CN113254903A - Image data processing method - Google Patents
Image data processing method Download PDFInfo
- Publication number
- CN113254903A CN113254903A CN202110730611.1A CN202110730611A CN113254903A CN 113254903 A CN113254903 A CN 113254903A CN 202110730611 A CN202110730611 A CN 202110730611A CN 113254903 A CN113254903 A CN 113254903A
- Authority
- CN
- China
- Prior art keywords
- image data
- neural network
- deep neural
- instruction information
- data set
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses an image data processing method, which comprises the following steps: acquiring video image data, and decoding each frame of image; preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network; and the deep neural network compares the data set with a preset image data set to determine a system interface for the equipment to enter. According to the image data processing method and device, the deep neural network is adopted to process the image data, the obtained image is divided into the face information and the instruction information, and then the independent matching operation is carried out, so that the device can enter different preset interfaces according to different unlocking actions while unlocking is carried out, and the user experience and the use efficiency of a user using intelligent devices such as a smart phone are greatly improved.
Description
Technical Field
The invention relates to an image processing method, in particular to an image data processing method.
Background
With the rapid development of multimedia technology, intelligent terminals such as smart phones and tablet computers have been widely used in various fields of work and life.
The intelligent terminals such as the existing intelligent mobile phone and the tablet personal computer mostly adopt the face recognition technology to unlock the mobile phone and pay, and the daily use of people is greatly facilitated, but the existing face recognition technology has a great progress space in application, for example, when the intelligent mobile phone is used in unlocking, most people all have a determined target, for example, after unlocking, WeChat is directly opened for use, the existing technology needs people to firstly unlock the intelligent mobile phone through face recognition, then manually click on the WeChat icon, and then use the WeChat, and when the intelligent terminal is used, the intelligent terminal is very inconvenient. If the user can directly enter the interface wanted by the user while unlocking, the user experience of the user using the intelligent devices such as the intelligent mobile phone can be greatly improved.
Disclosure of Invention
The present invention provides an image data processing method, which can directly enter an interface that a user wants to enter while performing face recognition unlocking on a device, so as to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
an image data processing method comprising the steps of:
acquiring video image data, and decoding each frame of image;
preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network;
and the deep neural network compares the data set with a preset image data set to determine a system interface for the equipment to enter.
As a further aspect of the present invention, the preset image dataset acquisition method includes the steps of:
acquiring video image data, and decoding each frame of image;
preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network;
the deep neural network learns and stores the data set to form a preset image data set, and specifies the corresponding relation between a preset image data set subset and each interface of the equipment system.
As a further aspect of the present invention, the step of learning the data set by the deep neural network is as follows:
(a) collecting a large amount of video image data, and marking original instruction information and original face information in the picture;
(b) preprocessing the pictures uniformly and generating a structured binary data file;
(c) feeding binary data files into a deep neural network model in batches, and obtaining prediction instruction information and prediction face information after reasoning and calculation;
(d) respectively comparing the predicted instruction information and the predicted face information with the original instruction information and the original face information correspondingly, and calculating loss values between the predicted instruction information and the predicted face information;
(e) updating weight parameters in the deep neural network through operation, and minimizing the loss value in (d);
(f) and (e) repeating the operations (c) to (e) until the number of iteration rounds is met, and finishing the training.
As a further scheme of the invention, the prediction instruction information is obtained by adopting a spectrum reconstruction algorithm to carry out reasoning calculation.
As a further scheme of the present invention, the steps of the spectrum reconstruction algorithm for performing inference calculation are as follows:
reconstructing the original structured binary data file by adopting a spectrum reconstruction algorithm to generate image segmentation data;
the image segmentation data includes prediction instruction information and predicted face information.
As a further aspect of the present invention, the method for comparing the data set with the preset image data set by the deep neural network is as follows:
comparing the predicted face information with the original face information to determine whether to unlock the equipment;
and if the predicted face information is matched with the original face information, matching and searching the predicted instruction information and the original instruction information, and if the matched original instruction information is inquired, executing an interface of the equipment system corresponding to the corresponding original instruction information.
Further, on the basis of the method, the invention also provides an image processing device, which comprises an image acquisition module, a decoding module, a preprocessing module, a storage module and a deep neural network module;
the image acquisition module is used for acquiring a video image;
the decoding module is used for decoding the acquired video image;
the preprocessing module is used for preprocessing each decoded frame image to obtain data sets with different dimensionalities and feeding the data sets into the deep neural network module;
the storage module is used for storing a preset image data set;
and the deep neural network module compares the data set with a preset image data set to determine a system interface for the equipment to enter.
Compared with the prior art, the invention has the beneficial effects that: according to the image data processing method, the deep neural network is adopted to process the image data, the obtained image is divided into the face information and the instruction information, and then the independent matching operation is carried out, so that the device can enter different preset interfaces according to different unlocking actions while unlocking is carried out, and the user experience and the use efficiency of the user using intelligent devices such as a smart phone are greatly improved.
Drawings
Fig. 1 is a flowchart of an image data processing method.
Fig. 2 is a flowchart of an image dataset acquisition method in an image data processing method.
Fig. 3 is a flowchart of learning a data set by a deep neural network in an image data processing method.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments 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 given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, in an embodiment of the present invention, an image data processing method includes the following steps:
video image data is acquired and each frame of image is decoded.
Specifically, the image data is acquired by using equipment such as a camera and the like, and then transmitted to a decoding module, so that each frame of image is decoded.
And preprocessing each decoded frame image by adopting a preprocessing module to obtain data sets with different dimensionalities and feeding the data sets into a deep neural network.
And the deep neural network compares the data set with a preset image data set to determine a system interface for the equipment to enter.
Specifically, referring to fig. 2, the preset image data set obtaining method includes the following steps:
acquiring video image data, and decoding each frame of image;
preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network;
the deep neural network learns and stores the data set to form a preset image data set, and specifies the corresponding relation between a preset image data set subset and each interface of the equipment system.
As a further aspect of the present invention, referring to fig. 3, the step of learning the data set by the deep neural network is as follows:
(a) collecting a large amount of video image data, and marking original instruction information and original face information in the picture;
(b) preprocessing the pictures uniformly and generating a structured binary data file;
(c) feeding binary data files into a deep neural network model in batches, and performing inference calculation by adopting a spectrum reconstruction algorithm to obtain predicted instruction information and predicted face information;
(d) respectively comparing the predicted instruction information and the predicted face information with the original instruction information and the original face information correspondingly, and calculating loss values between the predicted instruction information and the predicted face information;
(e) updating weight parameters in the deep neural network through operation, and minimizing the loss value in (d);
(f) and (e) repeating the operations (c) to (e) until the number of iteration rounds is met, and finishing the training.
Specifically, the steps of performing inference calculation by the spectrum reconstruction algorithm are as follows:
reconstructing the original structured binary data file by adopting a spectrum reconstruction algorithm to generate image segmentation data;
the image segmentation data includes prediction instruction information and predicted face information.
As a further aspect of the present invention, the method for comparing the data set with the preset image data set by the deep neural network is as follows:
comparing the predicted face information with the original face information to determine whether to unlock the equipment;
and if the predicted face information is matched with the original face information, matching and searching the predicted instruction information and the original instruction information, and if the matched original instruction information is inquired, executing an interface of the equipment system corresponding to the corresponding original instruction information.
On the basis of the method, the invention also provides image processing equipment which comprises an image acquisition module, a decoding module, a preprocessing module, a storage module and a deep neural network module;
the image acquisition module is used for acquiring a video image; specifically, the image acquisition module may be a camera module of a smart phone;
the decoding module is used for decoding the acquired video image;
the preprocessing module is used for preprocessing each decoded frame image to obtain data sets with different dimensionalities and feeding the data sets into the deep neural network module;
the storage module is used for storing a preset image data set;
and the deep neural network module compares the data set with a preset image data set to determine a system interface for the equipment to enter.
Specifically, when the method is applied, a plurality of photos with face image information and different gestures at the same time can be shot through a camera of the image acquisition module, such as an OK gesture, a biye gesture (scissor hand) and the like, the gestures are set as keys entering different user interfaces on equipment, for example, the OK gesture is set as the key for opening the WeChat, when the user unlocks facing the camera, the OK gesture is shown, and the equipment adopting the method can directly open the WeChat software after the equipment is unlocked.
In summary, according to the image data processing method, the deep neural network is adopted to process the image data, the obtained image is divided into the face information and the instruction information, and then the individual matching operation is performed, so that the device can enter different preset interfaces according to different unlocking actions while unlocking is performed, and the user experience and the use efficiency of the user using smart devices such as a smart phone are greatly improved.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (5)
1. An image data processing method, characterized by comprising the steps of:
acquiring video image data, and decoding each frame of image;
preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network;
the deep neural network compares the data set with a preset image data set to determine a system interface for the equipment to enter; the preset image data set acquisition method comprises the following steps:
acquiring video image data, and decoding each frame of image;
preprocessing each decoded frame image to obtain data sets with different dimensions and feeding the data sets into a deep neural network;
the deep neural network learns and stores the data set to form a preset image data set, and specifies the corresponding relation between a preset image data set subset and each interface of the equipment system.
2. The image data processing method of claim 1, wherein the deep neural network learns the data set as follows:
(a) collecting a large amount of video image data, and marking original instruction information and original face information in the picture;
(b) preprocessing the pictures uniformly and generating a structured binary data file;
(c) feeding binary data files into a deep neural network model in batches, and obtaining prediction instruction information and prediction face information after reasoning and calculation;
(d) respectively comparing the predicted instruction information and the predicted face information with the original instruction information and the original face information correspondingly, and calculating loss values between the predicted instruction information and the predicted face information;
(e) updating weight parameters in the deep neural network through operation, and minimizing the loss value in (d);
(f) and (e) repeating the operations (c) to (e) until the number of iteration rounds is met, and finishing the training.
3. The image data processing method according to claim 2, wherein the prediction instruction information is obtained by performing inference calculation using a spectral reconstruction algorithm.
4. The image data processing method according to claim 3, wherein the spectral reconstruction algorithm performs the following steps of performing the inference calculation:
reconstructing the original structured binary data file by adopting a spectrum reconstruction algorithm to generate image segmentation data;
the image segmentation data includes prediction instruction information and predicted face information.
5. The method of claim 4, wherein the deep neural network compares the data set with a preset image data set by:
comparing the predicted face information with the original face information to determine whether to unlock the equipment;
and if the predicted face information is matched with the original face information, matching and searching the predicted instruction information and the original instruction information, and if the matched original instruction information is inquired, executing an interface of the equipment system corresponding to the corresponding original instruction information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110730611.1A CN113254903A (en) | 2021-06-30 | 2021-06-30 | Image data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110730611.1A CN113254903A (en) | 2021-06-30 | 2021-06-30 | Image data processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113254903A true CN113254903A (en) | 2021-08-13 |
Family
ID=77190143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110730611.1A Pending CN113254903A (en) | 2021-06-30 | 2021-06-30 | Image data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113254903A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832700A (en) * | 2017-11-03 | 2018-03-23 | 全悉科技（北京）有限公司 | A kind of face identification method and system |
| CN110058777A (en) * | 2019-03-13 | 2019-07-26 | 华为技术有限公司 | The method and electronic equipment of shortcut function starting |
| CN111581620A (en) * | 2020-04-30 | 2020-08-25 | 新浪网技术(中国)有限公司 | User identification method and device |
| CN111950007A (en) * | 2020-08-13 | 2020-11-17 | 北京元心科技有限公司 | Text information hiding method and device, electronic equipment and storage medium |
-
2021
- 2021-06-30 CN CN202110730611.1A patent/CN113254903A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832700A (en) * | 2017-11-03 | 2018-03-23 | 全悉科技（北京）有限公司 | A kind of face identification method and system |
| CN110058777A (en) * | 2019-03-13 | 2019-07-26 | 华为技术有限公司 | The method and electronic equipment of shortcut function starting |
| CN111581620A (en) * | 2020-04-30 | 2020-08-25 | 新浪网技术(中国)有限公司 | User identification method and device |
| CN111950007A (en) * | 2020-08-13 | 2020-11-17 | 北京元心科技有限公司 | Text information hiding method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10713532B2 (en) | Image recognition method and apparatus | |
| CN106777318B (en) | Matrix decomposition cross-modal Hash retrieval method based on collaborative training | |
| Liu et al. | Generalized higher order orthogonal iteration for tensor learning and decomposition | |
| CN109934249A (en) | Data processing method, device, medium and calculating equipment | |
| CN113111842B (en) | Action recognition method, device, equipment and computer readable storage medium | |
| CN111126347B (en) | Human eye state identification method, device, terminal and readable storage medium | |
| CN112466314A (en) | Emotion voice data conversion method and device, computer equipment and storage medium | |
| CN108898549A (en) | Image processing method, picture processing unit and terminal device | |
| CN112395979A (en) | Image-based health state identification method, device, equipment and storage medium | |
| CN110610191A (en) | Elevator floor identification method and device and terminal equipment | |
| CN112528029A (en) | Text classification model processing method and device, computer equipment and storage medium | |
| CN109635706B (en) | Gesture recognition method, device, storage medium and device based on neural network | |
| CN114359582A (en) | Small sample feature extraction method based on neural network and related equipment | |
| CN111027582A (en) | Semi-supervised feature subspace learning method and device based on low-rank graph learning | |
| CN110909578A (en) | Low-resolution image recognition method and device and storage medium | |
| CN109376230B (en) | Crime prediction method, crime prediction system, storage medium, and server | |
| CN113254903A (en) | Image data processing method | |
| CN111709312B (en) | Local feature face recognition method based on combined main mode | |
| Kabakus | A novel sketch recognition model based on convolutional neural networks | |
| KR20190136441A (en) | Deep learning based real time 3d gesture recognition system and method using temporal and spatial normalization | |
| CN111428652B (en) | Biological characteristic management method, system, equipment and medium | |
| Wang et al. | Temporal sparse feature auto‐combination deep network for video action recognition | |
| CN112258392A (en) | Super-resolution image training method, device, medium and equipment | |
| Ren | The advance of generative model and variational autoencoder | |
| Tang et al. | Incomplete multi-view learning: Review, analysis, and prospects |
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: 20210813 |