conveyer belt or transmission chain can be industrial camera or CCD imaging mechanisms etc.
Image capture module.Since betel nut charging tray has certain volume size so that image taking mechanism is carried out to betel nut charging tray
It needs the betel nut charging tray repeatedly passed through to transmission to shoot when image taking obtains, location hole is previously provided on betel nut charging tray
Or locating slot etc. be used as mark point, image taking mechanism according to mark point on betel nut charging tray as betel nut charging tray center into
Row is continuously shot, and thus obtains multiple initial pictures of betel nut charging tray, and then image processing module is according to the label of betel nut charging tray
Point-to-points a initial pictures carry out coincidence splicing, thus obtain the image with complete betel nut charging tray, i.e. betel nut charging tray is worked as
Preceding image.
After obtaining the present image of betel nut charging tray, image processing module is by using the parameter model of target image to institute
It states target image and carries out parameter threshold processing, to obtain the two-value picture of the target image.For example, being joined using saturation degree
The target image is carried out picture saturation degree extraction, is handled by saturation analysis, the target image at this time by exponential model
For the two-value picture that saturation degree is clearly demarcated, for example, only retaining the two-value picture of Black-White, enabling simpler accurately in institute
It states and extracts required betel nut outer profile pictorial information in two-value picture.
After obtaining the two-value picture, image processing module by comparing each object pattern in the two-value picture face
The features such as product, length, width, filter smaller interfering object feature (for example, in picture equipment outer profile feature or Bin
Outer profile feature of bulky charging tray etc.), to obtain each betel nut foreign steamer in regime values range in the target image
Exterior feature figure, can so, it is possible to obtain betel nut outer profile clear from each betel nut outer profile in the position of the two-value picture
Distribution map.
When obtaining betel nut outer profile distribution map, image processing module carries out piecemeal cutting to the betel nut outer profile distribution map
Processing, to obtain multiple independent betel nut outer profile figures.In order to enable each independent betel nut outer profile figure into
When row segmentation, it is ensured that the edge of the betel nut outer profile in each betel nut outer profile figure, which will not be divided, to be removed, in the external square
Amplify 5~10 pixels at the boundary position of shape outward, then piecemeal cutting process carried out to the betel nut outer profile distribution map,
To obtain multiple independent betel nut outer profile figures.Specific dividing method is the external square of minimum for calculating betel nut outer profile
Shape carries out cutting drawing piece after then amplifying 5~10 pixels outward at the boundary position of the boundary rectangle, thus makes to cut
The every pictures for cutting out all include a betel nut.
Default computation model is prestored in computing module, computing module is used to respectively calculate betel nut outer profile figure,
Obtain the fruit caving Internal periphery in each betel nut outer profile figure；Computing module is additionally operable to calculate separately the fruit caving in each betel nut outer profile figure
The target size of Internal periphery；Computing module is additionally operable to be obtained according to the target size of the fruit caving Internal periphery in each betel nut outer profile figure
Preset halogen path and preset halogen flow.
It should be noted that default computation model is the training for the charging tray for being placed with multiple betel nuts by industrial camera acquisition
Image, after the training image for obtaining the charging tray of these betel nuts in batches, by manually to the betel nut intracavitary wheel in these training images
Exterior feature is marked, and is normalized, and thus obtains multiple processing images and believes to get to the data of betel nut intracavitary profile
Breath, for example, the area information of betel nut intracavitary profile, length information, width information, saturation infromation and chroma-luminance information
Deng.After completing normalized, convolutional neural networks training operation is carried out according to each processing image, is obtained described pre-designed
Calculate model；Also that is, by the area information of the betel nut intracavitary profile in each processing image, length information, width information, satisfying
Sorted out and be integrated into related to betel nut intracavitary outline data information to the degree data informations such as information and chroma-luminance information
The data network of connection, i.e. convolutional neural networks, wherein the convolutional neural networks are based on residual error neural network and full convolutional Neural
The network architecture, including 60 convolutional layers.By the default computation model that convolutional neural networks training operation obtains after optimization,
With registration, calculation amount is small, and speed is fast, the advantages that being easy to train.So after obtaining the betel nut outer profile figure, meter
It calculates module and calculation processing is carried out to the betel nut outer profile figure by default computation model, to obtain each betel nut outer profile
Fruit caving Internal periphery in figure.After obtaining fruit caving Internal periphery, computing module is by calculating the fruit caving lubrication groove in fruit caving Internal periphery figure
Profile surface product, fruit caving Internal periphery width and fruit caving Internal periphery length, and by the fruit caving Internal periphery area data being calculated by one
The scaling of certainty ratio calculates, to which preset halogen flow be calculated.Computing module passes through the fruit caving Internal periphery area of acquisition, fruit
The data informations such as intracavitary profile width and fruit caving Internal periphery length, and the fruit caving Internal periphery figure obtained is combined to calculate betel nut fruit caving
The central point of Internal periphery, and central point and betel nut intracavitary profile two-end-point are attached, to obtain preset halogen path.
Motion module is moved according to preset halogen path；Halogen module is selected to be located in motion module, order halogen module according to
Preset halogen flow order halogen operation.Control module be respectively used to controlled motion module and order halogen module execute motor performance with
Select halogen operation.
It should be noted that after calculating the preset halogen flow and preset halogen flow that obtain each betel nut, control module
It is moved according to the preset halogen path clustering motion module, orders halogen module in correspondence to make to be located in motion module
Betel nut fruit caving above moved, while movement, control module is according to obtained preset halogen data on flows information pair
It selects halogen module and order halogen control operation so that halogen module of ordering clicks and enters brine in betel nut intracavitary, thus reaches to a brine amount
Accurately control so that the betel nut quality of production is improved, meanwhile, it is precisely controlled to order halogen path, prevents to order halogen position
The case where setting offset.
In one embodiment, control module is XYZ three-axis moving modules, for driving halogen module progress tri- axis directions of XYZ of ordering
Reciprocating type displacement operation, thus enable to order to carry out a brine operation in halogen module flexible motion to corresponding betel nut nucleome,
It can carry out ordering halogen movement according to preset halogen path simultaneously.
In another embodiment, brine-adding system 10 further includes brine supply module 800, brine supply module 800 and halogen mould of ordering
Block 600 be connected to, brine supply module 800 using circuit cycle mode be used for order halogen module 600 input brine.Select halogen module
The length that 600 brine amount is opened and closed the time by solenoid valve controls.
On the one hand can be to order halogen module 600 to provide brine it should be noted that by the way that brine supply module 800 is arranged,
Invocation point halogen module 600 is set to be normally carried out a brine operation；On the other hand, brine supply module 800 uses circuit cycle side
Formula is used to input brine to ordering halogen module 600, to enable brine in brine supply module 800 and order between halogen module 600
Continuous shuttling movement is carried out, thereby guarantees that the brine in brine supply module 800 still is able under prolonged working environment
Set temperature value is kept, prevents brine from issuing the matter that changes in prolonged working environment, and be conducive to the cleaning replacement of brine
Operation.
Compared with prior art, the present invention has at least the following advantages：
The present invention order halogen method use image processing method, by charging tray betel nut nucleome carry out Image Acquisition,
Analyzing processing and calculating, so as to obtain selecting brine amount and ordering halogen path for each betel nut nucleome on charging tray, thus in reality
Betel nut order and can be accurately controlled to selecting brine amount and order halogen path in halogen production so that betel nut orders the matter that halogen processes
Amount effectively improves.
In order to be further described the above-mentioned inventive concept for ordering halogen method, specific embodiment is given below and is further discussed
It states.
Specific embodiment is as follows：
Brine-adding system is always divided into two large divisions:Image processing section and motion control portion.For image processing section,
Its workflow is as follows：It is first to prepare for work, betel nut picture is obtained by industrial camera, obtains original image.Batch obtains
After obtaining these pictures, by manually the betel nut fruit caving profile in image is marked, and it is normalized.It handles well
Picture be used for train one can automatic identification fruit caving profile convolutional neural networks.The network is based on residual error neural network
(ResNet) and full convolutional network (FCN) framework, containing 60 convolutional layers, but by optimization, there is registration, calculation amount is small,
Speed is fast, the advantages that being easy to train.Secondly, real time picture is obtained by industrial camera, sets area-of-interest, passes through plate
Positioning round orifice finds the center of plate.The saturation degree channel of picture HSV models is extracted, and by picture thresholding, is obtained
To two-value picture.It is analyzed by blobs, picture includes the outer contour of object within the vision at this time.More each blob
The features such as area, length, width, the smaller interference blob of filtering obtains the betel nut outer profile blob of normal range (NR).Then,
The picture that upper step is obtained carries out piecemeal processing.Specific dividing method is the minimum enclosed rectangle for calculating betel nut outer profile,
Amplify 5-10 pixel outward and cut picture, picture every small in this way includes a betel nut.And then, respectively to each small
Betel nut picture is handled, and enhances picture contrast using Retinex algorithm for image enhancement.The small picture that upper step is obtained
It inputs neural network and carries out propagated forward (Forward) calculating, obtain betel nut intracavitary profile.Finally, Internal periphery area is calculated,
In conjunction with information such as length, width, by area data by a certain proportion of scaling obtain ordering halogen amount number.
For the motion control portion of brine-adding system, main flow is as follows：Initial phase is carried out first, brine-adding system
Agitating shaft starts to act, pan feeding optoelectronic induction to plate, and pan feeding axis drives belt to start operation.Optoelectronic induction is imaged to plate,
Triggering camera is taken pictures three times at this time, is taken pictures interval depending on plate position, is usually moved 1/3 time of the length of plate
(variable).Halogen optoelectronic induction is selected to plate, supercharging axis pressurization, halogen syringe needle of ordering ejection brine.When brine amount is opened and closed by solenoid valve
Between length control.
Several embodiments of the invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the protection of the present invention
Range.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of halogen method of ordering, which is characterized in that include the following steps：
Acquisition is placed with the present image of the charging tray of multiple betel nuts, carries out target area selection to the present image, obtains mesh
Logo image；
The parameter model of the target image is established, for carrying out parameter threshold processing to the target image, obtains two-value
Picture；
Condition is identified according to default outer profile, the identification operation of betel nut outer profile is carried out to the two-value picture, to obtain outside betel nut
Contoured profile figure；
Respectively according to the boundary rectangle of each betel nut outer profile, piecemeal cutting process is carried out to the betel nut outer profile distribution map, with
Obtain multiple independent betel nut outer profile figures；
According to default computation model, the betel nut outer profile figure is calculated respectively, is obtained in each betel nut outer profile figure
Fruit caving Internal periphery；
Calculate separately the target size of the fruit caving Internal periphery in each betel nut outer profile figure；
Preset halogen path and preset halogen are obtained according to the target size of the fruit caving Internal periphery in each betel nut outer profile figure
Flow；
Controlled motion module is moved according to preset halogen path, and corresponding control is located at the point in the motion module
Halogen module order halogen operation according to the preset halogen flow.
2. halogen method according to claim 1 of ordering, which is characterized in that the multiple of charging tray that acquisition is placed with multiple betel nuts work as
Preceding image carries out the predeterminable area selection to each present image according to the mark point on betel nut charging tray, obtains multiple first
Each initial pictures are carried out image mosaic operation, obtain the target image by beginning image.
3. halogen method according to claim 1 of ordering, which is characterized in that the default computation model obtains as follows
：
Acquisition is placed with the training image of the charging tray of multiple betel nuts；
Fruit caving Internal periphery marking operation is carried out to each training image, and is normalized, multiple processing images are obtained；
Convolutional neural networks training operation is carried out according to each processing image, obtains the default computation model；Wherein, described
Convolutional neural networks are based on residual error neural network and full convolutional neural networks framework, including 60 convolutional layers.
4. halogen method according to claim 1 of ordering, which is characterized in that carry out calculating it to the betel nut outer profile figure
Before, further include following steps：
Picture superposition processing is carried out to the betel nut outer profile figure.
5. halogen method according to claim 1 of ordering, which is characterized in that the parameter model of the target image is joined for saturation degree
Exponential model.
6. halogen method according to claim 1 of ordering, which is characterized in that in the betel nut outer profile distribution map, according to each
The boundary rectangle of betel nut outer profile amplifies 5~10 pixels outward at the boundary position of the boundary rectangle, then to described
Betel nut outer profile distribution map carries out piecemeal cutting process, to obtain multiple independent betel nut outer profile figures.
7. halogen method according to claim 1 of ordering, which is characterized in that the target size of the fruit caving Internal periphery includes fruit caving
Internal periphery area, fruit caving Internal periphery width and fruit caving Internal periphery length, according to the fruit caving in each betel nut outer profile figure
Internal periphery area, the fruit caving Internal periphery width and the fruit caving Internal periphery length, to obtain the preset halogen flow.
8. a kind of brine-adding system, which is characterized in that including：
Image capture module, described image acquisition module are used to acquire the present image for the charging tray for being placed with multiple betel nuts；
Image processing module, described image processing module are used to carry out target area selection to the present image, obtain target
Image；Described image processing module is additionally operable to establish the parameter model of the target image, for being carried out to the target image
Parameter thresholdization processing, obtains two-value picture；Described image processing module is additionally operable to according to outer profile identification condition is preset, to institute
It states two-value picture and carries out the identification operation of betel nut outer profile, to obtain betel nut outer profile distribution map；Described image processing module is also used
According to the boundary rectangle of each betel nut outer profile, piecemeal cutting process is carried out to the betel nut outer profile distribution map respectively, to obtain
To multiple independent betel nut outer profile figures；
Computing module, the default computation model is prestored in the computing module, and the computing module is used for respectively to described
Betel nut outer profile figure is calculated, and the fruit caving Internal periphery in each betel nut outer profile figure is obtained；The computing module is additionally operable to
Calculate separately the target size of the fruit caving Internal periphery in each betel nut outer profile figure；The computing module is additionally operable to according to each institute
The target size for stating the fruit caving Internal periphery in betel nut outer profile figure obtains preset halogen path and preset halogen flow；
Motion module, the motion module are moved according to preset halogen path；
Halogen module is selected, the halogen module of ordering is located in the motion module, and the halogen module of ordering is according to the preset halogen flow
Order halogen operation；
Control module, the control module be respectively used to control the motion module and it is described order halogen module execute motor performance and
Select halogen operation.
9. brine-adding system according to claim 1, which is characterized in that the control module is XYZ three-axis moving modules, is used
It is operated in driving the reciprocating type displacement for ordering halogen module progress tri- axis directions of XYZ.
10. brine-adding system according to claim 1, which is characterized in that the brine-adding system further includes brine supply module,
The brine supply module is connected to the halogen module of ordering, and the brine supply module is used for using circuit cycle mode to described
Select halogen module input brine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810439169.5A CN108634241A (en) | 2018-05-09 | 2018-05-09 | Select halogen method and its brine-adding system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810439169.5A CN108634241A (en) | 2018-05-09 | 2018-05-09 | Select halogen method and its brine-adding system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108634241A true CN108634241A (en) | 2018-10-12 |
Family
ID=63753771
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810439169.5A Withdrawn CN108634241A (en) | 2018-05-09 | 2018-05-09 | Select halogen method and its brine-adding system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108634241A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109704005A (en) * | 2019-02-15 | 2019-05-03 | 武汉菲舍控制技术有限公司 | The deviation correcting device and method for correcting error of production line, V belt translation equipment, V belt translation equipment |
| CN110338378A (en) * | 2019-08-01 | 2019-10-18 | 苏州感知线智能科技有限公司 | It is a kind of for betel nut order halogen vision to just with bootstrap technique and device |
| CN112438415A (en) * | 2019-09-02 | 2021-03-05 | 惠州市三协精密有限公司 | Cavity entering method and system for preserved areca nuts and storage medium |
-
2018
- 2018-05-09 CN CN201810439169.5A patent/CN108634241A/en not_active Withdrawn
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109704005A (en) * | 2019-02-15 | 2019-05-03 | 武汉菲舍控制技术有限公司 | The deviation correcting device and method for correcting error of production line, V belt translation equipment, V belt translation equipment |
| CN109704005B (en) * | 2019-02-15 | 2024-05-31 | 武汉菲舍控制技术有限公司 | Production line, belt transmission equipment, deviation correcting device of belt transmission equipment and deviation correcting method |
| CN110338378A (en) * | 2019-08-01 | 2019-10-18 | 苏州感知线智能科技有限公司 | It is a kind of for betel nut order halogen vision to just with bootstrap technique and device |
| CN112438415A (en) * | 2019-09-02 | 2021-03-05 | 惠州市三协精密有限公司 | Cavity entering method and system for preserved areca nuts and storage medium |
| CN112438415B (en) * | 2019-09-02 | 2023-11-24 | 惠州市三协精密有限公司 | Betel nut preserved fruit cavity entering method, betel nut preserved fruit cavity entering system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aquino et al. | vitisBerry: An Android-smartphone application to early evaluate the number of grapevine berries by means of image analysis | |
| CN108634241A (en) | Select halogen method and its brine-adding system | |
| CN109410168A (en) | For determining the modeling method of the convolutional neural networks model of the classification of the subgraph block in image | |
| CN102704215B (en) | Automatic cutting method of embroidery cloth based on combination of DST file parsing and machine vision | |
| CN102214306B (en) | Leaf disease spot identification method and device | |
| CN113012150A (en) | Feature-fused high-density rice field unmanned aerial vehicle image rice ear counting method | |
| CN104132897B (en) | A kind of nitrogenous measuring method of plant leaf blade based on handheld device and device | |
| CN110245593A (en) | A kind of images of gestures extraction method of key frame based on image similarity | |
| CN106529006A (en) | Depth image-based broiler growth model fitting method and apparatus | |
| CN113313692B (en) | Automatic banana young plant identification and counting method based on aerial visible light image | |
| CN112767382A (en) | Fry counting method based on deep learning | |
| CN108311411A (en) | A kind of cordyceps sinensis intelligence sorting system and its application method | |
| CN106803257A (en) | The dividing method of scab in a kind of crop disease leaf image | |
| CN107194348A (en) | The domain color recognition methods of target object in a kind of image | |
| CN110827304A (en) | Traditional Chinese medicine tongue image positioning method and system based on deep convolutional network and level set method | |
| CN113269191A (en) | Crop leaf disease identification method and device and storage medium | |
| CN113887389A (en) | Quality inspection and classification method for traditional Chinese medicine decoction pieces based on image recognition | |
| CN111369497B (en) | Walking type tree fruit continuous counting method and device | |
| CN105195438A (en) | Embedded type automatic pearl sorting device and method based on image recognition | |
| CN118053154A (en) | Oyster mushroom growth monitoring method and device, electronic equipment and storage medium | |
| CN111738310B (en) | Material classification method, device, electronic equipment and storage medium | |
| CN110414369A (en) | A kind of training method and device of ox face | |
| CN109840498A (en) | A kind of real-time pedestrian detection method and neural network, target detection layer | |
| CN112966698A (en) | Freshwater fish image real-time identification method based on lightweight convolutional network | |
| CN107491459A (en) | The search method and device of three-dimensional image |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181012 |
|
| WW01 | Invention patent application withdrawn after publication |