CN108765372A - Identify the method, apparatus and computer readable storage medium of shoe information - Google Patents
Identify the method, apparatus and computer readable storage medium of shoe information Download PDFInfo
- Publication number
- CN108765372A CN108765372A CN201810383564.6A CN201810383564A CN108765372A CN 108765372 A CN108765372 A CN 108765372A CN 201810383564 A CN201810383564 A CN 201810383564A CN 108765372 A CN108765372 A CN 108765372A
- Authority
- CN
- China
- Prior art keywords
- information
- shoes
- image
- shoe
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000003708 edge detection Methods 0.000 claims abstract description 11
- 210000005036 nerve Anatomy 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 abstract description 19
- 230000009467 reduction Effects 0.000 description 15
- 238000003062 neural network model Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000011946 reduction process Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 4
- 238000009499 grossing Methods 0.000 description 3
- 238000005406 washing Methods 0.000 description 3
- 206010049244 Ankyloglossia congenital Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a kind of methods of identification shoe information, belong to shoes automatic cleaning technology field.This method includes：By a plurality of image collecting device, obtain the original image for being placed into the shoes on rotating platform, edge detection process carried out to original image, determine include shoe information feature image, wherein shoe information includes：It is one or more among brand message, type information, length information, width information and elevation information；Shoe information in identification feature picture determines one or more among brand message, type information, length information, width information and the elevation information of shoes.Shoe information can be detected automatically, and the shoe information detected can provide information support for automatic shoe cleaning equipment.The embodiment of the present invention additionally provides a kind of device and computer readable storage medium of identification shoe information.
Description
Technical field
The present invention relates to shoes automatic cleaning technology field, more particularly to the method, apparatus of a kind of identification shoe information and
Computer readable storage medium.
Background technology
With the development of the progress and artificial intelligence of science and technology, intelligent algorithm is also more and more applied to daily life
In.But in existing automatic shoe cleaning equipment, lack the device that shoe information is identified, to be washed according to one kind
Shoes method cleans all shoes, is easily damaged shoes or does not wash clean clearly.
Invention content
An embodiment of the present invention provides a kind of methods of identification shoe information, can provide information branch for automatic shoe cleaning equipment
Support.
In order to the embodiment to disclosure some aspects there are one basic understanding, simple summary is shown below.It should
Summarized section is not extensive overview, nor to determine key/critical component or describe the protection domain of these embodiments.
Its sole purpose is that some concepts are presented with simple form, in this, as the preamble of following detailed description.
According to a first aspect of the embodiments of the present invention, a kind of method of identification shoe information is provided.
In some optional embodiments, this method includes：
By a plurality of image collecting device, the original image for being placed into the shoes on rotating platform is obtained；
To the original image carry out edge detection process, determine include shoe information feature image, wherein it is described
Shoe information includes：It is one or more among brand message, type information, length information, width information and elevation information；
It identifies the shoe information in the feature image, determines the brand message, type information, length of the shoes
It is one or more among information, width information and elevation information.
In some optional embodiments, the original image for obtaining the shoes for being placed into rotating platform, including：
Rotate the rotating platform；
The one or more the first original images of the shoes are obtained by the image collecting device of first position and are stored；
The one or more the second original images of the shoes are obtained by the image acquiring device of the second position and are stored.
In some optional embodiments, the method further includes：
The three-dimensional of the shoes is built according to one or more the first picture and one or more second picture
Model；
Store the threedimensional model of the shoes.
In some optional embodiments, the shoe information in the identification feature image determines the shoes
It is one or more among the brand message of son, type information, length information, width information and elevation information, including following two
It is one or more among kind step：
The feature image is input to the default artificial nerve network model, determine the shoes brand message and
It is one or more among type information；
The distance between the object edge in the feature image is obtained, determines length information, the width letter of the shoes
It is one or more in breath and elevation information.
According to a second aspect of the embodiments of the present invention, a kind of device of identification shoe information is provided.
In some optional embodiments, the device of the identification shoe information includes：
Image collection module, for by a plurality of image collecting device, obtaining the shoes being placed on rotating platform
The original image of son；
Detection module, for the original image carry out edge detection process, determine include shoe information feature
Picture, wherein the shoe information includes：Among brand message, type information, length information, width information and elevation information
It is one or more；
Identification module, the shoe information in the feature image for identification, determine the shoes brand message,
It is one or more among type information, length information, width information and elevation information.
In some optional embodiments, described image acquisition module includes：
Driving unit, for rotating the rotating platform；
First image acquisition unit obtains one or more the of the shoes by the image collecting device of first position
One original image simultaneously stores；
Second image acquisition unit obtains one or more the of the shoes by the image acquiring device of the second position
Two original images simultaneously store.
Further include three-dimensional modeling module in some optional embodiments, for according to one or more the first figure
Piece and one or more second picture build the threedimensional model of the shoes；Store the threedimensional model of the shoes.
In some optional embodiments, the identification module includes the one or more of following two recognition units：
First identification module determines institute for the feature image to be input to the default artificial nerve network model
State the brand message of shoes and one or more among type information；
Second identification module, for obtaining the distance between object edge in the feature image, determine the shoes
It is one or more in length information, width information and elevation information.
In some optional embodiments, the device of the identification shoe information includes：
Processor；
Memory for storing executable instruction；
Wherein, the processor is configured as：
By a plurality of image collecting device, the original image for being placed into the shoes on rotating platform is obtained；
To the original image carry out edge detection process, determine include the shoe information feature image, wherein
The shoe information includes：One kind or more among brand message, type information, length information, width information and elevation information
Kind；
It identifies the shoe information in the feature image, determines the brand message, type information, length of the shoes
It is one or more among information, width information and elevation information.
According to a third aspect of the embodiments of the present invention, a kind of computer readable storage medium is provided.
In some optional embodiments, the computer readable storage medium is stored thereon with computer program, works as institute
State the method that the shoe information of identification hereinbefore is realized when computer program is executed by processor.
The embodiment of the present invention has an advantageous effect in that：Brand message, type information, the length letter of shoes can be detected automatically
Breath, width information and elevation information etc., the shoe information being detected above can provide information support for automatic shoe cleaning equipment, to
Automatic shoe cleaning equipment can be used a variety of method for washing shoe and be cleaned to shoes, and that not only washes is clean, be also less prone to damage shoes.
It should be understood that above general description and following detailed description is only exemplary and explanatory, not
It can the limitation present invention.
Description of the drawings
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the present invention
Example, and be used to explain the principle of the present invention together with specification.
Fig. 1 is a kind of flow diagram of identification shoe information shown according to an exemplary embodiment；
Fig. 2 is a kind of position view of image collecting device shown according to an exemplary embodiment；
Fig. 3 is a kind of flow diagram of acquisition original image shown according to an exemplary embodiment；
Fig. 4 is a kind of flow diagram of establishment threedimensional model shown according to an exemplary embodiment；
Fig. 5 is that a kind of flow of feature image for determining to include shoe information shown according to an exemplary embodiment is shown
It is intended to；
Fig. 6 is the flow signal of the edge feature in a kind of acquisition noise reduction picture shown according to an exemplary embodiment
Figure；
Fig. 7 is a kind of schematic diagram of the gradient direction of pixel shown according to an exemplary embodiment；
Fig. 8 is that a kind of flow for training default artificial nerve network model shown according to an exemplary embodiment is illustrated
Figure；
Fig. 9 is a kind of block diagram of the device of identification shoe information shown according to an exemplary embodiment.
Specific implementation mode
The following description and drawings fully show specific embodiments of the present invention, to enable those skilled in the art to
Put into practice them.Embodiment only represents possible variation.Unless explicitly requested, otherwise individual components and functionality is optional, and
And the sequence of operation can change.The part of some embodiments and feature can be included in or replace other embodiments
Part and feature.The range of embodiment of the present invention includes the entire scope of claims and the institute of claims
There is obtainable equivalent.Herein, each embodiment can individually or generally be indicated that this is only with term " invention "
It is merely for convenience, and if in fact disclosing the invention more than one, be not meant to automatically limit the range of the application
For any single invention or inventive concept.Herein, relational terms such as first and second and the like are used only for one
Entity, which either operates to distinguish with another entity or operation, to be existed without requiring or implying between these entities or operation
Any actual relationship or sequence.Moreover, the terms "include", "comprise" or its any other variant be intended to it is non-exclusive
Property include, so that process, method or equipment including a series of elements include not only those elements, but also to include
Other elements that are not explicitly listed.Each embodiment herein is described by the way of progressive, and each embodiment stresses
Be all difference from other examples, just to refer each other for identical similar portion between each embodiment.For implementing
For structure, product etc. disclosed in example, since it is corresponding with part disclosed in embodiment, so fairly simple, the phase of description
Place is closed referring to method part illustration.
According to a first aspect of the embodiments of the present invention, a kind of method of identification shoe information is provided.
As shown in Figure 1, in some optional embodiments, the method for identifying shoe information includes：
S101, by a plurality of image collecting device, obtain the original image for being placed into the shoes on rotating platform；
S102, to original image carry out edge detection process, determine include shoe information feature image, wherein shoes
Sub-information includes：It is one or more among brand message, type information, length information, width information and elevation information；
Shoe information in S103, identification feature picture determines brand message, type information, length information, the width of shoes
It is one or more among degree information and elevation information.
Brand message, type information, length information, width information and the height that the technical program can detect shoes automatically are believed
Breath etc., the information being detected above can provide information support for automatic shoe cleaning equipment, thus automatic shoe cleaning equipment can be used it is a variety of
Method for washing shoe cleans shoes, and that not only washes is clean, is also less prone to damage shoes.
Wherein, the brand message of shoes includes that may include that female brand and sub-brand name, female brand refer to total brand of shoes,
Refer to producing that the brand of the company of the shoes, sub-brand name refer to the brand of certain a series of shoes in some cases.Shoes
Son type include but not limited to frenulum sports and leisure (skin) shoes, without frenulum sports and leisure (skin) shoes, pumps, middle side shoes, boots,
High-heeled shoes, cooling slippers etc..
The width information of shoes is determined for example, by using the technical program, and the width information is transmitted to and implements this method
The automatic shoe cleaning equipment that used equipment is connected, automatic shoe cleaning equipment can adjust fixture according to the width information of shoes
Width ensures that shoes clean up, while shoes being avoided to be damaged.
Image collecting device in S101 can be camera.Rotating platform is that horizontal positioned can revolve along the horizontal plane
The device turned, when shoes are placed on rotating platform, if rotating platform rotates, shoes can be with rotating platform common rotation.
The image collecting device in fixed position, which is arranged, can obtain the original image of shoes of different angle.
As shown in Fig. 2, in some alternative embodiments, image collecting device includes the image being arranged in first position
Harvester 11 and the image collecting device 12 in the second position is set；Wherein, first position is located at the top of rotating platform 13,
The image collecting device 11 in first position, which is arranged, can obtain the original image of shoes of depression angle, and the second position is located at rotation
The side of platform 13, the image collecting device 12 in the second position, which is arranged, can obtain the original image of shoes of side view angle.?
The image collecting device being arranged in the second position in present embodiment refers to the original graph that can acquire the shoes of side view angle
The position of piece a, however it is not limited to position, an image collecting device.Optionally, the image collector in the second position is set
It is set to two, wherein two image collecting devices are oppositely arranged；Optionally, the image collector in the second position is arranged to be set to
Three, wherein three image collecting devices surround and form equilateral triangle.
As shown in figure 3, in some optional embodiments, the original image for the shoes for being placed into rotating platform, packet are obtained
It includes：
S301, the rotating platform is rotated；
S302, the one or more the first original images that shoes are obtained by the image collecting device of first position are simultaneously deposited
Storage；
S303, the one or more the second original images that shoes are obtained by the image acquiring device of the second position are simultaneously deposited
Storage.
The original image of the shoes of different angle can be obtained using the technical program, the picture of these different angles can be entirely square
The state (cleannes, damaged condition etc.) of the display shoes of position before cleaning.During daily cleaning shoes, often
Because whether shoes occur damaging and generate dispute in cleaning process, and after using the technical program, to shoes before cleaning shoes
The photo of son is stored, you can is backed up to the damaged condition of the shoes before cleaning, if occurring entangling about what shoes damaged
Confusingly, by compare the shoes before the cleaning that has backed up damaged condition and cleaning after shoes damaged condition, you can assert
Responsible party solves unnecessary dispute.
In some alternative embodiments, the sum of the first original image and the second original image is 6~12.Example
Such as：6,8,10,12.
In some alternative embodiments, the rotation sample platform is rotated, including：The rotary flat is at the uniform velocity rotated with setting speed
Platform；Or, rotating the rotation sample platform with set angle stepping.Set angle is related with the quantity of original image for needing to shoot, and needs
The quantity for the original image wanted is bigger, and set angle is smaller, and the quantity of the original image needed is smaller, and set angle is bigger.It can
Selection of land, set angle are 30 °~60 °.
As shown in figure 4, for the further state of accurate backup shoes before cleaning, in some optional embodiments,
Further include：
S401, the threedimensional model that shoes are built according to the one or more the first pictures and one or more second picture；
S402, the threedimensional model for storing shoes.
Using the technical program, the original image of shoes is synthesized to the threedimensional model of shoes, can more intuitively be reflected clear
The state of shoes, the usage experience of user are good before washing.
As shown in figure 5, for S102, in some alternative embodiments, edge detection process is carried out to original image,
Determine include shoe information feature image, including：
S501, noise reduction process is carried out to original image to obtain noise reduction picture；
Edge feature in S502, acquisition noise reduction picture；
Object edge in S503, selected characteristic edge obtains feature image.
The technical program can accurately obtain the feature image of shoes.Noise in original image concentrates on high-frequency signal, very
It is easily recognizable as pseudo-edge, so in S501, needs to carry out noise reduction process to original image.Optionally, flat using Gauss
Sliding filtering and noise reduction method carries out noise reduction process to original image, reduces the probability for pseudo-edge occur.
Wherein, Gaussian smoothing filter goes dry method to can remove high-frequency noise, and since image edge information is also high-frequency signal,
Therefore need while removing high-frequency noise, retain the high-frequency signal of image border.Still optionally further, in Gaussian smoothing filter
In Denoising Algorithm, it is filtered using setting Gaussian filter core, wherein set Gaussian filter core and refer to size as (2k+1)
The Gaussian filter core of × (2k+1), also, k is positive integer.The Gaussian filter core is obtained, including：
Wherein, 1≤i, j≤(2k+1).
Such as σ=1.4, k=1, Gaussian filter core is at this time：
Noise reduction process is carried out to original image using Gaussian smoothing Denoising Algorithm, including：Use Gaussian filter core and image
Carry out convolution；Its convolution results is exactly filtered image.
For example, Gaussian filter core is：
The window that will be filtered is：
Pixel is obtained after filtering is：
Wherein * is convolution symbol.
As shown in fig. 6, in S502, the edge feature in noise reduction picture is obtained, including：
S601, the detection direction for determining setting pixel；
Wherein, detection direction includes horizontal direction, vertical direction and diagonally opposed.
S602, the gradient intensity and gradient direction for setting pixel in detection direction are obtained；
Wherein, each detection direction includes both direction, such as horizontal edge includes to the left and to the right both direction；
Gradient intensity is：
Gradient direction is：
By taking window A as an example,
The pixel that sets in window A is in the horizontal direction with the gradient intensity of vertical direction as pixel e：
Wherein, GxFor pixel e gradient intensities in the horizontal direction, SxFor the setting operator in horizontal direction；
GyIt is pixel e in the gradient intensity of vertical direction, SyFor the upper setting operator of vertical direction.
The gradient intensity of pixel e is：
The gradient direction of pixel e is：
The gradient of S603, the gradient intensity that pixel will be set and two pixels on positive gradient direction and negative gradient direction
Intensity is compared；
Wherein, optionally, to be compared to determine using linear difference between two adjacent pixels of cross-domain gradient direction
Pixel gradient.Such as：
As shown in fig. 7, gradient is divided into 8 directions, respectively with E, NE, N, NW, W, SW, S, SE, wherein 0 represent 0 °~
45 °, 1 represents 45 °~90 °, and 2 represent -90 °~-45 °, and 3 represent -45 °~0 °.The gradient direction of pixel P is θ, then pixel
The gradient linearity interpolation of P1 and P2 is：
Gp1=(1-tan θ) × E+tan θ × NE
Gp2=(1-tan θ) × W+tan θ × SW
S604, the gradient for being more than two pixels on positive gradient direction and negative gradient direction when the gradient of setting pixel are strong
When spending, to set pixel as the pixel on setting edge.
The gradient intensity of pixel P is GpIf Gp≥Gp1, and Gp≥Gp2, then pixel P be characterized the pixel on edge
Point.Otherwise, by the gradient intensity G of pixel PpIt sets to 0.
The pixel chosen by above method can indicate the edge feature in image, still, however it remains due to
Some edge pixel points caused by noise and color change.
In S503, the object edge in selected characteristic edge, optionally, when the gradient intensity of setting pixel is more than setting
When high threshold, which is labeled as strong edge pixel；
When the gradient intensity for setting pixel is less than setting high threshold and is more than setting Low threshold, by the setting pixel
Point is labeled as weak edge pixel point；
When the gradient intensity for setting pixel is less than setting Low threshold, the gradient intensity of the setting pixel is set to 0；
Using strong edge pixel as the pixel in object edge.
The technical program can distinguish the actual characteristic edge in original image and the pseudo-edge as caused by noise etc..Its
In, using the edge that strong edge pixel is formed as the actual characteristic edge in original image, it is with the edge that weak edge forms
The pseudo-edge as caused by noise etc..
Optionally, Low threshold is set as 0.05~0.2, sets high threshold as 0.25~0.4；
Optionally, Low threshold is set as 0.05, sets high threshold as 0.25；
Optionally, Low threshold is set as 0.1；High threshold is set as 0.3；
Optionally, Low threshold is set as 0.15；High threshold is set as 0.35；
Optionally, Low threshold is set as 0.2；High threshold is set as 0.4.
In practical application, in weak edge pixel point, there are still the pictures at the actual characteristic edge of characterizing part original image
Vegetarian refreshments, in order to pick out these pixels, in some alternative embodiments, when determining the weak edge pixel point and its adjacent
In pixel, there are when one or more strong edge pixels, using the weak edge pixel point as the pixel in object edge.
It can avoid omitting the pixel of composition object edge so that object edge is more complete.
In S502, the edge feature obtained in noise reduction picture is calculated using binaryzation in some alternative embodiments
Method handles noise reduction picture, obtains edge feature.Preferably, noise reduction picture is handled using Otsu algorithm, obtains edge feature.
It is corresponding when handling noise reduction picture using Binarization methods, in S503, take distance maximum in the horizontal direction
Two edges as object edge；In the vertical direction apart from maximum two edges as object element.
Identification is using the feature image acquired in the technical program, you can determine the length informations of shoes, width information and
It is one or more among elevation information.
In a kind of optional embodiment, in S503, the object edge in selected characteristic edge, including：To feature
Edge carries out morphology operations processing, obtains object edge.
In some optional embodiments, the shoe information in identification feature picture determines the brand, type, length of shoes
It is one or more including one or more among following two steps among degree, width and height：
The first step：Feature image is input to default artificial nerve network model, determines the brand and type of shoes
Among it is one or more；
Second of step：The distance between object edge in feature image is obtained, determines length, width and the height of shoes
In it is one or more.
In the first step, artificial nerve network model includes but not limited to：Convolutional neural networks model, cycle nerve
Network model and deep neural network model.Further, default artificial nerve network model includes but not limited to：Default convolution
Neural network model, preset loop neural network model and predetermined depth neural network model.
Default neural network model refers to the neural network model that training finishes.As shown in figure 8, in a kind of optional reality
It applies in mode, training nerve neural network model is included the following steps with obtaining default neural network model：
S801, the product that fisrt feature picture and fisrt feature picture are characterized are inputted into the first artificial nerve network model
One or both of board and type；
S802, the first identification error for obtaining the first artificial neural network model；
S803, when the first identification error is less than or equal to the first given threshold, determine default artificial nerve network model.
Can determine using the technical solution in the first step one kind among the brand message and type information of shoes or
It is a variety of.
According to a second aspect of the embodiments of the present invention, a kind of device of identification shoe information is provided.
As shown in figure 9, in some optional embodiments, the device of the identification shoe information includes：
Image collection module 21, for by a plurality of image collecting device, acquisition to be placed on rotating platform
The original image of shoes；
Detection module 22, for original image carry out edge detection process, determine include shoe information characteristic pattern
Piece, wherein shoe information includes：One kind among brand message, type information, length information, width information and elevation information
Or it is a variety of；
Identification module 23, the shoe information in feature image for identification, determine the brand messages of shoes, type information,
It is one or more among length information, width information and elevation information.
In some alternative embodiments, detection module includes：
Noise reduction unit carries out noise reduction process to obtain noise reduction picture to original image；
Feature unit obtains the edge feature in noise reduction picture；
Object element, the object edge in selected characteristic edge obtain feature image.
In some alternative embodiments, detection direction of the feature unit for determining setting pixel obtains setting picture
Element will set the gradient intensity of pixel and along positive gradient direction and negative gradient side in the gradient intensity and gradient direction of detection direction
The gradient intensity of two upward pixels is compared, when the gradient of setting pixel is more than along positive gradient direction and negative gradient direction
When the gradient intensity of two pixels of upper score, to set pixel as the pixel on setting edge.
In some alternative embodiments, object element is used to set high threshold when the gradient intensity of setting pixel is more than
When value, which is labeled as strong edge pixel；
When the gradient intensity for setting pixel is less than setting high threshold and is more than setting Low threshold, by the setting pixel
Point is labeled as weak edge pixel point；
When the gradient intensity for setting pixel is less than setting Low threshold, the gradient intensity of the setting pixel is set to 0；
Using strong edge pixel as the pixel in object edge.
Optionally, Low threshold is set as 0.05~0.2, sets high threshold as 0.25~0.4；
Optionally, Low threshold is set as 0.05, sets high threshold as 0.25；
Optionally, Low threshold is set as 0.1；High threshold is set as 0.3；
Optionally, Low threshold is set as 0.15；High threshold is set as 0.35；
Optionally, Low threshold is set as 0.2；High threshold is set as 0.4.
In some alternative embodiments, noise reduction unit is used to handle noise reduction picture using Binarization methods, obtains special
Levy edge.
Optionally, noise reduction picture is handled using Otsu algorithm, obtains edge feature.
In some alternative embodiments, object element is made for taking in the horizontal direction apart from maximum two edges
For object edge；In the vertical direction apart from maximum two edges as object element.
In some optional embodiments, image collection module includes：
Driving unit, for rotating the rotating platform；
First image acquisition unit obtains the one or more the first former of shoes by the image collecting device of first position
Beginning picture simultaneously stores；
Second image acquisition unit obtains the one or more the second former of shoes by the image acquiring device of the second position
Beginning picture simultaneously stores.
In some optional embodiments, identify that the device of shoe information further includes three-dimensional modeling module, for according to one
Or multiple first pictures and one or more second picture structure shoes threedimensional model；Store the threedimensional model of shoes.
In some optional embodiments, identification module includes the one or more of following two recognition units：
First recognition unit determines the brand of shoes for feature image to be input to default artificial nerve network model
It is one or more among information and type information；
Second recognition unit, for obtaining the distance between object element in feature image, determine shoes length information,
It is one or more in width information and elevation information.
In some optional embodiments, identify that the device of shoe information further includes training module, for artificial to first
One or both of brand and the type that fisrt feature picture and fisrt feature picture are characterized are inputted in neural network model,
The first identification error for obtaining the first artificial neural network model, when the first identification error is less than or equal to the first given threshold
When, determine default artificial nerve network model.
In some optional embodiments, identify that the device of shoe information includes：
Processor；
Memory for storing executable instruction；
Wherein, processor is configured as：
By a plurality of image collecting device, the original image for being placed into the shoes on rotating platform is obtained；
To original image carry out edge detection process, determine include shoe information feature image, wherein shoe information
Including：It is one or more among brand message, type information, length information, width information and elevation information；
Shoe information in identification feature picture determines brand message, type information, length information, the width letter of shoes
It is one or more among breath and elevation information.
Optionally, the method or apparatus of previously described identification shoe information can be realized in network side server, or
Person can also realize in the terminal, alternatively, being realized in dedicated control device.
In some optional embodiments, a kind of computer readable storage medium is provided, computer program is stored thereon with, when
The computer program realizes previously described method when being executed by processor.Above computer readable storage medium storing program for executing includes read-only
Memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), tape and
Light storage device etc..
Those of ordinary skill in the art may realize that lists described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, depends on the specific application and design constraint of technical solution.Those of skill in the art
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.It is apparent to those skilled in the art that for convenience and simplicity of description, foregoing description
The specific work process of system and device, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
It should be understood that the invention is not limited in the flow and structure that are described above and are shown in the accompanying drawings,
And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is only limited by the attached claims
System.
Claims (10)
1. a kind of method of identification shoe information, which is characterized in that including：
By a plurality of image collecting device, the original image for being placed into the shoes on rotating platform is obtained；
To the original image carry out edge detection process, determine include shoe information feature image, wherein the shoes
Information includes：It is one or more among brand message, type information, length information, width information and elevation information；
It identifies the shoe information in the feature image, determines brand message, type information, the length letter of the shoes
It is one or more among breath, width information and elevation information.
2. according to the method described in claim 1, it is characterized in that, the original graph for obtaining the shoes for being placed into rotating platform
Piece, including：
Rotate the rotating platform；
The one or more the first original images of the shoes are obtained by the image collecting device of first position and are stored；
The one or more the second original images of the shoes are obtained by the image acquiring device of the second position and are stored.
3. according to the method described in claim 2, it is characterized in that, further including：
The threedimensional model of the shoes is built according to one or more the first picture and one or more second picture；
Store the threedimensional model of the shoes.
4. according to the method described in claim 1, it is characterized in that, the shoes in the identification feature image are believed
Breath, determines one kind or more among brand message, type information, length information, width information and the elevation information of the shoes
Kind, including it is one or more among following two steps：
The feature image is input to the default artificial nerve network model, determines the brand message and type of the shoes
It is one or more among information；
Obtain the distance between the object edge in the feature image, determine the length informations of the shoes, width information and
It is one or more in elevation information.
5. a kind of device of identification shoe information, which is characterized in that including：
Image collection module, for by a plurality of image collecting device, obtaining the shoes being placed on rotating platform
Original image；
Detection module, for the original image carry out edge detection process, determine include shoe information feature image,
Wherein, the shoe information includes：One kind among brand message, type information, length information, width information and elevation information
Or it is a variety of；
Identification module, the shoe information in the feature image, determines brand message, the type of the shoes for identification
It is one or more among information, length information, width information and elevation information.
6. device according to claim 5, which is characterized in that described image acquisition module includes：
Driving unit, for rotating the rotating platform；
First image acquisition unit obtains the one or more the first former of the shoes by the image collecting device of first position
Beginning picture simultaneously stores；
Second image acquisition unit obtains the one or more the second former of the shoes by the image acquiring device of the second position
Beginning picture simultaneously stores.
7. device according to claim 6, which is characterized in that further include three-dimensional modeling module, for according to described one
Or multiple first pictures and one or more second picture build the threedimensional model of the shoes；Store the three of the shoes
Dimension module.
8. device according to claim 5, which is characterized in that the identification module includes the one of following two recognition units
Kind is a variety of：
First identification module determines the shoes for the feature image to be input to the default artificial nerve network model
It is one or more among the brand message and type information of son；
Second identification module, for obtaining the distance between object edge in the feature image, determine the length of the shoes
It is one or more in information, width information and elevation information.
9. a kind of device of identification shoe information, which is characterized in that including：
Processor；
Memory for storing executable instruction；
Wherein, the processor is configured as：
By a plurality of image collecting device, the original image for being placed into the shoes on rotating platform is obtained；
To the original image carry out edge detection process, determine include shoe information feature image, wherein the shoes
Information includes：It is one or more among brand message, type information, length information, width information and elevation information；
It identifies the shoe information in the feature image, determines brand message, type information, the length letter of the shoes
It is one or more among breath, width information and elevation information.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that when the computer journey
The method that identification shoe information according to any one of claims 1 to 4 is realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810383564.6A CN108765372A (en) | 2018-04-26 | 2018-04-26 | Identify the method, apparatus and computer readable storage medium of shoe information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810383564.6A CN108765372A (en) | 2018-04-26 | 2018-04-26 | Identify the method, apparatus and computer readable storage medium of shoe information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108765372A true CN108765372A (en) | 2018-11-06 |
Family
ID=64012038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810383564.6A Pending CN108765372A (en) | 2018-04-26 | 2018-04-26 | Identify the method, apparatus and computer readable storage medium of shoe information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108765372A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109315945A (en) * | 2018-12-14 | 2019-02-12 | 佛山市瑞创智能科技有限公司 | A kind of intelligent recognition shoe chest |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102743144A (en) * | 2012-07-13 | 2012-10-24 | 重庆理工大学 | Intelligent dry-cleaning type shoe cleaning machine and dry cleaning method |
| CN104123528A (en) * | 2013-04-24 | 2014-10-29 | 阿里巴巴集团控股有限公司 | Image recognition method and apparatus |
| CN104463851A (en) * | 2014-11-19 | 2015-03-25 | 哈尔滨工业大学深圳研究生院 | Automatic shoe sole edge line tracking method based on robot |
| CN107374562A (en) * | 2017-08-31 | 2017-11-24 | 成都南红鞋业有限公司 | A kind of footwear automatic flushing device |
| CN107944917A (en) * | 2017-11-27 | 2018-04-20 | 深圳码隆科技有限公司 | The data capture method and equipment and computer-readable storage medium of a kind of footwear |
| CN107943955A (en) * | 2017-11-24 | 2018-04-20 | 谭云 | A kind of clothing information collecting device, information management system and method |
-
2018
- 2018-04-26 CN CN201810383564.6A patent/CN108765372A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102743144A (en) * | 2012-07-13 | 2012-10-24 | 重庆理工大学 | Intelligent dry-cleaning type shoe cleaning machine and dry cleaning method |
| CN104123528A (en) * | 2013-04-24 | 2014-10-29 | 阿里巴巴集团控股有限公司 | Image recognition method and apparatus |
| CN104463851A (en) * | 2014-11-19 | 2015-03-25 | 哈尔滨工业大学深圳研究生院 | Automatic shoe sole edge line tracking method based on robot |
| CN107374562A (en) * | 2017-08-31 | 2017-11-24 | 成都南红鞋业有限公司 | A kind of footwear automatic flushing device |
| CN107943955A (en) * | 2017-11-24 | 2018-04-20 | 谭云 | A kind of clothing information collecting device, information management system and method |
| CN107944917A (en) * | 2017-11-27 | 2018-04-20 | 深圳码隆科技有限公司 | The data capture method and equipment and computer-readable storage medium of a kind of footwear |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109315945A (en) * | 2018-12-14 | 2019-02-12 | 佛山市瑞创智能科技有限公司 | A kind of intelligent recognition shoe chest |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109008887A (en) | A kind of method for washing shoe, device, shoe washing machine and computer readable storage medium | |
| US10463257B2 (en) | System and method for identifying physical properties of feet | |
| US10163256B2 (en) | Method and system for generating a three-dimensional model | |
| CN104463159B (en) | A kind of image processing method and device for positioning iris | |
| CN110381369A (en) | Determination method, apparatus, equipment and the storage medium of recommendation information implantation position | |
| CN108765278A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
| CN108377374B (en) | Method and system for generating depth information related to an image | |
| CN106575450A (en) | Augmented reality content rendering via albedo models, systems and methods | |
| Ofir et al. | On detection of faint edges in noisy images | |
| CN110936370A (en) | Cleaning robot control method and device | |
| CN110189390B (en) | Monocular vision SLAM method and system | |
| CN113011403B (en) | Gesture recognition method, system, medium and device | |
| CN110458855A (en) | Image extraction method and Related product | |
| CN108615045A (en) | Screen the method, apparatus and equipment of the image of capsule endoscope shooting | |
| Fang et al. | Separable kernel for image deblurring | |
| CN108765372A (en) | Identify the method, apparatus and computer readable storage medium of shoe information | |
| CN108256520A (en) | A kind of method, terminal device and computer readable storage medium for identifying the coin time | |
| CN103733207B (en) | Image partition method | |
| Liu et al. | Deblurring saturated night image with function-form kernel | |
| CN110009664A (en) | A kind of infrared object tracking method and device based on response diagram fusion | |
| CN109376782A (en) | Support vector machines cataract stage division and device based on eye image feature | |
| CN107292822A (en) | The joining method and device of a kind of image | |
| CN109146834A (en) | Car damage identification method and device, computer readable storage medium, terminal | |
| CN106462966B (en) | Method for identifying color block area in image and image processing device | |
| Valgma et al. | Iterative closest point based 3d object reconstruction using rgb-d acquisition devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |
|
| RJ01 | Rejection of invention patent application after publication |