CN108564019A - A kind of face identification method and system - Google Patents

A kind of face identification method and system Download PDF

Info

Publication number
CN108564019A
CN108564019A CN201810305276.9A CN201810305276A CN108564019A CN 108564019 A CN108564019 A CN 108564019A CN 201810305276 A CN201810305276 A CN 201810305276A CN 108564019 A CN108564019 A CN 108564019A
Authority
CN
China
Prior art keywords
image
face
grader
recognized
images
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
Application number
CN201810305276.9A
Other languages
Chinese (zh)
Inventor
支洪伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shades Of Vision Technology (dalian) Co Ltd
Original Assignee
Shades Of Vision Technology (dalian) Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shades Of Vision Technology (dalian) Co Ltd filed Critical Shades Of Vision Technology (dalian) Co Ltd
Priority to CN201810305276.9A priority Critical patent/CN108564019A/en
Publication of CN108564019A publication Critical patent/CN108564019A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172Classification, e.g. identification

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

The present invention discloses a kind of face identification method and system, the method includes:Obtain pending image;The pending image is pre-processed, pretreatment image is obtained;Histogram equalization is carried out to the pretreatment image and is normalized, rectangular image is obtained;To the rectangular image and difference carry out thresholding processing, obtain images to be recognized;Multiple grader nodes are created using statistics boosting;Recognition of face is carried out to the images to be recognized using each grader node.The face identification method of the present invention has identification accurate by image preprocessing, and identification is quick, has higher reduction degree to face, and the identification scanning system of face can be completed without body contact.

Description

A kind of face identification method and system
Technical field
The present invention relates to field of face identification, more particularly to a kind of face identification method and system.
Background technology
Face recognition technology is the face feature based on people, and the facial image or video flowing to input first determine whether it With the presence or absence of face, if there is face, then the position of each face, size and each major facial organ are further provided Location information.And according to these information, the identity characteristic contained in each face is further extracted, and by itself and known people Face is compared, to identify the identity of each face.
A series of practical the relevant technologies for including structure face identification system of recognition of face of broad sense, including facial image are adopted Collection, Face detection, recognition of face pretreatment, identity validation and identity finder etc.;And the recognition of face of narrow sense refers in particular to pass through people Face carries out the technology or system of identity validation or identity finder.
Modern biological identification technology has had reached certain height, and people pursue the effect of the identification in more complicated environment Fruit, and current face identification system is unable to fully meet this demand of people.
Invention content
The object of the present invention is to provide a kind of face identification method and systems, for improving the precision of recognition of face.
To achieve the above object, the present invention provides following schemes:
A kind of face identification method, the method includes:
Obtain pending image;
The pending image is pre-processed, pretreatment image is obtained;
Histogram equalization is carried out to the pretreatment image and is normalized, rectangular image is obtained;
To the rectangular image and difference carry out thresholding processing, obtain images to be recognized;
Multiple grader nodes are created using statistics boosting;
Recognition of face is carried out to the images to be recognized using each grader node.
Optionally, described that recognition of face is carried out to the images to be recognized using each grader node, it specifically includes:
Each grader node of stating is divided into from small to large by resolution ratio:First grader node, the second grader section Point ...;
Recognition of face is carried out to images to be recognized using the first grader node, obtains judging result;
If the judging result indicates that the images to be recognized includes face information, using the second grader node pair The images to be recognized carries out recognition of face, and so on, to the last a grader node judges the images to be recognized Including face information, determine that the pending image is facial image.
Optionally, described pre-process includes:In digitlization, geometric transformation, normalization, smooth, recovery and enhancing at least It is a kind of.
Optionally, described normalize includes:At least one of gray scale normalization, geometrical normalization and transform normalization.
Optionally, described to include smoothly:Median method, part are averaging at least one of method and k line neighbour's methods of average.
Optionally, described enhance includes:At least one of contrast broadening, logarithmic transformation, density stratification and histogram.
Optionally, the geometric transformation is systematic error for correcting image capturing system and instrumented site with chance error The transformation that difference is carried out.
A kind of face identification system, the system comprises:
Pending image collection module, for obtaining pending image;
Preprocessing module obtains pretreatment image for being pre-processed to the pending image;
Module is equalized and normalized, for carrying out histogram equalization to the pretreatment image and normalizing, is obtained Rectangular image;
Thresholding processing module, for the rectangular image and difference carry out thresholding processing, obtain to be identified Image;
Grader node creation module, for creating multiple grader nodes using statistics boosting;
Face recognition module, for carrying out recognition of face to the images to be recognized using each grader node.
According to specific embodiment provided by the invention, the invention discloses following technique effects:
The face identification method of the present invention has identification accurate, identification is quick, has to face higher by image preprocessing Reduction degree, can complete the identification scanning system of face without body contact.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings Obtain other attached drawings.
Fig. 1 is the flow chart of the present inventor's face recognition method;
Fig. 2 is the structure connection figure of face identification system of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of face identification method and systems, for improving the precision of recognition of face.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real Applying mode, the present invention is described in further detail.
The present invention provides a kind of face identification methods, and Fig. 1 is the flow chart of the present inventor's face recognition method, such as Fig. 1 institutes Show, the method includes:
Step 11:Obtain pending image;
Step 12:The pending image is pre-processed, pretreatment image is obtained.
The pretreatment includes:At least one of digitlization, geometric transformation, normalization, smooth, recovery and enhancing.
The normalization includes:At least one of gray scale normalization, geometrical normalization and transform normalization.
It is described to include smoothly:Median method, part are averaging at least one of method and k line neighbour's methods of average.
The enhancing includes:At least one of contrast broadening, logarithmic transformation, density stratification and histogram.
The geometric transformation for systematic error and instrumented site for correcting image capturing system random error institute into Capable transformation.
Step 13:Histogram equalization is carried out to the pretreatment image and is normalized, rectangular image is obtained;
Step 14:To the rectangular image and difference carry out thresholding processing, obtain images to be recognized;
Step 15:Multiple grader nodes are created using statistics boosting;
Step 16:Recognition of face is carried out to the images to be recognized using each grader node, is specifically included:
Each grader node of stating is divided into from small to large by resolution ratio:First grader node, the second grader section Point ...;
Recognition of face is carried out to images to be recognized using the first grader node, obtains judging result;
If the judging result indicates that the images to be recognized includes face information, using the second grader node pair The images to be recognized carries out recognition of face, and so on, to the last a grader node judges the images to be recognized Including face information, determine that the pending image is facial image.
In image recognition, the quality of picture quality directly affects the design and effect precision of recognizer, then in addition to Outside optimization that can be algorithmically, preconditioning technique occupies critically important factor in entire project, however people often ignore this A bit.
Each character image information sorting, is out given identification module identification by image preprocessing, this process is known as Image preprocessing.
The main purpose of image preprocessing is to eliminate information unrelated in image, restores useful real information, enhancing has The detectability and simplified data to the maximum extent for closing information, to improve feature extraction, image segmentation, matching and identification can By property.Preprocessing process generally have digitlization, geometric transformation, normalization, it is smooth, restore and enhancing.
Filtering:Filtering (Wave fi ltering) is by the operation that specific band frequency filters out in signal, is to inhibit and prevent The important measures only interfered.(note:Mean value, intermediate value, gaussian filtering can influence the clarity of image to a certain extent, and one As be all just to be filtered when having apparent noise.)
The normalization of image and histogram have similitude, can be to a certain extent the pixel of 0-255 by image understanding Between value becomes 0-1, its distribution distance is reduced.(note:When target and background gray scale difference are larger with preferably, otherwise can go out Existing adhesion leads to not divide.)
Smooth (filtering) of image is a kind of inhibition to gradation of image saltus step, and the sharpening of image is then on the contrary, it is to figure A kind of enhancing of the Gray Level Jump part of picture, the variations in detail information of prominent image.(note:Smooth and sharpening belongs to filtering, all All it is first to establish a filter, but difference lies in the mode operator of selection is different with the same function).It sharpens and often also " increases Image sharpening peace can be slided and is used in combination by noise by force ", if image itself has apparent noise, can it is first smooth again It sharpens, if there is noise after image sharpening, can further go to be smoothed.
Digitlization:The gray value of one width original photo is the continuous function of space variable (successive value of position).In M × N Photo gray scale is sampled on dot matrix and is quantified and (is classified as one of 2b tonal gradation), can obtain what computer can be handled Digital picture.In order to enable digital picture to rebuild original image, just there is certain requirement to the size of M, N and b value.It is receiving Within the scope of the space and gray scale resolution capability of device, the numerical value of M, N and b are bigger, and the quality of reconstruction image is better.When sampling week When phase is equal to or less than the half in minimum details period in original image, the frequency spectrum of reconstruction image is equal to the frequency spectrum of original image, Therefore reconstruction image can be identical with original image.Since the product of M, N and b three determine piece image in a computer Amount of storage, therefore need under conditions of amount of storage is certain to select suitable M, N and b value according to the heterogeneity of image, with Obtain best treatment effect.
Geometric transformation:The change that the random error of systematic error and instrumented site for correcting image capturing system is carried out It changes.For the systematic error of satellite image, the distortion as caused by the factors such as earth rotation, scanning mirror speed and map projection can To be indicated with model, and eliminated by geometric transformation.Random error error as caused by attitude of flight vehicle and height change is difficult To be showed with model, so being usually after systematic error is repaired, by the figure being observed and known correct geometry The figure of position compares, and the mesh of transformation is reached with a certain number of ground control point solution bivariate polynomial group of functions in figure 's.
Normalization:Certain features of image are made to change a kind of graphics standard form with invariance in given change.Figure Certain properties of picture, for example, object area and perimeter, originally for coordinate rotation for just have constant property.General In the case of, certain factors or the influence converted to image properties can be eliminated or be weakened by normalized, to It can be selected as measuring the foundation of image.Such as the uncontrollable remote sensing image of illumination, the normalization of grey level histogram for Image analysis is very necessary.Gray scale normalization, geometrical normalization and transform normalization are obtain image invariance three Kind method for normalizing.
Smoothly:It is smoothly the technology for eliminating random noise in image, is the basic demand to smoothing technique, is eliminating noise While become blurred image outline or lines.Common smoothing method has median method, local averaging method and k lines Neighbour's method of average, regional area size can be fixed, and can also be point by point with gray value size variation.In addition, sometimes Application space frequency domain band-pass filtering method.
It restores:It corrects the image caused by a variety of causes to degenerate, the image that reconstruction or estimation obtain is made to approach as far as possible Image field of the ideal without degeneration, in practical applications it occur frequently that image degradation phenomena.Such as the disturbance of big air-flow, optical system The relative motion of aberration, camera and object can all make remote sensing images degenerate.Basic recovery technique is the degeneration acquisition Image g (x, y) regards the convolution of degenrate function h (x, y) and ideal image f (x, y) as, and there are relationships for their Fourier transformation G (u, v=H (u, v) F (u, v).After degenrate function being determined according to degradation mechanism, so that it may find out F (u, v) from this relational expression, then use Fourier inversion finds out f (x, y).Usually it is referred to as inverse filter.When practical application, since H (u, v) is flat with uv is left The distance of face origin increases and declines rapidly, in order to avoid the reinforcing of noise in high-frequency range, when u2+v2 is more than a certain boundary value When W, M (u, v) [what refers to] (distance of uv plane origins) is made to be equal to 1.The selection of W0 should make H (u, v) in u2+v2≤W ranges It is not in inside zero.The algebraic method of image restoration is based on least square method optimum criterion.Seek a valuation, makes excellent It is minimum to spend criterion function value.This method is fairly simple, can derive least square method Wiener filter.When there is no noises When, Wiener filter becomes ideal inverse filter.
Enhancing:Header length in image is reinforced and inhibited, to improve the visual effect of image, or image is turned Become the form more suitable for machine processing, in order to data pick-up or identification.Such as:One Image Intensified System can pass through High-pass filter protrudes the contour line of image, to enable the machine to measure the shape and perimeter of contour line.Image enhancement skill There are many method, contrast broadening, logarithmic transformation, density stratification and histogram equalization etc. can be used in changing image tone and dash forward art Go out details.Often experiment is repeated and can be only achieved satisfied effect with different methods when practical application.
The protection point of the present invention is that image will be pre-processed, and pretreated image is accurate, needs algorithm Accurate calculation, the detector of Face datection need the function of automatic decision identification face, the detection of face can be divided Class identifies, realizes quick recognition detection face.
Face datection
The Haar classifier of openCV is a supervised classifier, carries out histogram equalization to image first and normalizes To onesize, then whether label the inside includes the object to be monitored.It is initially by Paul Viola and Michael Jones Design, therefore referred to as Viola Jones detectors.Viola Jones graders use AdaBoost in cascade each node To learn the multilayer Tree Classifier of the low reject rate of a high detection rate.The present invention improves the prior art, specifically includes Several aspects under several:
1. using class Haar input feature vectors:To rectangular image area and difference carry out thresholding.(the words is to making With the definition of class Haar input feature vectors)
2. integral image techniques accelerate the calculating of the value of 45 ° of rotations of rectangular area, for accelerating class Haar inputs special The calculating of sign.
3. using statistics boosting come create two class problems (face and non-face) grader node (high pass rate, Low reject rate).
4. Weak Classifier node composition screening type cascade.I.e. the first classifiers are optimal, can pass through the figure comprising object As region, at the same allow some do not include object by image pass through;Second classifiers be suboptimum grader, and have compared with Low reject rate.And so on, to boosting graders, if there is face that can detect, while it is non-to refuse sub-fraction Face, and it is passed along next grader, it is for low reject rate.And so on, the last one grader will be almost all of Non-face all refusals fall, only remaining human face region.As long as image-region has passed through entire cascade, then it is assumed that there is object in the inside.
Although this technology is suitable for Face datection, it is not limited to Face datection, it may also be used for the detection of other objects, such as vapour The front of vehicle, aircraft etc., is detected side below.When detecting, trained Parameter File is first imported, wherein Haarcascade_frontalface_alt2.xml is preferable to the recognition effect of frontal faces, haarcascade_ Profileface.xml is preferable to the detection result of side face.
The present invention also provides a kind of face identification system, Fig. 2 is the structure connection figure of face identification system of the present invention.Such as Shown in Fig. 2, the system comprises:
Pending image collection module 21, for obtaining pending image;
Preprocessing module 22 obtains pretreatment image for being pre-processed to the pending image;
Module 23 is equalized and normalized, for carrying out histogram equalization to the pretreatment image and normalizing, is obtained To rectangular image;
Thresholding processing module 24, for the rectangular image and difference carry out thresholding processing, obtain waiting knowing Other image;
Grader node creation module 25, for creating multiple grader nodes using statistics boosting;
Face recognition module 26, for carrying out recognition of face to the images to be recognized using each grader node.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For system disclosed in embodiment For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is said referring to method part It is bright.
Principle and implementation of the present invention are described for specific case used herein, and above example is said The bright method and its core concept for being merely used to help understand the present invention;Meanwhile for those of ordinary skill in the art, foundation The thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not It is interpreted as limitation of the present invention.

Claims (8)

1. a kind of face identification method, which is characterized in that the method includes:
Obtain pending image;
The pending image is pre-processed, pretreatment image is obtained;
Histogram equalization is carried out to the pretreatment image and is normalized, rectangular image is obtained;
To the rectangular image and difference carry out thresholding processing, obtain images to be recognized;
Multiple grader nodes are created using statistics boosting;
Recognition of face is carried out to the images to be recognized using each grader node.
2. face identification method according to claim 1, which is characterized in that described to use each grader node to institute It states images to be recognized and carries out recognition of face, specifically include:
Each grader node of stating is divided into from small to large by resolution ratio:First grader node, the second grader section Point ...;
Recognition of face is carried out to images to be recognized using the first grader node, obtains judging result;
If the judging result indicates that the images to be recognized includes face information, using the second grader node to described Images to be recognized carries out recognition of face, and so on, to the last a grader node judges that the images to be recognized includes Face information determines that the pending image is facial image.
3. face identification method according to claim 1, which is characterized in that the pretreatment includes:Digitlization, geometry become Change, normalize, smoothly, restore and enhancing at least one of.
4. face identification method according to claim 3, which is characterized in that the normalization includes:It is gray scale normalization, several At least one of what normalization and transform normalization.
5. face identification method according to claim 3, which is characterized in that described to include smoothly:Median method, part ask flat Equal at least one of method and k line neighbour's methods of average.
6. face identification method according to claim 3, which is characterized in that the enhancing includes:Contrast broadening, logarithm become It changes, at least one of density stratification and histogram.
7. face identification method according to claim 3, which is characterized in that the geometric transformation is to be adopted for correcting image The transformation that the systematic error of collecting system and the random error of instrumented site are carried out.
8. a kind of face identification system, which is characterized in that the system comprises:
Pending image collection module, for obtaining pending image;
Preprocessing module obtains pretreatment image for being pre-processed to the pending image;
Module is equalized and normalized, for carrying out histogram equalization to the pretreatment image and normalizing, obtains rectangle Image;
Thresholding processing module, for the rectangular image and difference carry out thresholding processing, obtain images to be recognized;
Grader node creation module, for creating multiple grader nodes using statistics boosting;
Face recognition module, for carrying out recognition of face to the images to be recognized using each grader node.
CN201810305276.9A 2018-04-08 2018-04-08 A kind of face identification method and system Pending CN108564019A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810305276.9A CN108564019A (en) 2018-04-08 2018-04-08 A kind of face identification method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810305276.9A CN108564019A (en) 2018-04-08 2018-04-08 A kind of face identification method and system

Publications (1)

Publication Number Publication Date
CN108564019A true CN108564019A (en) 2018-09-21

Family

ID=63534147

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810305276.9A Pending CN108564019A (en) 2018-04-08 2018-04-08 A kind of face identification method and system

Country Status (1)

Country Link
CN (1) CN108564019A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800640A (en) * 2018-12-14 2019-05-24 天津大学 A kind of smoking detection method based on Two-dimensional Surfaces fitting
CN109829995A (en) * 2018-12-14 2019-05-31 浙江工业大学 A kind of recognition of face classroom is registered system
CN111508093A (en) * 2020-04-26 2020-08-07 王素云 Road toll budget system for expressway
CN112183550A (en) * 2020-09-09 2021-01-05 上海鹰觉科技有限公司 Ship classification and identification method based on deep learning
CN115619410A (en) * 2022-10-19 2023-01-17 闫雪 Self-adaptive financial payment platform

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101430759A (en) * 2008-12-04 2009-05-13 上海大学 Optimized recognition pretreatment method for human face
CN105488456A (en) * 2015-11-23 2016-04-13 中国科学院自动化研究所 Adaptive rejection threshold adjustment subspace learning based human face detection method
CN106815560A (en) * 2016-12-22 2017-06-09 广州大学 It is a kind of to be applied to the face identification method that self adaptation drives seat
CN107273783A (en) * 2016-08-23 2017-10-20 苏州金脑袋智能系统工程有限公司 Face identification system and its method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101430759A (en) * 2008-12-04 2009-05-13 上海大学 Optimized recognition pretreatment method for human face
CN101430759B (en) * 2008-12-04 2010-11-17 上海大学 Optimized recognition pretreatment method for human face
CN105488456A (en) * 2015-11-23 2016-04-13 中国科学院自动化研究所 Adaptive rejection threshold adjustment subspace learning based human face detection method
CN107273783A (en) * 2016-08-23 2017-10-20 苏州金脑袋智能系统工程有限公司 Face identification system and its method
CN106815560A (en) * 2016-12-22 2017-06-09 广州大学 It is a kind of to be applied to the face identification method that self adaptation drives seat

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
LIUPH_: "基于openCV实现人脸检测", 《HTTPS://BLOG.CSDN.NET/LIUPH_/ARTICLE/DETAILS/51941705》 *
VIOLA P等: "Rapid Object Detection using a Boosted Cascade of Simple Features", 《PROCEEDINGS OF THE 2001 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 *
中国大百科全书出版社编辑部: "《中国大百科全书 自动化控制与系统工程》", 30 April 1992 *
单园园等: "改进AdaBoost与SVM快速人脸检测", 《烟台大学学报(自然科学与工程版)》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800640A (en) * 2018-12-14 2019-05-24 天津大学 A kind of smoking detection method based on Two-dimensional Surfaces fitting
CN109829995A (en) * 2018-12-14 2019-05-31 浙江工业大学 A kind of recognition of face classroom is registered system
CN111508093A (en) * 2020-04-26 2020-08-07 王素云 Road toll budget system for expressway
CN112183550A (en) * 2020-09-09 2021-01-05 上海鹰觉科技有限公司 Ship classification and identification method based on deep learning
CN115619410A (en) * 2022-10-19 2023-01-17 闫雪 Self-adaptive financial payment platform
CN115619410B (en) * 2022-10-19 2024-01-26 闫雪 Self-adaptive financial payment platform

Similar Documents

Publication Publication Date Title
CN108564019A (en) A kind of face identification method and system
CN112819772B (en) High-precision rapid pattern detection and recognition method
Kumawat et al. A robust edge detection algorithm based on feature-based image registration (FBIR) using improved canny with fuzzy logic (ICWFL)
Nie et al. Two-dimensional extension of variance-based thresholding for image segmentation
CN111985427A (en) Living body detection method, living body detection apparatus, and readable storage medium
CN114387591A (en) License plate recognition method, system, equipment and storage medium
CN112686248A (en) Certificate increase and decrease type detection method and device, readable storage medium and terminal
CN109712134B (en) Iris image quality evaluation method and device and electronic equipment
Pujar et al. Medical image segmentation based on vigorous smoothing and edge detection ideology
CN112101283A (en) Intelligent identification method and system for traffic signs
CN116524269A (en) Visual recognition detection system
CN106525027A (en) Star sensor star point extracting method based on local binary pattern
Teutsch et al. Noise resistant gradient calculation and edge detection using local binary patterns
Barbu Automatic edge detection solution using anisotropic diffusion-based multi-scale image analysis and fine-to-coarse tracking
Bandarupally et al. Detection of Military Targets from Satellite Images using Deep Convolutional Neural Networks
CN114529518A (en) Image pyramid and NLM-based image enhancement method for cryoelectron microscope
Gautam et al. A GUI for automatic extraction of signature from image document
CN110188601B (en) Airport remote sensing image detection method based on learning
Adamski et al. Binary handwriting image enhancement by directional field-guided morphology
Steckenrider et al. Continuum detection and predictive-corrective classification of crack networks
Ho et al. Nom document background removal using generative adversarial network
Favorskaya et al. The off-line signature verification based on structural similarity
CN117542067B (en) Region labeling form recognition method based on visual recognition
Chehdi et al. A blind system to identify and filter degradations affecting an image
Bhatnagar Image Segmentation Using Canny Edge Detection Technique Kirti

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20180921