CN105224950A - The recognition methods of filter classification and device - Google Patents
The recognition methods of filter classification and device Download PDFInfo
- Publication number
- CN105224950A CN105224950A CN201510634596.5A CN201510634596A CN105224950A CN 105224950 A CN105224950 A CN 105224950A CN 201510634596 A CN201510634596 A CN 201510634596A CN 105224950 A CN105224950 A CN 105224950A
- Authority
- CN
- China
- Prior art keywords
- image
- filter
- identified
- category database
- classification
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Abstract
Present disclose provides the recognition methods of a kind of filter classification and device, belong to technical field of image processing.Method comprises: the characteristics of image extracting image to be identified, and described image to be identified is the image after filter process; According to the characteristics of image of filter category database and described image to be identified, obtain other matching degree of each filter class in described image to be identified and described filter category database; The filter classification described matching degree being met preset standard is defined as filter classification corresponding to described image to be identified.The disclosure, by extracting the characteristics of image of image to be identified, according to filter category database, realizes according to other function of characteristics of image Auto-matching filter class.In one embodiment, the disclosure, according to the method for crucial words and Image Feature Matching, can also realize carrying out according to filtering effects image the function named to corresponding filter classification.In one embodiment, the disclosure obtains filter category database by degree of depth learning algorithm.
Description
Technical field
The disclosure relates to technical field of image processing, particularly relates to the recognition methods of a kind of filter classification and device.
Background technology
At present, have image procossing to apply in various terminal, along with the development of image processing techniques, the application of these image procossing also has increasing function, such as filter function, and filter function may be used for realizing the multiple image with different-effect.
In order to obtain abundanter filtering effects, terminal is except providing some intrinsic filters, additionally provide filter params and regulate option, make user can obtain self-defined filter by regulating filter params, and then obtain corresponding effect image by this self-defined filter.
But, in existing image procossing application, mostly title due to different filtering effects is to number or the words irrelevant with filtering effects, as " vividly " " aestheticism " etc., which kind of and do not identify through the image of filter process and be through filter process and obtain, thus cause being difficult to from a large amount of filters, find corresponding filter to reappear effect.
Summary of the invention
For overcoming Problems existing in correlation technique, the disclosure provides the recognition methods of a kind of filter classification and device.
According to the first aspect of disclosure embodiment, the recognition methods of a kind of filter classification is provided, comprises:
Extract the characteristics of image of image to be identified, this image to be identified is the image after filter process;
According to the characteristics of image of filter category database with this image to be identified, obtain other matching degree of each filter class in this image to be identified and this filter category database, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image;
The filter classification this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
In the first possibility implementation of first aspect of the present disclosure, the method also comprises:
Collect learning sample, this learning sample is the image after the reason of multiple filter class other places;
Extract the characteristics of image of every piece image in this learning sample;
According to characteristics of image and the corresponding filter classification of this every piece image, set up this filter category database.
In the second possibility implementation of first aspect of the present disclosure, this filter category database also comprises the crucial words for describing each other image effect of filter class;
The method also comprises:
According to the characteristics of image of this filter category database with this image to be identified, obtain the crucial words corresponding with this image to be identified;
Crucial words corresponding for this image to be identified is defined as filter title corresponding to this image to be identified.
In the third possibility implementation of first aspect of the present disclosure, this is according to the characteristics of image of filter category database with this image to be identified, and before obtaining the crucial words corresponding with this image to be identified, the method also comprises:
The set of structure alternative names, this alternative names set is used for storage key word;
According to the crucial words in this alternative names set, collect the image of the predetermined number mated with each crucial words;
Extract the characteristics of image of the image that this mates with each crucial words;
According to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, set up this filter category database.
In the 4th kind of possibility implementation of first aspect of the present disclosure, the method also comprises:
When the characteristics of image of this image to be identified and other matching degree of arbitrary filter class all do not meet this preset standard, determine that this image to be identified is the image without filter process.
According to the second aspect of disclosure embodiment, a kind of filter classification recognition device is provided, comprises:
Characteristic extracting module, for extracting the characteristics of image of image to be identified, this image to be identified is the image after filter process;
Matching degree acquisition module, for the characteristics of image according to filter category database and this image to be identified, obtain other matching degree of each filter class in this image to be identified and this filter category database, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image;
Determination module, the filter classification for this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
In the first possibility implementation of second aspect of the present disclosure, this device also comprises:
Collect module, for collecting learning sample, this learning sample is the image after the reason of multiple filter class other places;
This characteristic extracting module is also for extracting the characteristics of image of every piece image in this learning sample;
First sets up module, for according to the characteristics of image of this every piece image and corresponding filter classification, sets up this filter category database.
In the second possibility implementation of second aspect of the present disclosure, this filter category database also comprises the crucial words for describing each other image effect of filter class;
This device also comprises:
Crucial words acquisition module, for the characteristics of image according to this filter category database and this image to be identified, obtains the crucial words corresponding with this image to be identified;
This determination module is also for being defined as filter title corresponding to this image to be identified by crucial words corresponding for this image to be identified.
In the third possibility implementation of second aspect of the present disclosure, this device also comprises:
Constructing module, for constructing alternative names set, this alternative names set is used for storage key word;
Collection module, for according to the crucial words in this alternative names set, collects the image of the predetermined number mated with each crucial words;
This characteristic extracting module is also for extracting the characteristics of image of the image that this mates with each crucial words;
Second sets up module, for according to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, sets up this filter category database.
May in implementation at the 4th kind of second aspect of the present disclosure, this determination module also for:
When the characteristics of image of this image to be identified and other matching degree of arbitrary filter class all do not meet this preset standard, determine that this image to be identified is the image without filter process.
The third aspect, additionally provides a kind of filter classification recognition device, comprising:
Processor;
For the storer of the executable instruction of storage of processor;
Wherein, this processor is configured to:
Extract the characteristics of image of image to be identified, this image to be identified is the image after filter process;
According to the characteristics of image of filter category database with this image to be identified, obtain other matching degree of each filter class in this image to be identified and this filter category database, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image;
The filter classification this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
The beneficial effect that the technical scheme that disclosure embodiment provides is brought is:
The disclosure, by extracting the characteristics of image of image to be identified, according to filter category database, realizes according to other function of characteristics of image Auto-matching filter class.In one embodiment, the disclosure, according to the method for crucial words and Image Feature Matching, can also realize carrying out according to filtering effects image the function named to corresponding filter classification.In one embodiment, the disclosure obtains filter category database by degree of depth learning algorithm.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the disclosure.
Accompanying drawing explanation
Accompanying drawing to be herein merged in instructions and to form the part of this instructions, shows and meets embodiment of the present disclosure, and is used from instructions one and explains principle of the present disclosure.
Fig. 1 is the process flow diagram of a kind of filter classification recognition methods according to an exemplary embodiment.
Fig. 2 is the process flow diagram of a kind of filter classification recognition methods according to an exemplary embodiment.
Fig. 3 is a kind of filter classification recognition device block diagram according to an exemplary embodiment.
Fig. 4 is the block diagram of a kind of filter classification recognition device 400 according to an exemplary embodiment.
Embodiment
For making object of the present disclosure, technical scheme and advantage clearly, below in conjunction with accompanying drawing, disclosure embodiment is described in further detail.
Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the disclosure.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that aspects more of the present disclosure are consistent.
Fig. 1 is the process flow diagram of a kind of filter classification recognition methods according to an exemplary embodiment, and as shown in Figure 1, the recognition methods of filter classification is used for, in terminal, comprising the following steps.
In a step 101, extract the characteristics of image of image to be identified, this image to be identified is the image after filter process.
In a step 102, according to the characteristics of image of filter category database with this image to be identified, obtain other matching degree of each filter class in this image to be identified and this filter category database, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image.
In step 103, the filter classification this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
The method that disclosure embodiment provides, by extracting the characteristics of image of image to be identified, according to the filter category database obtained, realizes according to other function of characteristics of image Auto-matching filter class.
In the first possibility implementation of the present disclosure, the method also comprises:
Collect learning sample, this learning sample is the image after the reason of multiple filter class other places;
Extract the characteristics of image of every piece image in this learning sample;
According to characteristics of image and the corresponding filter classification of this every piece image, set up this filter category database.
In the second possibility implementation of the present disclosure, this filter category database also comprises the crucial words for describing each other image effect of filter class;
The method also comprises:
According to the characteristics of image of this filter category database with this image to be identified, obtain the crucial words corresponding with this image to be identified;
Crucial words corresponding for this image to be identified is defined as filter title corresponding to this image to be identified.
In the third possibility implementation of the present disclosure, this is according to the characteristics of image of filter category database with this image to be identified, and before obtaining the crucial words corresponding with this image to be identified, the method also comprises:
The set of structure alternative names, this alternative names set is used for storage key word;
According to the crucial words in this alternative names set, collect the image of the predetermined number mated with each crucial words;
Extract the characteristics of image of the image that this mates with each crucial words;
According to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, set up this filter category database.
In the 4th kind of possibility implementation of the present disclosure, this is according to the characteristics of image of filter category database with this image, and obtain in this image to be identified and this filter category database after other matching degree of each filter class, the method also comprises:
When the characteristics of image of this image to be identified and other matching degree of arbitrary filter class all do not meet this preset standard, determine that this image to be identified is the image without filter process.
Above-mentioned all alternatives, can adopt and combine arbitrarily formation embodiment of the present disclosure, this is no longer going to repeat them.
Fig. 2 is the process flow diagram of a kind of filter classification recognition methods according to an exemplary embodiment.With reference to Fig. 2, this embodiment specifically comprises:
In step 201, collect learning sample, this learning sample is the image after the reason of multiple filter class other places.
The effect of learning sample is understood all by part.In the present embodiment, in order to obtain the characteristics of image of image after different classes of filter process, the some width images through often kind of filter process can be obtained, to form learning sample.Such as, the filter of every kind can process the corresponding effect images of 1000 width, by this 1000 width effect image, obtains the characteristics of image of image after often kind of filter process.The effect image quantity that every kind filter makes can be arranged voluntarily by person skilled, and it should be noted that, within the specific limits, this quantitative value is larger, and the characteristics of image of acquisition more can represent the treatment effect of corresponding filter.
In another embodiment, learning sample can also comprise the effect image through self-defined filter process.Illustratively, the acquisition methods of this self-defined filter can be: terminal provides filter params to regulate option, makes user by regulating filter options, can be obtained from definition filter, and obtains the effect image through this self-defined filter process further.By adding in this learning sample by this effect image of this self-defined filter process, dirigibility and the applicability of method described in the present embodiment can be improved.
It should be noted that, the acquisition of this learning sample can also be the process of a real-time update.Such as, when adding arbitrary self-defined filter, in learning sample, increase the corresponding effect image of this self-defined filter process.This learning sample can also be obtained by additive method, and disclosure embodiment is not construed as limiting this.
In step 202., the characteristics of image of every piece image in this learning sample is extracted.
Before the characteristics of image extracting every piece image in this learning sample, when this learning sample comprises the different image of size, also need to do pre-service to the image of this learning sample, such as, make each image size of this learning sample identical.Such as, the size of image in this learning sample can be reset to 256*256, also can reset to other suitable sizes, disclosure embodiment is not construed as limiting this.
In the disclosed embodiments, convolutional neural networks can be adopted to extract the characteristics of image of every piece image in this learning sample.Convolutional neural networks has the shared characteristic of local weight because of it, can reduce the training parameter of network, thus can reduce the complicacy of web results, accelerates training speed.
Disclosure embodiment does not repeat the specific algorithm aspect relating to convolutional neural networks in this step 202, except employing convolutional neural networks extracts except characteristics of image, also can adopt the extraction of other degree of depth learning algorithms realization to characteristics of image, disclosure embodiment is not construed as limiting this.
In step 203, according to characteristics of image and the corresponding filter classification of this every piece image, this filter category database is set up.In one embodiment, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image.
According to the characteristics of image after every kind filter process of extracting in step 202 and corresponding filter classification, set up filter category database, this filter category database is used for according to the characteristics of image of image after the process of arbitrary classification filter, determine to process this image use the filter classification of filter.
In the disclosure one embodiment, this filter category database can also comprise the crucial words for describing each other image effect of filter class.When this filter category database also comprises the crucial words for describing each other image effect of filter class, the method for building up of this filter category database can also comprise: the set of structure alternative names, and this alternative names set is used for storage key word; According to the crucial words in this alternative names set, collect the image of the predetermined number mated with each crucial words; Extract the characteristics of image of the image that this mates with each crucial words; According to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, set up this filter category database.
Illustratively, the crucial words in alternative names set is used for the characteristics of image of accurate description image after filter process, as " black and white " " shade " " mirror image " etc.According to the crucial words in this alternative set, the image that can mate from network search with crucial words, also can choose the image that can describe its characteristics of image with key word word from the image after filter process.By the image composition image collection ImGroup got, extract the characteristics of image of every piece image in this image collection ImGroup, and then set up filter category database.Obtaining in the image mated with crucial words, except the above-mentioned acquisition methods of employing, can also adopt other acquisition methods according to actual conditions, disclosure embodiment is not construed as limiting this.
In another embodiment, according to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, can also set up for naming other database of filter class separately.
It should be noted that, the step of above-mentioned building database also can think the process by training study Sample Establishing model.When for name other database of filter class with for identifying that other database of filter class is separate time, can think that above-mentioned steps is the process of establishing of two models: one is the process by training study Sample Establishing filter classification model of cognition, one is the process setting up filter classification naming model according to image collection ImGroup and alternative names set.
In step 204, extract the characteristics of image of image to be identified, this image to be identified is the image after filter process.
In the image characteristic extracting method of image to be identified and step 202, image characteristic extracting method in like manner, do not repeat herein.
In step 205, according to the characteristics of image of filter category database with this image to be identified, other matching degree of each filter class in this image to be identified and this filter category database is obtained.
In the disclosed embodiments, matching degree for two width images are described characteristics of image between similarity degree.According to the characteristics of image of the image to be identified extracted in step 204, the matching degree of this image to be identified characteristics of image corresponding with each filter classification in this filter category database can be calculated respectively.
In step 206, the filter classification this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
Determine that the corresponding filter class method for distinguishing of image to be identified can be according to matching degree: in other matching degree of each filter class the image to be identified got from step 205 and this filter category database, obtain maximum matching degree, and obtaining the filter classification corresponding with this maximum matching degree further, this filter classification can be defined as filter classification corresponding to this image to be identified.That is to say, this preset standard can be the maximum standard of matching degree.
In order to obtain filter classification corresponding to image to be identified more accurately, this preset standard can also be set to: after determining maximum matching degree, judge whether this maximum matching degree is greater than predetermined threshold value, when this maximum matching degree is greater than this predetermined threshold value, filter classification corresponding for this maximum matching degree is defined as filter classification corresponding to this image to be identified.
When the characteristics of image of this image to be identified and other matching degree of arbitrary filter class all do not meet this preset standard, determine that this image to be identified is the image without filter process.That is to say, according to other matching degree of each filter class in the characteristics of image of this image to be identified and this filter category database, obtain maximum matching degree wherein, when this maximum matching degree is not more than predetermined threshold value, determine that this image to be identified is the image without filter process.
This preset standard can carry out specific aim setting according to actual conditions, and disclosure embodiment is to how to determine that the filter classification of image to be identified is not construed as limiting according to matching degree.
In step 207, according to the characteristics of image of this filter category database with this image to be identified, the crucial words corresponding with this image to be identified is obtained; Crucial words corresponding for this image to be identified is defined as filter title corresponding to this image to be identified.
According to the characteristics of image of the image to be identified extracted in step 205, use SVM (support vector machine, SupportVectorMachine) sorter obtains crucial words corresponding with the characteristics of image of this image to be identified in this filter category database, this crucial words can be defined as filter title corresponding to this image to be identified, except using SVM classifier, other sorters with similar effect can also be used to obtain corresponding crucial words according to filter classification, and disclosure embodiment is not construed as limiting this.
Above-mentioned filter classification identifying can be applied in the application of arbitrary image procossing, such as, can at image procossing interface display identification filter class option, when the trigger action to this identification filter class option being detected, by the recognition methods of above-mentioned filter classification, the image that current interface shows is identified, and display filter classification recognition result.In one embodiment, name filter class option can also being shown, when the trigger action to this other option of name filter class being detected, showing other name result of filter class to this current interface of process showing image.Illustratively, when the filter classification that the filter recognized provides for present terminal, the filter classification that this recognizes can also be shown on interface, directly obtain the image with identical filtering effects to facilitate user.
In another embodiment, after user obtains self-defined filter by adjustment filter params, when wanting to name this self-defined filter, piece image is appointed by this self-defined filter process, and the image obtained after filter process, namely this image can be used as identifying this other image of self-defined filter class.Illustratively, when the trigger action to name filter class option being detected, can be named by above-mentioned filter classification naming method, and on screen, show name result.Illustratively, except show name result on screen except, also can showing the option whether upgrading filter title simultaneously, when detecting that user selects the option upgrading filter title, name result being updated to the correspondence position waiting to name filter icon.
The method that disclosure embodiment provides, by extracting the characteristics of image of image to be identified, according to the filter category database obtained by degree of depth learning algorithm, is realized according to other function of characteristics of image Auto-matching filter class.According to the method for crucial words and Image Feature Matching, can also realize carrying out according to filtering effects image the function named to corresponding filter classification.Further, extract characteristics of image by convolutional neural networks, can also complexity be reduced, improve feature extraction efficiency, use SVM classifier to determine that crucial words can also improve name accuracy rate.
Fig. 3 is a kind of filter classification recognition device block diagram according to an exemplary embodiment.With reference to Fig. 3, this device comprises characteristic extracting module 301, matching degree acquisition module 302, determination module 303.
Characteristic extracting module 301, for extracting the characteristics of image of image to be identified, this image to be identified is the image after filter process;
Matching degree acquisition module 302, for the characteristics of image according to filter category database and this image to be identified, obtain other matching degree of each filter class in this image to be identified and this filter category database, this filter category database is at least for preserving multiple filter classification and adopting the plurality of filter class other places to manage the characteristics of image of rear image;
Determination module 303, the filter classification for this matching degree being met preset standard is defined as filter classification corresponding to this image to be identified.
In the first possibility implementation that the disclosure provides, this device also comprises:
Collect module, for collecting learning sample, this learning sample is the image after different classes of filter process;
This characteristic extracting module 301 is also for extracting the characteristics of image of every piece image in this learning sample;
First sets up module, for according to the characteristics of image of this every piece image and corresponding filter classification, sets up this filter category database.
In the second possibility implementation that the disclosure provides, this filter category database also comprises the crucial words for describing each other image effect of filter class;
This device also comprises:
Crucial words acquisition module, for the characteristics of image according to this filter category database and this image to be identified, obtains the crucial words corresponding with this image to be identified;
This determination module 303 is also for being defined as filter title corresponding to this image to be identified by crucial words corresponding for this image to be identified.
In the third possibility implementation that the disclosure provides, this device also comprises:
Constructing module, for constructing alternative names set, this alternative names set is used for storage key word;
Collection module, for according to the crucial words in this alternative names set, collects the image of the predetermined number mated with each crucial words;
This characteristic extracting module 301 is also for extracting the characteristics of image of the image that this mates with each crucial words;
Second sets up module, for according to the crucial words in this alternative names set and the characteristics of image being somebody's turn to do the image that mate with each crucial words, sets up this filter category database.
The 4th kind that provides in the disclosure may in implementation, this determination module 303 also for:
When the characteristics of image of this image to be identified and other matching degree of arbitrary filter class all do not meet this preset standard, determine that this image to be identified is the image without filter process.
About the device in above-described embodiment, wherein the concrete mode of modules executable operations has been described in detail in about the embodiment of the method, will not elaborate explanation herein.
Fig. 4 is the block diagram of a kind of filter classification recognition device 400 according to an exemplary embodiment.Such as, device 400 can be mobile phone, computing machine, digital broadcast terminal, messaging devices, game console, tablet device, Medical Devices, body-building equipment, personal digital assistant etc.
With reference to Fig. 4, device 400 can comprise following one or more assembly: processing components 402, storer 404, power supply module 406, multimedia groupware 404, audio-frequency assembly 410, I/O (I/O) interface 412, sensor module 414, and communications component 416.
The integrated operation of the usual control device 400 of processing components 402, such as with display, call, data communication, camera operation and record operate the operation be associated.Processing components 402 can comprise one or more processor 420 to perform instruction, to complete all or part of step of above-mentioned method.In addition, processing components 402 can comprise one or more module, and what be convenient between processing components 402 and other assemblies is mutual.Such as, processing components 402 can comprise multi-media module, mutual with what facilitate between multimedia groupware 408 and processing components 402.
Storer 404 is configured to store various types of data to be supported in the operation of device 400.The example of these data comprises for any application program of operation on device 400 or the instruction of method, contact data, telephone book data, message, picture, video etc.Storer 404 can be realized by the volatibility of any type or non-volatile memory device or their combination, as static RAM (SRAM), Electrically Erasable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory EPROM (EPROM), programmable read only memory (PROM), ROM (read-only memory) (ROM), magnetic store, flash memory, disk or CD.
The various assemblies that power supply module 406 is device 400 provide electric power.Power supply module 406 can comprise power-supply management system, one or more power supply, and other and the assembly generating, manage and distribute electric power for device 400 and be associated.
Multimedia groupware 408 is included in the screen providing an output interface between described device 400 and user.In certain embodiments, screen can comprise liquid crystal display (LCD) and touch panel (TP).If screen comprises touch panel, screen may be implemented as touch-screen, to receive the input signal from user.Touch panel comprises one or more touch sensor with the gesture on sensing touch, slip and touch panel.Described touch sensor can the border of not only sensing touch or sliding action, but also detects the duration relevant to described touch or slide and pressure.In certain embodiments, multimedia groupware 408 comprises a front-facing camera and/or post-positioned pick-up head.When device 400 is in operator scheme, during as screening-mode or video mode, front-facing camera and/or post-positioned pick-up head can receive outside multi-medium data.Each front-facing camera and post-positioned pick-up head can be fixing optical lens systems or have focal length and optical zoom ability.
Audio-frequency assembly 410 is configured to export and/or input audio signal.Such as, audio-frequency assembly 410 comprises a microphone (MIC), and when device 400 is in operator scheme, during as call model, logging mode and speech recognition mode, microphone is configured to receive external audio signal.The sound signal received can be stored in storer 404 further or be sent via communications component 416.In certain embodiments, audio-frequency assembly 410 also comprises a loudspeaker, for output audio signal.
I/O interface 412 is for providing interface between processing components 402 and peripheral interface module, and above-mentioned peripheral interface module can be keyboard, some striking wheel, button etc.These buttons can include but not limited to: home button, volume button, start button and locking press button.
Sensor module 414 comprises one or more sensor, for providing the state estimation of various aspects for device 400.Such as, sensor module 414 can detect the opening/closing state of device 400, the relative positioning of assembly, such as described assembly is display and the keypad of device 400, the position of all right pick-up unit 400 of sensor module 414 or device 400 1 assemblies changes, the presence or absence that user contacts with device 400, the temperature variation of device 400 orientation or acceleration/deceleration and device 400.Sensor module 414 can comprise proximity transducer, be configured to without any physical contact time detect near the existence of object.Sensor module 414 can also comprise optical sensor, as CMOS or ccd image sensor, for using in imaging applications.In certain embodiments, this sensor module 414 can also comprise acceleration transducer, gyro sensor, Magnetic Sensor, pressure transducer or temperature sensor.
Communications component 416 is configured to the communication being convenient to wired or wireless mode between device 400 and other equipment.Device 400 can access the wireless network based on communication standard, as WiFi, 2G or 3G, or their combination.In one exemplary embodiment, communications component 416 receives from the broadcast singal of external broadcasting management system or broadcast related information via broadcast channel.In one exemplary embodiment, described communications component 416 also comprises near-field communication (NFC) module, to promote junction service.Such as, can based on radio-frequency (RF) identification (RFID) technology in NFC module, Infrared Data Association (IrDA) technology, ultra broadband (UWB) technology, bluetooth (BT) technology and other technologies realize.
In the exemplary embodiment, device 400 can be realized, for performing the recognition methods of above-mentioned filter classification by one or more application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD) (PLD), field programmable gate array (FPGA), controller, microcontroller, microprocessor or other electronic components.
In the exemplary embodiment, additionally provide a kind of non-transitory computer-readable recording medium comprising instruction, such as, comprise the storer 404 of instruction, above-mentioned instruction can perform said method by the processor 420 of device 400.Such as, described non-transitory computer-readable recording medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and optical data storage devices etc.
In the exemplary embodiment, additionally provide a kind of non-transitory computer-readable recording medium, when the instruction in described storage medium is performed by the processor of mobile terminal, make mobile terminal can perform the recognition methods of above-mentioned filter classification.
Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the present disclosure.The application is intended to contain any modification of the present disclosure, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present disclosure and comprised the undocumented common practise in the art of the disclosure or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope of the present disclosure and spirit are pointed out by claim below.
Should be understood that, the disclosure is not limited to precision architecture described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope.The scope of the present disclosure is only limited by appended claim.
Claims (11)
1. the recognition methods of filter classification, is characterized in that, described method comprises:
Extract the characteristics of image of image to be identified, described image to be identified is the image after filter process;
According to the characteristics of image of filter category database and described image to be identified, obtain other matching degree of each filter class in described image to be identified and described filter category database, described filter category database is at least for preserving multiple filter classification and adopting the characteristics of image of the rear image of described multiple filter class other places reason;
The filter classification described matching degree being met preset standard is defined as filter classification corresponding to described image to be identified.
2. method according to claim 1, is characterized in that, described method also comprises:
Collect learning sample, described learning sample is the image after the reason of multiple filter class other places;
Extract the characteristics of image of every piece image in described learning sample;
According to characteristics of image and the corresponding filter classification of described every piece image, set up described filter category database.
3. method according to claim 1, is characterized in that, described filter category database also comprises the crucial words for describing each other image effect of filter class;
Described method also comprises:
According to the characteristics of image of described filter category database and described image to be identified, obtain the crucial words corresponding with described image to be identified;
Crucial words corresponding for described image to be identified is defined as filter title corresponding to described image to be identified.
4. method according to claim 3, is characterized in that, the described characteristics of image according to filter category database and described image to be identified, and before obtaining the crucial words corresponding with described image to be identified, described method also comprises:
The set of structure alternative names, described alternative names set is used for storage key word;
According to the crucial words in described alternative names set, collect the image of the predetermined number mated with each crucial words;
The characteristics of image of the image mated with each crucial words described in extracting;
According to the characteristics of image of the image that the crucial words in described alternative names set and described and each crucial words mate, set up described filter category database.
5. method according to claim 1, is characterized in that, described method also comprises:
When the characteristics of image of described image to be identified and other matching degree of arbitrary filter class all do not meet described preset standard, determine that described image to be identified is the image without filter process.
6. a filter classification recognition device, is characterized in that, described device comprises:
Characteristic extracting module, for extracting the characteristics of image of image to be identified, described image to be identified is the image after filter process;
Matching degree acquisition module, for the characteristics of image according to filter category database and described image to be identified, obtain other matching degree of each filter class in described image to be identified and described filter category database, described filter category database is at least for preserving multiple filter classification and adopting the characteristics of image of the rear image of described multiple filter class other places reason;
Determination module, the filter classification for described matching degree being met preset standard is defined as filter classification corresponding to described image to be identified.
7. device according to claim 6, is characterized in that, described device also comprises:
Collect module, for collecting learning sample, described learning sample is the image after the reason of multiple filter class other places;
Described characteristic extracting module is also for extracting the characteristics of image of every piece image in described learning sample;
First sets up module, for according to the characteristics of image of described every piece image and corresponding filter classification, sets up described filter category database.
8. device according to claim 6, is characterized in that, described filter category database also comprises the crucial words for describing each other image effect of filter class;
Described device also comprises:
Crucial words acquisition module, for the characteristics of image according to described filter category database and described image to be identified, obtains the crucial words corresponding with described image to be identified;
Described determination module is also for being defined as filter title corresponding to described image to be identified by crucial words corresponding for described image to be identified.
9. device according to claim 8, is characterized in that, described device also comprises:
Constructing module, for constructing alternative names set, described alternative names set is used for storage key word;
Collection module, for according to the crucial words in described alternative names set, collects the image of the predetermined number mated with each crucial words;
Described characteristic extracting module is also for extracting the characteristics of image of the described image mated with each crucial words;
Second sets up module, for the characteristics of image of image mated according to the crucial words in described alternative names set and described and each crucial words, sets up described filter category database.
10. device according to claim 6, is characterized in that, described determination module also for:
When the characteristics of image of described image to be identified and other matching degree of arbitrary filter class all do not meet described preset standard, determine that described image to be identified is the image without filter process.
11. 1 kinds of filter classification recognition devices, is characterized in that, comprising:
Processor;
For the storer of the executable instruction of storage of processor;
Wherein, described processor is configured to:
Extract the characteristics of image of image to be identified, described image to be identified is the image after filter process;
According to the characteristics of image of filter category database and described image to be identified, obtain other matching degree of each filter class in described image to be identified and described filter category database, described filter category database is at least for preserving multiple filter classification and adopting the characteristics of image of the rear image of described multiple filter class other places reason;
The filter classification described matching degree being met preset standard is defined as filter classification corresponding to described image to be identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510634596.5A CN105224950A (en) | 2015-09-29 | 2015-09-29 | The recognition methods of filter classification and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510634596.5A CN105224950A (en) | 2015-09-29 | 2015-09-29 | The recognition methods of filter classification and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105224950A true CN105224950A (en) | 2016-01-06 |
Family
ID=54993910
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510634596.5A Pending CN105224950A (en) | 2015-09-29 | 2015-09-29 | The recognition methods of filter classification and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105224950A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809174A (en) * | 2016-03-29 | 2016-07-27 | 北京小米移动软件有限公司 | Method and device for identifying image |
| CN109089042A (en) * | 2018-08-30 | 2018-12-25 | Oppo广东移动通信有限公司 | The recognition methods of image procossing mode, device, storage medium and mobile terminal |
| CN109739414A (en) * | 2018-12-29 | 2019-05-10 | 努比亚技术有限公司 | A kind of image processing method, mobile terminal, computer readable storage medium |
| WO2020119254A1 (en) * | 2018-12-10 | 2020-06-18 | 北京达佳互联信息技术有限公司 | Method and device for filter recommendation, electronic equipment, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101394460A (en) * | 2007-09-21 | 2009-03-25 | 索尼株式会社 | Image processing apparatus, image processing method, image processing program, and image capturing apparatus |
| CN104281843A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Image recognition method and system based on self-adaptive features and classification model selection |
| CN104636759A (en) * | 2015-02-28 | 2015-05-20 | 成都品果科技有限公司 | Method for obtaining picture recommending filter information and picture filter information recommending system |
| CN104854859A (en) * | 2012-12-21 | 2015-08-19 | 谷歌公司 | Recommending transformations for photography |
-
2015
- 2015-09-29 CN CN201510634596.5A patent/CN105224950A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101394460A (en) * | 2007-09-21 | 2009-03-25 | 索尼株式会社 | Image processing apparatus, image processing method, image processing program, and image capturing apparatus |
| CN104854859A (en) * | 2012-12-21 | 2015-08-19 | 谷歌公司 | Recommending transformations for photography |
| CN104281843A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Image recognition method and system based on self-adaptive features and classification model selection |
| CN104636759A (en) * | 2015-02-28 | 2015-05-20 | 成都品果科技有限公司 | Method for obtaining picture recommending filter information and picture filter information recommending system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809174A (en) * | 2016-03-29 | 2016-07-27 | 北京小米移动软件有限公司 | Method and device for identifying image |
| CN105809174B (en) * | 2016-03-29 | 2019-04-30 | 北京小米移动软件有限公司 | Identify the method and device of image |
| CN109089042A (en) * | 2018-08-30 | 2018-12-25 | Oppo广东移动通信有限公司 | The recognition methods of image procossing mode, device, storage medium and mobile terminal |
| WO2020119254A1 (en) * | 2018-12-10 | 2020-06-18 | 北京达佳互联信息技术有限公司 | Method and device for filter recommendation, electronic equipment, and storage medium |
| CN109739414A (en) * | 2018-12-29 | 2019-05-10 | 努比亚技术有限公司 | A kind of image processing method, mobile terminal, computer readable storage medium |
| CN109739414B (en) * | 2018-12-29 | 2021-12-14 | 努比亚技术有限公司 | Picture processing method, mobile terminal and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104850828A (en) | Person identification method and person identification device | |
| CN105095873A (en) | Picture sharing method and apparatus | |
| CN104753766A (en) | Expression sending method and device | |
| CN105517112A (en) | Method and device for displaying WiFi network information | |
| CN105302315A (en) | Image processing method and device | |
| CN105095881A (en) | Method, apparatus and terminal for face identification | |
| CN105335712A (en) | Image recognition method, device and terminal | |
| CN104035558A (en) | Terminal device control method and device | |
| CN105096144A (en) | Social relation analysis method and social relation analysis device | |
| CN104933170A (en) | Information exhibition method and device | |
| CN104717366A (en) | Method and device for recommending contact photos | |
| CN104869574A (en) | Wireless network connection method and apparatus | |
| CN104700353A (en) | Image filter generating method and device | |
| CN105426515A (en) | Video classification method and apparatus | |
| CN105046231A (en) | Face detection method and device | |
| CN106250921A (en) | Image processing method and device | |
| CN104731880A (en) | Image ordering method and device | |
| CN105488511A (en) | Image identification method and device | |
| CN105184313A (en) | Classification model construction method and device | |
| CN106652113A (en) | Access control method and device | |
| CN105426485A (en) | Image combination method and device, intelligent terminal and server | |
| CN105069426A (en) | Similar picture determining method and apparatus | |
| CN105205494A (en) | Similar picture identification method and device | |
| CN105354560A (en) | Fingerprint identification method and device | |
| CN104933419A (en) | Method and device for obtaining iris images and iris identification equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20160106 |