CN113609928B - Smart city management system based on cloud computing and image recognition - Google Patents
Smart city management system based on cloud computing and image recognition Download PDFInfo
- Publication number
- CN113609928B CN113609928B CN202110813899.9A CN202110813899A CN113609928B CN 113609928 B CN113609928 B CN 113609928B CN 202110813899 A CN202110813899 A CN 202110813899A CN 113609928 B CN113609928 B CN 113609928B
- Authority
- CN
- China
- Prior art keywords
- image
- face
- matching
- repairing
- restoration
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000004088 simulation Methods 0.000 claims abstract description 13
- 238000004891 communication Methods 0.000 claims abstract description 8
- 230000001815 facial effect Effects 0.000 claims description 66
- 239000000284 extract Substances 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 47
- 238000007726 management method Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 3
- 238000012806 monitoring device Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a smart city management system based on cloud computing and image recognition, which comprises: the security terminal and the security image processing platform are in communication connection. The security image processing platform comprises a face analysis module, an image matching module, a simulation restoration module and a database, and all the modules are in communication connection. The face analysis module acquires the characteristics of a face to be repaired according to the original face three-dimensional model and the face repairing part; the image matching module obtains a reconstructed image spanning tree according to the characteristics of the surface to be repaired, generates a first repairing constraint function according to the maximum loss value and the minimum loss value of the reconstructed image spanning tree so as to obtain a first repairing matching image, and then obtains the repairing matching degree of the first repairing matching image according to a second repairing constraint function so as to obtain a second repairing matching image; and the simulation repairing module is used for fitting the second repairing matching image and the three-dimensional model of the face to be repaired to obtain a standard three-dimensional model of the face.
Description
Technical Field
The invention relates to the field of cloud computing and smart cities, in particular to a smart city management system based on cloud computing and image recognition.
Background
Cloud computing is an increasing, usage and delivery model for internet-based related services, typically involving the provision of dynamically scalable and often virtualized resources over the internet. Cloud is a metaphor of network and internet. In the past, telecommunications networks were often represented by clouds and later also by the abstraction of the internet and the underlying infrastructure. Narrow-sense cloud computing refers to a delivery and use mode of an IT infrastructure, and refers to acquiring required resources in an on-demand and easily-extensible manner through a network; the generalized cloud computing refers to a delivery and use mode of a service, and refers to obtaining a required service in an on-demand and easily-extensible manner through a network.
The smart city utilizes various information technologies or innovative concepts to get through and integrate the system and service of the city, so as to improve the efficiency of resource application, optimize city management and service, and improve the quality of life of citizens. The smart city is a city informatization advanced form which fully applies a new generation of information technology to various industries in the city and is based on the innovation of the next generation of knowledge society, realizes the deep integration of informatization, industrialization and urbanization, improves the urbanization quality, realizes the fine and dynamic management, improves the city management effect and improves the quality of life of citizens. In the prior art, the aim of automatically identifying a target object cannot be achieved by combining an image identification technology with cloud computing.
Disclosure of Invention
In view of the above, the present invention provides a smart city management system based on cloud computing and image recognition, which includes: the security terminal is in communication connection with the security image processing platform; the security image processing platform comprises a face analysis module, an image matching module, a simulation repairing module and a database, wherein the modules are in communication connection;
the security terminal sends a security image processing request to the security image processing platform; the security image processing request comprises an original face three-dimensional model, a face repairing image set and a face repairing part;
the face analysis module generates a three-dimensional face model to be repaired according to the original three-dimensional face model and the face repairing part, and extracts the features of the face to be repaired of the three-dimensional face model to be repaired;
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired, performs multi-order reconstruction according to the position relation among all repairing points to obtain a multi-order reconstructed image, and analyzes the structure of the multi-order reconstructed image to construct a reconstructed image generating tree of the multi-order reconstructed image;
the image matching module generates a first repairing constraint function according to the maximum loss value and the minimum loss value of the reconstructed image generation tree and takes the facial repairing images which are in the facial repairing image set and accord with the first repairing constraint function as first repairing matching images;
the image matching module generates a second restoration constraint function according to the first restoration matching images and the characteristics of the surface to be restored, acquires the restoration matching degree of each first restoration matching image according to the second restoration constraint function, and then takes the first restoration matching image with the maximum restoration matching degree as a second restoration matching image;
the simulation restoration module determines the fitting position and the fitting direction of the restoration matching image and the three-dimensional model of the face to be restored, fits the three-dimensional model of the face to be restored and the restoration matching image according to the fitting position and the fitting direction to obtain a standard three-dimensional model of the face, and then sends the standard three-dimensional model of the face to be restored and the restoration matching image to the corresponding security terminal.
According to a preferred embodiment, the face analysis module extracting the features of the face to be repaired of the three-dimensional model of the face to be repaired comprises:
the face analysis module acquires the characteristic value of each characteristic point of the three-dimensional model of the face to be repaired, takes the characteristic points with the characteristic values larger than the characteristic threshold value as the repair points, and then carries out weighted average on the characteristic values of all the repair points to obtain an average characteristic value;
the face analysis module compares the characteristic value of each repair point with the average characteristic value, takes the repair points with the characteristic value more than or equal to the average characteristic value as a first class of repair points, and takes the repair points with the characteristic value less than the average characteristic value as a second class of repair points;
the face analysis module generates a first type of repair point set according to all the first type of repair points and generates a second type of repair point set according to all the second type of repair points.
According to a preferred embodiment, the face analysis module extracting the features of the face to be repaired of the three-dimensional model of the face to be repaired comprises:
the face analysis module acquires the distance between each first-class repair point and other first-class repair points in the first-class repair point set, and connects each first-class repair point with other first-class repair points closest to the first-class repair point;
the face analysis module acquires the distance between each second-class repair point and other second-class repair points in the second-class repair point set, and connects each second-class repair point with other second-class repair points closest to the second-class repair point;
and the face analysis module obtains the features of the face to be repaired of the three-dimensional model of the face to be repaired according to the connection relation among all the repairing points of the three-dimensional model of the face to be repaired.
According to a preferred embodiment, the first repair constraint function is:
|MinP-MinQ i | 2 ≤θ 1 and|MaxP-MaxQ i | 2 ≤θ 2
wherein, theta 1 Is a first constraint threshold, θ 2 For the second constraint threshold, minP is the minimum loss value of the surface to be repaired, maxP is the maximum loss value of the surface to be repaired, minQ i For the minimum loss value, maxQ, of the ith face-repairing image in the set of face-repairing images i The maximum loss value of the ith facial restoration image in the facial restoration image set.
According to a preferred embodiment, the generating, by the image matching module, the second repair constraint function according to the first repair matching image and the feature of the surface to be repaired includes:
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired and acquires a repairing point set according to all repairing points of the face to be repaired;
the image matching module extracts the connection surface characteristics of each first restoration matching image, acquires all characteristic points of the connection surface of each first restoration matching image according to the connection surface characteristics of each first restoration matching image, and then acquires a connection surface characteristic point set of each first restoration matching image according to all characteristic points of the connection surface of each first restoration matching image;
and the image matching module generates a second restoration constraint function according to the restoration point set and the connection surface characteristic point set of each first restoration matching image.
According to a preferred embodiment, the obtaining, by the image matching module, a second repair matched image according to the second repair constraint function includes:
the image matching module calculates the similarity between the characteristics of the surface to be repaired and the characteristics of the connecting surface of each first repairing matching image according to a second repairing constraint function so as to obtain the repairing matching degree of each first repairing matching image;
the image matching module selects the first restoration matching image with the maximum restoration matching degree as a second restoration matching image.
According to a preferred embodiment, the second repair constraint function is:
wherein d is the repair match degree, r is a distance function, T 1 In order to repair the point set,
and matching the connecting surface feature point set of the image for the jth first repairing.
According to a preferred embodiment, the distance function comprises a first distance function and a second distance function;
the first distance function is:
T 1 in order to repair the point set,
and matching the set of connection surface feature points of the image for the jth first restoration.
The second distance function is:
T 1 in order to repair the point set,
and matching the connecting surface feature point set of the image for the jth first repairing.
According to a preferred embodiment, the three-dimensional model of the face to be repaired is the three-dimensional model of the face after the face repairing part in the original three-dimensional model of the face is removed; the facial repair image set comprises facial images which are collected by all monitoring devices and are related to the original facial three-dimensional model. The security protection terminal is the terminal equipment that security protection personnel used, and it includes: smart phones, tablet computers, desktop computers, and smart watches.
The invention has the following beneficial effects: according to the smart city management system based on cloud computing and image recognition, the purpose of clearly displaying the facial features of the target object is achieved by combining the cloud computing with the image recognition to repair the original facial three-dimensional image of the target object, the accuracy of object recognition is improved, and meanwhile waste of manpower and material resources is reduced.
Drawings
Fig. 1 is a block diagram illustrating a smart city management system based on cloud computing and image recognition according to an exemplary embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be described below clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the embodiments described are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments herein without making any creative effort, shall fall within the scope of protection.
Referring to fig. 1, in one embodiment, a smart city management system based on cloud computing and image recognition includes: the security terminal is in communication connection with the security image processing platform;
the security image processing platform comprises a face analysis module, an image matching module, a simulation restoration module and a database, and all the modules are in communication connection;
the security terminal sends a security image processing request to the security image processing platform; the security image processing request comprises an original face three-dimensional model, a face repairing image set and a face repairing part;
the face analysis module generates a three-dimensional face model to be repaired according to the original three-dimensional face model and the face repairing part, and extracts the features of the face to be repaired of the three-dimensional face model to be repaired;
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired, performs multi-order reconstruction according to the position relation among all repairing points to obtain a multi-order reconstructed image, and analyzes the structure of the multi-order reconstructed image to construct a reconstructed image generating tree of the multi-order reconstructed image;
the image matching module generates a first repairing constraint function according to the maximum loss value and the minimum loss value of the reconstructed image generation tree and takes the facial repairing images which are in the facial repairing image set and accord with the first repairing constraint function as first repairing matching images;
the image matching module generates a second restoration constraint function according to the first restoration matching images and the characteristics of the surface to be restored, acquires the restoration matching degree of each first restoration matching image according to the second restoration constraint function, and then takes the first restoration matching image with the maximum restoration matching degree as a second restoration matching image;
the simulation restoration module determines the fitting position and the fitting direction of the restoration matching image and the three-dimensional model of the face to be restored, fits the three-dimensional model of the face to be restored and the restoration matching image according to the fitting position and the fitting direction to obtain a standard three-dimensional model of the face, and then sends the standard three-dimensional model of the face to be restored and the restoration matching image to a corresponding security terminal.
For the purposes of promoting an understanding, the principles and operation of the present invention are described in detail below.
Specifically, in one embodiment, a smart city management method based on cloud computing and image recognition may include:
s1, the security terminal sends a security image processing request to a security image processing platform.
The security protection terminal is the terminal equipment that security protection personnel used, and it includes: smart phones, tablet computers, desktop computers, and smart watches. The security image processing request comprises an original face three-dimensional model, a face repairing image set and a face repairing part and is used for instructing the security image processing platform to repair the original face three-dimensional model.
The original face three-dimensional model is obtained by fusing the face images of the person. The personnel face images are obtained by multi-dimensional shooting of the monitoring equipment from different angles. The facial repair part is a facial part which needs to be repaired by security personnel according to an original facial three-dimensional model of the personnel, and the facial part can be divided according to human facial features, for example, the facial parts divided according to the human facial features are as follows: eyes, nose, lips, brow, and ears. The facial part can also be a part obtained by subdivision according to the division result of facial features.
And S2, the face analysis module generates a three-dimensional model of the face to be repaired according to the original three-dimensional model of the face and the face repairing part and extracts the features of the face to be repaired of the three-dimensional model of the face to be repaired.
In one embodiment, the three-dimensional model of the face to be repaired is generated by finding the face repairing part of the user from the original three-dimensional model according to the face repairing part in the security image processing request by the face analysis module, that is, finding the corresponding face part from the original three-dimensional model of the face by the face analysis module according to the part information indicated by the face repairing part, marking the face repairing part in the original three-dimensional model of the face by the face analysis module according to the face part, and removing the face repairing part in the original three-dimensional model of the face by the face analysis module according to the mark to obtain the three-dimensional model of the face to be repaired. For example, a facial repair site in the three-dimensional model of the original face may be marked with a grid of different colors, and the color of the facial repair site is changed to be transparent to achieve the effect of removal.
In one embodiment, the face analysis module extracting the features of the face to be repaired of the three-dimensional model of the face to be repaired comprises:
the face analysis module acquires the characteristic value of each characteristic point of the three-dimensional model of the face to be repaired, takes the characteristic points with the characteristic values larger than the characteristic threshold value as the repair points, and then carries out weighted average on the characteristic values of all the repair points to obtain an average characteristic value;
the face analysis module compares the characteristic value of each repair point with the average characteristic value, takes the repair points with the characteristic value more than or equal to the average characteristic value as a first class of repair points, and takes the repair points with the characteristic value less than the average characteristic value as a second class of repair points;
the face analysis module generates a first type of repair point set according to all the first type of repair points and generates a second type of repair point set according to all the second type of repair points.
In one embodiment, the face analysis module extracting the features of the face to be repaired of the three-dimensional model of the face to be repaired comprises:
the face analysis module acquires the distance between each first-class repair point and other first-class repair points in the first-class repair point set, and connects each first-class repair point with other first-class repair points closest to the first-class repair point;
the face analysis module acquires the distance between each second type repair point and other second type repair points in the second type repair point set, and connects each second type repair point with other second type repair points closest to the second type repair point;
and the face analysis module obtains the features of the face to be repaired of the three-dimensional model of the face to be repaired according to the connection relation among all the repairing points of the three-dimensional model of the face to be repaired. The characteristics of the surface to be repaired can be used for the image matching module to carry out characteristic reconstruction so as to screen out the facial repair image which is most matched with the surface to be repaired.
And S3, the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired, performs multi-order reconstruction according to the position relation among all repairing points to obtain a multi-order reconstructed picture, and analyzes the structure of the multi-order reconstructed picture to construct a reconstructed picture generating tree of the multi-order reconstructed picture.
And S4, the image matching module generates a first repairing constraint function according to the maximum loss value and the minimum loss value of the reconstructed image generation tree and takes the facial repairing images which are in the facial repairing image set and accord with the first repairing constraint function as first repairing matching images.
In another embodiment, the image matching module selecting a facial restoration image from the set of facial restoration images that conforms to the first restoration constraint function may include:
the image matching module selects one face restoration image from the face restoration image set as a target face restoration image;
the image matching module judges whether the target face repairing image accords with a first repairing constraint function or not;
when the target facial restoration image conforms to the first restoration constraint function, the image matching module adds the target facial restoration image to the first restoration matching image set and deletes the target facial restoration image from the facial restoration image set;
when the target facial restoration image does not conform to the first restoration constraint function, the image matching module deletes the target facial restoration image from the facial restoration image set;
the image matching module repeats the above steps until all the face restoration images in the face restoration image set are analyzed to obtain a first restoration matching image set.
Specifically, the first repairing constraint function is used for performing primary screening on the facial repairing in the facial repairing image set to obtain a facial repairing image meeting the first repairing constraint function, and obtaining a first repairing matching image set according to all the facial repairing images meeting the first repairing constraint function. The first repairing matching image set is a set of all facial repairing images meeting a first repairing constraint function, so that the security image processing platform selects a second repairing matching image from the first repairing matching image set according to a second repairing constraint function and fits the second repairing matching image to a to-be-repaired facial three-dimensional model to obtain a standard facial three-dimensional model.
In one embodiment, the first repair constraint function is:
|MinP-MinQ i | 2 ≤θ 1 and|MaxP-MaxQ i | 2 ≤θ 2
wherein, theta 1 Is a first constraint threshold, θ 2 For the second constraint threshold, minP is the minimum loss value of the surface to be repaired, maxP is the maximum loss value of the surface to be repaired, minQ i For the minimum loss value, maxQ, of the ith face-repairing image in the set of face-repairing images i The maximum loss value of the ith facial restoration image in the facial restoration image set.
The first constraint threshold value can be a threshold value preset by the system according to actual conditions, and the second constraint threshold value can be a threshold value preset by the system according to actual conditions.
In any method related to object feature extraction, when the features of a target object are converted into corresponding parameters, that is, the features of the target object are converted into data which can be used for numerical calculation, the features of the target object are point-collected, so that different feature points of the target object are selected from the point set according to corresponding selection rules, and the data amount and the calculation amount of feature conversion are reduced, therefore, the features of the target object are lost in the feature extraction process.
The maximum loss value is the maximum loss value of the characteristics of the surface to be repaired in the process that the image matching module constructs the multi-order reconstruction image according to the characteristics of the surface to be repaired and reconstructs the image to generate the tree. And the minimum loss value is the minimum loss value of the characteristics of the surface to be repaired in the process of constructing a multi-order reconstruction image and reconstructing the image spanning tree according to the characteristics of the surface to be repaired by the image matching module.
And S5, the image matching module generates a second restoration constraint function according to the first restoration matching images and the characteristics of the surface to be restored, acquires the restoration matching degree of each first restoration matching image according to the second restoration constraint function, and then takes the first restoration matching image with the maximum restoration matching degree as a second restoration matching image.
In one embodiment, the generating, by the image matching module, the second repair constraint function according to the first repair matching image and the feature of the surface to be repaired includes:
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired and acquires a repairing point set according to all repairing points of the face to be repaired;
the image matching module extracts the connection surface characteristics of each first restoration matching image, acquires all characteristic points of the connection surface of each first restoration matching image according to the connection surface characteristics of each first restoration matching image, and then acquires a connection surface characteristic point set of each first restoration matching image according to all characteristic points of the connection surface of each first restoration matching image;
and the image matching module generates a second restoration constraint function according to the restoration point set and the connection surface characteristic point set of each first restoration matching image.
In one embodiment, the obtaining, by the image matching module, a second repair matched image according to the second repair constraint function includes:
the image matching module calculates the similarity between the characteristics of the surface to be repaired and the characteristics of the connecting surface of each first repairing matching image according to a second repairing constraint function so as to obtain the repairing matching degree of each first repairing matching image;
the image matching module selects the first restoration matching image with the maximum restoration matching degree as a second restoration matching image.
And the second restoration constraint function is used for accurately matching the first restoration matching images in the first restoration matching image set so as to find out a final second restoration matching image of the three-dimensional model of the face to be restored.
In one embodiment, the second repair constraint function is:
wherein d is the repair match degree, r is a distance function, T 1 In order to repair the point set,
and matching the connecting surface feature point set of the image for the jth first repairing.
In one embodiment, the distance function includes a first distance function and a second distance function.
The first distance function is:
T 1 in order to repair the point set, the point set is repaired,
and matching the set of connection surface feature points of the image for the jth first restoration.
The second distance function is:
T 1 in order to repair the point set,
and matching the connecting surface feature point set of the image for the jth first repairing.
And the image matching module selects a second restoration matching image from the first restoration matching image set according to a second restoration constraint function and sends the second restoration matching image to the simulation restoration module so that the simulation restoration module fits the to-be-restored facial three-dimensional model and the second restoration matching image to obtain a standard facial three-dimensional model.
And S6, fitting the second repairing matching image and the to-be-repaired face three-dimensional model by the simulating repairing module to obtain a standard face three-dimensional model and sending the standard face three-dimensional model to a corresponding security terminal.
In one embodiment, the fitting the repair matching image and the three-dimensional model of the face to be repaired by the simulation repair module to obtain a standard three-dimensional model of the face comprises:
and the simulation repairing module determines the fitting position and the fitting direction of the repairing matching image and the three-dimensional model of the face to be repaired, and then fits the three-dimensional model of the face to be repaired and the repairing matching image according to the fitting position and the fitting direction to obtain a standard three-dimensional model of the face.
The three-dimensional model of the face to be repaired is the three-dimensional model of the face after the face repairing part in the original three-dimensional model of the face is removed; the facial repair image set comprises facial images which are collected by all monitoring devices and are related to the original facial three-dimensional model.
In one embodiment, the original facial three-dimensional model is a facial three-dimensional image obtained by performing image modeling according to a facial image of a target person acquired by a monitoring device, and the facial repair part is a facial part which is considered by security personnel that the original facial three-dimensional model cannot clearly display facial part features of the target person, namely, a facial part needing to be repaired. The facial repair image set is a facial image of the target person acquired by other monitoring equipment.
And the simulation restoration module fits the second restoration matching image and the three-dimensional model of the face to be restored to obtain a standard three-dimensional model of the face, and sends the standard three-dimensional model of the face to be restored to a corresponding security terminal, and the security terminal performs personnel identification on the received standard three-dimensional model of the face through a UI (user interface).
According to the smart city management system based on cloud computing and image recognition, the cloud computing and the image recognition are combined to repair the original facial three-dimensional image of the target object, so that the purpose of clearly displaying the facial features of the target object is achieved, the object recognition accuracy is improved, and meanwhile waste of manpower and material resources is reduced.
In addition, functional units in the embodiments herein may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions in the present invention substantially or partially contribute to the prior art, or all or part of the technical solutions may be embodied in the form of a software product stored in a storage medium, which includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention, and the foregoing storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Claims (5)
1. A smart city management system based on cloud computing and image recognition is characterized by comprising: the security terminal is in communication connection with the security image processing platform; the security image processing platform comprises a face analysis module, an image matching module, a simulation restoration module and a database, and all the modules are in communication connection;
the security terminal sends a security image processing request to the security image processing platform; the security image processing request comprises an original face three-dimensional model, a face repairing image set and a face repairing part;
the face analysis module generates a three-dimensional face model to be repaired according to the original three-dimensional face model and the face repairing part, and extracts the features of the face to be repaired of the three-dimensional face model to be repaired; the three-dimensional model of the face to be repaired is the three-dimensional model of the face after the face repairing part in the original three-dimensional model of the face is removed; the facial analysis module extracts the characteristics of the face to be repaired of the three-dimensional model of the face to be repaired, and the facial analysis module comprises the following steps:
the face analysis module acquires the characteristic value of each characteristic point of the three-dimensional model of the face to be repaired, takes the characteristic points with the characteristic values larger than the characteristic threshold value as the repair points, and then carries out weighted average on the characteristic values of all the repair points to obtain an average characteristic value; the face analysis module compares the characteristic value of each repair point with the average characteristic value, takes the repair points with the characteristic value more than or equal to the average characteristic value as a first class of repair points, and takes the repair points with the characteristic value less than the average characteristic value as a second class of repair points; the face analysis module generates a first type of repair point set according to all the first type of repair points and generates a second type of repair point set according to all the second type of repair points; the face analysis module acquires the distance between each first-class repair point and other first-class repair points in the first-class repair point set, and connects each first-class repair point with other first-class repair points closest to the first-class repair point; the face analysis module acquires the distance between each second-class repair point and other second-class repair points in the second-class repair point set, and connects each second-class repair point with other second-class repair points closest to the second-class repair point; the face analysis module obtains the features of the face to be repaired of the three-dimensional model of the face to be repaired according to the connection relation among all the repairing points of the three-dimensional model of the face to be repaired;
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired, performs multi-order reconstruction according to the position relation among all repairing points to obtain a multi-order reconstructed image, and analyzes the structure of the multi-order reconstructed image to construct a reconstructed image generating tree of the multi-order reconstructed image;
the image matching module generates a first repairing constraint function according to the maximum loss value and the minimum loss value of the reconstructed image generation tree and takes the facial repairing images which are in the facial repairing image set and accord with the first repairing constraint function as first repairing matching images; the maximum loss value is the maximum loss value of the characteristics of the surface to be repaired in the process that the image matching module constructs a multi-order reconstruction image according to the characteristics of the surface to be repaired and the reconstruction image generates a tree; the minimum loss value is the minimum loss value of the characteristics of the surface to be repaired in the process that the image matching module constructs a multi-order reconstruction graph and a reconstruction graph spanning tree according to the characteristics of the surface to be repaired; the first repair constraint function is:
|MinP-MinQ i | 2 ≤θ 1 and|MaxP-MaxQ i | 2 ≤θ 2
wherein, theta 1 Is a first constraint threshold, θ 2 For the second constraint threshold, minP is the minimum loss value of the surface to be repaired, maxP is the maximum loss value of the surface to be repaired, minQ i For the minimum loss value, maxQ, of the ith face-repairing image in the set of face-repairing images i The maximum loss value of the ith facial restoration image in the facial restoration image set;
the image matching module generates a second restoration constraint function according to the first restoration matching images and the characteristics of the surface to be restored, acquires the restoration matching degree of each first restoration matching image according to the second restoration constraint function, and then takes the first restoration matching image with the maximum restoration matching degree as a second restoration matching image; the second repair constraint function is:
wherein d is the repair match degree, r is a distance function, T 1 In order to repair the point set,
a connecting surface feature point set of the jth first repairing matching image;
and the simulation restoration module determines the fitting position and the fitting direction of the second restoration matching image and the three-dimensional model of the face to be restored, fits the three-dimensional model of the face to be restored and the second restoration matching image according to the fitting position and the fitting direction to obtain a standard three-dimensional model of the face, and then sends the standard three-dimensional model of the face to be restored and the second restoration matching image to the corresponding security terminal.
2. The system of claim 1, wherein the security terminal is a terminal device used by security personnel, and the system comprises: smart phones, tablet computers, desktop computers, and smart watches.
3. The system of claim 2, wherein the image matching module generating the second repair constraint function according to the first repair matching image and the surface feature to be repaired comprises:
the image matching module acquires all repairing points of the face to be repaired according to the features of the face to be repaired of the three-dimensional model of the face to be repaired and acquires a repairing point set according to all repairing points of the face to be repaired;
the image matching module extracts the connection surface characteristics of each first restoration matching image, acquires all characteristic points of the connection surface of each first restoration matching image according to the connection surface characteristics of each first restoration matching image, and then acquires a connection surface characteristic point set of each first restoration matching image according to all characteristic points of the connection surface of each first restoration matching image;
and the image matching module generates a second restoration constraint function according to the restoration point set and the connection surface characteristic point set of each first restoration matching image.
4. The system of claim 3, wherein the image matching module deriving the second repair match image according to the second repair constraint function comprises:
the image matching module calculates the similarity between the characteristics of the surface to be repaired and the characteristics of the connecting surface of each first repairing matching image according to a second repairing constraint function so as to obtain the repairing matching degree of each first repairing matching image;
the image matching module selects the first restoration matching image with the maximum restoration matching degree as a second restoration matching image.
5. The system of claim 4, wherein the distance function comprises a first distance function and a second distance function;
the first distance function is:
the second distance function is:
T 1 in order to repair the point set,
and matching the connecting surface feature point set of the image for the jth first repairing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110813899.9A CN113609928B (en) | 2021-07-19 | 2021-07-19 | Smart city management system based on cloud computing and image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110813899.9A CN113609928B (en) | 2021-07-19 | 2021-07-19 | Smart city management system based on cloud computing and image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113609928A CN113609928A (en) | 2021-11-05 |
| CN113609928B true CN113609928B (en) | 2022-12-20 |
Family
ID=78337881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110813899.9A Active CN113609928B (en) | 2021-07-19 | 2021-07-19 | Smart city management system based on cloud computing and image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609928B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107656978A (en) * | 2017-09-06 | 2018-02-02 | 国网上海市电力公司 | A kind of diversity data restorative procedure based on functional dependence |
| CN108171659A (en) * | 2017-12-01 | 2018-06-15 | 天津大学 | A kind of image repair method based on K-SVD dictionaries |
| CN112700388A (en) * | 2021-01-11 | 2021-04-23 | 清华大学 | Image restoration method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201215944D0 (en) * | 2012-09-06 | 2012-10-24 | Univ Manchester | Image processing apparatus and method for fittng a deformable shape model to an image using random forests |
| CN109697688B (en) * | 2017-10-20 | 2023-08-04 | 虹软科技股份有限公司 | Method and device for image processing |
| CN111724310B (en) * | 2019-03-21 | 2023-08-15 | 马上消费金融股份有限公司 | Training method of image restoration model, image restoration method and device |
| CN111415419B (en) * | 2020-03-19 | 2021-10-29 | 西安知北信息技术有限公司 | Method and system for making tooth restoration model based on multi-source image |
-
2021
- 2021-07-19 CN CN202110813899.9A patent/CN113609928B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107656978A (en) * | 2017-09-06 | 2018-02-02 | 国网上海市电力公司 | A kind of diversity data restorative procedure based on functional dependence |
| CN108171659A (en) * | 2017-12-01 | 2018-06-15 | 天津大学 | A kind of image repair method based on K-SVD dictionaries |
| CN112700388A (en) * | 2021-01-11 | 2021-04-23 | 清华大学 | Image restoration method and device |
Non-Patent Citations (3)
| Title |
|---|
| The minimum spanning tree in stereo vision;Efron Manik;《IOP Conference Series: Materials Science and Engineering》;20180930;全文 * |
| 基于Kinect的三维数据修复和融合算法;韩紫婷等;《电子世界》;20180923(第18期);全文 * |
| 基于最小生成树的规则图像碎片复原算法;赵林等;《计算机技术与发展》;20160505(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113609928A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220222925A1 (en) | Artificial intelligence-based image processing method and apparatus, device, and storage medium | |
| CN111814620A (en) | Face image quality evaluation model establishing method, optimization method, medium and device | |
| CN110414502B (en) | Image processing method and device, electronic equipment and computer readable medium | |
| US7257250B2 (en) | System, method, and program product for extracting a multiresolution quadrilateral-based subdivision surface representation from an arbitrary two-manifold polygon mesh | |
| CN112651342B (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN112231416B (en) | Knowledge graph body updating method and device, computer equipment and storage medium | |
| Nouri et al. | 3d blind mesh quality assessment index | |
| US20230047748A1 (en) | Method of fusing image, and method of training image fusion model | |
| CN112668638A (en) | Image aesthetic quality evaluation and semantic recognition combined classification method and system | |
| CN115147261A (en) | Image processing method, device, storage medium, equipment and product | |
| CN113313092B (en) | Handwritten signature recognition method, and claims settlement automation processing method, device and equipment | |
| CN114120413A (en) | Model training method, image synthesis method, device, equipment and program product | |
| CN114529750A (en) | Image classification method, device, equipment and storage medium | |
| CN114049290A (en) | Image processing method, device, equipment and storage medium | |
| CN113609928B (en) | Smart city management system based on cloud computing and image recognition | |
| CN113177466A (en) | Identity recognition method and device based on face image, electronic equipment and medium | |
| CN113673501B (en) | OCR classification method, system, electronic device and storage medium | |
| CN113378723B (en) | Automatic safety recognition system for hidden danger of power transmission and transformation line based on depth residual error network | |
| CN116431827A (en) | Information processing method, information processing device, storage medium and computer equipment | |
| CN112580538B (en) | Customer service personnel scheduling method, device, equipment and storage medium | |
| CN112069877B (en) | Face information identification method based on edge information and attention mechanism | |
| CN115545088A (en) | Model construction method, classification method and device and electronic equipment | |
| CN113592732B (en) | Image processing method based on big data and intelligent security | |
| CN113903071A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN112365189A (en) | Case distribution method and device |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221201 Address after: X1301-G543, No. 106, Fengze East Road, Nansha District, Guangzhou, Guangdong 510000 (for office use only) (JM) Applicant after: GUANGZHOU YATIAN NETWORK TECHNOLOGY Co.,Ltd. Address before: 650106 Yunnan Software Park, No. 220, 2nd Ring West Road, high tech Zone, Kunming, Yunnan Applicant before: Yang Wei |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |