CN103324697B - A kind of android application searches mountain vallage application elimination method based on icon contrast - Google Patents

Abstract

The present invention is the mountain vallage application elimination method of a kind of android application searches, and it comprises the following steps: captured qualified application by reptile network；Judge whether this qualified application grabbed exists same icon with the application of storage in authentication database, if the application existed in authentication database and grab bag name is different but application that icon is identical, can be judged as that mountain vallage is applied, if not existing, then this qualified application belongs to new application；Judge whether this qualified application grabbed has identical signature with this application in authentication database；If the application of storage is the most identical with the application Bao Mingyu signature grabbed in authentication database, legal copy can be judged as；If judged result is no, it is judged as that mountain vallage is applied.This invention uses bag name, signature and the mode of icon certification, effectively eliminates the mountain vallage application in retrieving.

Description

A kind of android application searches mountain vallage application elimination method based on icon contrast

Technical field

The present invention relates to a kind of search elimination method, particularly to a kind of android application searches mountain vallage application elimination method based on icon contrast and signature comparison.

Background technology

Android application searches field, presently, there are a serious problem, it is simply that mountain vallage application emerges in an endless stream, and various application markets all exist similar (or identical) application that a large number of users does not appears to differentiate, and make user at a loss as to what to do；The generator of mountain vallage application, has two kinds of main packing manner: one is not change former application bag name, uses own signature to issue after adding malicious code；Two is amendment application bag name, uses own signature to issue simultaneously.Even above Google Play, there is also the mountain vallage software of many this forms, in third party market, this problem is the most universal.In manual examination and verification, this problem reasonable can carry out artificial filter by the knowledge of auditor in the most well-known application.But for the most well-known software, and in the whole network application captures, because with simple standard, artificial or web crawlers all cannot judge whether an application is mountain vallage application, therefore this problem is more thorny.

Refer to Chinese invention patent: " web search method and device ", application number: CN201210548469.X, disclose a kind of web search method and device, browser searches hurdle loads preset search engine classification information and generates search engine list of categories；The search keyword of input is received in browser searches hurdle input frame；Receive the selection information to the search engine classification in described search engine list of categories；The search engine list that search engine classification selected by loading is corresponding；Receive the selection information to the search engine in described search engine list, and initiate searching request according to described search keyword to selected search engine.Using search key to form search engine list in this patent, do not carry out differentiating and confirming in search procedure, the information retrieved easily produces the similar information that cannot differentiate.

In view of drawbacks described above, creator of the present invention obtains this creation finally through research for a long time and practice.

Summary of the invention

It is an object of the invention to provide the mountain vallage application elimination method of a kind of android application searches, in order to overcome above-mentioned technological deficiency.

For achieving the above object, the present invention provides the mountain vallage application elimination method of a kind of android application searches, and it comprises the following steps:

Step a, is captured qualified application by reptile network；

Step b, it is judged that the qualified application grabbed in above-mentioned steps a whether with authentication database present in application there is identical bag name；If application exists identical bag name present in the qualified application grabbed and authentication database, then jump to following step d；If not existing, then jump to following step c；

Step c, judge whether this qualified application grabbed exists same icon with the application of storage in authentication database, if the application existed in authentication database and grab bag name is different but application that icon is identical, can be judged as that mountain vallage is applied, then be jumped to following step e；If not existing, then this qualified application belongs to new application, jumps to following step f；

Step d, it is judged that whether this qualified application grabbed has identical signature with this application in authentication database；If the application of storage is the most identical with the application Bao Mingyu signature grabbed in authentication database, legal copy can be judged as, jumped to following step f；If judged result is no, it is judged as that mountain vallage is applied, is jumped to following step e；

Step e, rejects this application grabbed from network, jumps to following step h；

Step f, adds this application grabbed to legal application references list；

Step g, the legal application references list that will obtain in above-mentioned steps f, collated rear transmission presents to user to front end；

Step h, terminates.

It is also preferred that the left the forming process of described authentication database is:

Step b1, it would be desirable to be included the legal application into authentication database and send certification request to authentication database；

Step b2, the legal application filed a request is authenticated by described authentication database；

Step b3, will be resolved by the legal application of above-mentioned steps b2 certification, obtains legal author's signature；

Step b4, obtains legal application icon；

Step b5, obtains legal copy application bag name from certification legal copy is applied and is stored in authentication signature storehouse；

Step b6, becomes long string of sequence code by legal copy author signature through hash function computing and for storage and compares, digital signature be stored in authentication signature storehouse with the form of above-mentioned long string sequence code；

The bag name of all of legal application is stored to authentication signature storehouse by step b7；

Step b8, stores the icon of all of legal application to application icon storehouse；

Step b9, the application icon stored to described application icon storehouse through above-mentioned steps b8 forms described application icon storehouse；

Step b10, stores to the authentication signature in described authentication signature storehouse and the legal bag name of storage through above-mentioned steps b7 and forms described authentication signature storehouse；

Step b11, described authentication signature storehouse and application icon storehouse form described authentication database.

It is also preferred that the left the process of icon contrast is in described step c:

Apply present in step c1, the application grabbed and authentication database and there is not identical bag name；

Step c2, obtains the application icon grabbed；

Step c3, application icon above-mentioned steps c2 grabbed is sent to application icon storehouse and detects；

Step c4, detects the icon in all application icon storehouses；The icon that if the application icon existed in authentication database and grab is identical, bag name is different, then jump to step c5；If not existing, then jump to step c6；

Step c5, jumps to above-mentioned steps e；

Step c6, jumps to above-mentioned steps f.

It is also preferred that the left the process of signature comparison is in described step d:

Present in step d1, the application searched out and authentication database there is identical bag name in application；

Step d2, captures the signature of this application grabbed；

Step d3, sends to the censorship of authentication signature storehouse by the signing messages that described step d1 obtains；

Step d4, positions the signature corresponding to bag name identical in the authentication database detected in above-mentioned steps d2 in authentication signature storehouse；

Step d5, the signature detecting the censorship of described step d3 institute is the most identical with the authentication signature of location in described step d4；If identical, then jump to following step d6；If it is different, then jump to following step d7；

Step d6, jumps to above-mentioned steps e；

Step d7, jumps to above-mentioned steps f.

It is also preferred that the left in above-mentioned steps a, user input key word in front end, the key word that described reptile network is inputted according to user scans in network-wide basis, and crawl meets the android of keyword condition and applies and form Preliminary List.

It is also preferred that the left the process in above-mentioned steps c contrasted the icon in all application icon storehouses, using the method for template matching to compare, its process is:

Step c11, unifies size by the icon of the icon in authentication database with the application grabbed；

Icon in step c12, relatively described authentication database and the average color of the icon of the application grabbed, if average color difference is less than a preset value k (k ＞ 0), then jumps to following step c13, otherwise, jump to above-mentioned steps e；

Icon in step c13, relatively described authentication database is the template matching of multiplication based on standard relevant matches operation with the icon of the application grabbed, and calculates matching degree R₁If, R₁More than predetermined threshold value K₁(K₁>=0), then following step c14 is performed；If it is not, then jump to above-mentioned steps e；

The template matching of the multiplication operation of the icon relative value to average based on standard relevant matches with the icon of the application grabbed in step c14, relatively described authentication database, calculates matching degree R₂If, R₂More than predetermined threshold value K₂(K₂>=0), then following step c15 is performed；If it is not, then jump to above-mentioned steps e；

The template matching that icon in step c15, relatively described authentication database mates based on standard deviation with the icon of the application grabbed, calculates matching degree R₃If, R₃More than predetermined threshold value K₃(K₃>=0), then device above-mentioned steps f is jumped；If it is not, then jump to above-mentioned steps e.

It is also preferred that the left matching degree R in above-mentioned steps c13₁, calculate according to following formula (1),

$R_1(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T(x^{\′},y^{\′})*I(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}T{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}I{(x+x^{\′},y+y^{\′})}^2}}---\left(1\right)$

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed；

Matching degree R in above-mentioned steps c14₂, calculate according to following formula (2),

$R_2(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T^{\′}(x^{\′},y^{\′})*I^{\′}(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}{T}^{\′}{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}{I}^{\′}{(x+x^{\′},y+y^{\′})}^2}}---\left(2\right)$

Wherein, T ' and I ' is defined by following formula (3), (4),

T ' (x ', y ')=T (x ', y ')-1/ (w h) ∑_{x ", y ″}T (x ", y ") (3)

I ' (x+x ', y+y ')=I (x+x ', y+y ')-1/ (w h) ∑_{x ", y ″}T (x+x ", y+y ") (4)

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；X "; y " representing the transverse and longitudinal coordinate figure of the truncated picture of the icon of the application grabbed respectively, w represents the width of truncated picture, and h represents the height of truncated picture, T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed；

Matching degree R in above-mentioned steps c15₃, calculate according to following formula (5),

$R_3(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T(x^{\′},y^{\′})-I(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}T{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}I{(x+x^{\′},y+y^{\′})}^2}}---\left(5\right)$

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed.

It is also preferred that the left described preset value k is 10 degree；Described threshold k₁Value is 0.99；Described threshold k₂Value is 0.95；Described threshold k₃Value is 0.01.

It is also preferred that the left described authentication database includes an authentication signature storehouse and an application icon storehouse, the bag name in described authentication signature storehouse and signature one_to_one corresponding.

It is also preferred that the left the execution program of the judge process of above-mentioned bag name, signature and icon is stored in a contrast device, described contrast device obtains treating in reptile network respectively and captures application and the bag name of application in authentication database, signature, icon information；And by result transmission a to index data base, forming described legal application references list, collated rear transmission is to front end.

The beneficial effects of the present invention is compared with the prior art: the mountain vallage application elimination method that android of the present invention application captures, successively bag name is carried out by needing the application captured application in an authentication database, signature, the comparison of icon, can effectively differentiate that the application grabbed belongs to the mountain vallage application to legal copy application amendment bag name, or the application grabbed is brand-new application, or the application grabbed belongs to succession original bag name and legal copy application is carried out mountain vallage, or it is identical with the application bag name grabbed to there is application in authentication database, the multiple situation of the mountain vallage application signing different, through three kinds of different way of contrast, effectively reject mountain vallage application；In judge process, it is with reference to a preformed authentication database, and this authentication database includes application icon storehouse and authentication signature storehouse, enables described icon certification and signature authentication to carry out in order.

Accompanying drawing explanation

Fig. 1 is the flow chart of the mountain vallage application elimination method first embodiment of android application searches of the present invention；

Fig. 2 is the flow chart that the authentication database of the mountain vallage application elimination method of android application searches of the present invention realizes process；

Fig. 3 is the flow chart of the icon contrast implementation method of the mountain vallage application elimination method of android application searches of the present invention；

Fig. 4 do android application searches of the present invention mountain vallage application elimination method in the flow chart of icon template matching process；

Fig. 5 is the flow chart of the signature comparison implementation method of the mountain vallage application elimination method of android application searches of the present invention.

Detailed description of the invention

Below in conjunction with accompanying drawing, to the present invention, above-mentioned and other technical characteristic and advantage are described in more detail.

Embodiment one:

Android application searches mountain vallage application elimination method of the present invention can realize according to the following steps:

Referring to shown in Fig. 1, it is the flow chart of mountain vallage application elimination method first embodiment of android application searches of the present invention, and this process is:

Step a101: captured qualified application by reptile network, particularly as follows: user inputs key word in front end, reptile network grabs and meets the android of keyword condition and apply and form Preliminary List；

Step a102: judge the application that grabs whether with authentication database present in application there is identical bag name, authentication database is delivered in application in the list that described step a101 is formed successively, the bag name that the full storehouse of Bao Mingyu of the application grabbed is applied is compared, it may be judged whether there is identical bag name；If application exists identical bag name present in the application grabbed and authentication database, then jump to step a104；If not existing, then jump to step a103；Authentication database in this invention includes an application icon storehouse and an authentication signature storehouse, wherein, the bag name in described authentication signature storehouse and signature one_to_one corresponding.

Step a103: judge this application grabbed whether with authentication database present in application there is same icon, it is specially, if judging to there is not identical bag name through described step a102, then the icon of the application grabbed is determined whether, if the determination result is YES, the application existed in authentication authorization and accounting data base and grab bag name is different but application that icon is identical, can be judged as that mountain vallage is applied, then be jumped to step a105；If not existing, the application situation that bag name is different, icon is the most different existing in authentication authorization and accounting data base from grabbing, belong to new application, then jump to step a106；

Step a104: judge whether this application grabbed has identical signature with this application in authentication database；If judging to there is identical bag name through described step 102, then determining whether the signature of the application grabbed, if the determination result is YES, it is the most identical with the application Bao Mingyu signature grabbed to there is application in authentication authorization and accounting data base, legal copy can be judged as, jumped to step a106；

If judging to there is identical bag name through step a102, then determining whether the signature of the application grabbed, if judged result is no, and in authentication database, existence application is identical from the application bag name grabbed, it is different to sign, mountain vallage can be judged as, jumped to step a106.

Step a105: will be judged to that bag name is identical, signature is different through above-mentioned steps a102, a103, a104 step, or bag name is different, icon is identical, then this application grabbed rejected from network；

Step a106: will be judged to that bag name is identical, it is identical to sign through above-mentioned steps a102, a103, a104 step, or bag name is different, icon is different, then add this application grabbed to legal application references list；

Step a107: screen the legal application references list obtained by above-mentioned steps a101, a102, a103, a104, a106, collated rear transmission presents to user to front end.

The mountain vallage rejected by above-mentioned steps a101, a102, a103, a104, a105 screening is applied, and will completely delete, and will not again grab by the most crawled engine from application network.

Referring to shown in Fig. 2, it is the flow chart of authentication database forming process of mountain vallage application elimination method of android application searches of the present invention, and this process is:

Step a201: legal application request certification；Specifically, wish that being included the legal application into authentication database sends certification request to authentication database；

Step a202: the legal application filed a request is authenticated by authentication database；Specifically, first authentication database detects the most legal application of this application, perform following step a203, a204, a205 respectively；

Step a203: obtain legal author's signature；Specifically, resolve this application afterwards by the described step 202 of legal certification, obtain legal author's signature；

Step a204: obtain legal application icon；Specifically, obtain legal application icon from authentication application and be stored in application icon storehouse；

Step a205: obtain legal bag name；Specifically, obtain legal copy application bag name from certification legal copy is applied and be stored in authentication signature storehouse；

Step a206: hashed functional operation becomes sequence code；After being specially acquisition legal copy application author's signature, legal copy author signature is become long string of sequence code through hash function computing and for storage and compares, digital signature is stored in authentication signature storehouse with the form of above-mentioned long string sequence code；

Step a207: store to authentication signature storehouse；Specifically, the bag name of all of legal application is stored to authentication signature storehouse；

Step a208: store to application icon storehouse；Specifically, the icon of all of legal application is stored to application icon storehouse；

Step a209: form application icon storehouse；Specifically, store the application icon formation application icon storehouse to application icon storehouse through step 208；

Step a210: form authentication signature storehouse；Specifically, store the legal bag name formation authentication signature storehouse of the authentication signature to authentication signature storehouse and storage through step 207；

Step a211: form authentication database, specifically, authentication signature storehouse and application icon storehouse form authentication database.

Step a103: judge this application grabbed whether with authentication database present in application there is same icon, it is specially, if judging to there is not identical bag name through described step a102, then the icon of the application grabbed is determined whether, if the determination result is YES, the application existed in authentication authorization and accounting data base and grab bag name is different but application that icon is identical, can be judged as that mountain vallage is applied, then be jumped to step a105；If not existing, the application situation that bag name is different, icon is the most different existing in authentication authorization and accounting data base from grabbing, belong to new application, then jump to step a106；

Referring to shown in Fig. 3, it is the flow chart of icon contrast implementation method of mountain vallage application elimination method of android application searches of the present invention；

Step a300: apply present in the application grabbed and authentication database and there is not identical bag name；

Step a301: obtain the application icon that grabs, if specifically, judged result is to be not detected by there is the application of identical bag name, the then application icon grabbed in authentication database；

Step a302: to the censorship of application icon storehouse, detects specifically, the application icon that described step a301 grabs is sent to application icon storehouse；

Step a303: image detects: whether template matching, it is specially, icon in all application icon storehouses is detected, scope is all application icons in application icon storehouse, the method using template matching in the present embodiment detects, and the icon of all application icons in authentication database with this crawl application is carried out comparison one by one；It is specially, if mating through step a303 testing result, authentication database then exists an application and the application grabbed, there is the situation that icon is identical, bag name is different, can assert, this application grabbed belongs to the mountain vallage application to legal copy application amendment bag name, is judged to mountain vallage and jumps to step a304；If testing result is not mated, then existing application bag name differs with authentication database, icon also differs, and belongs to brand-new application in this application grabbed, and is not belonging to mountain vallage and jumps to step a305；

Step a304: then (with above-mentioned steps a105) is rejected in this application grabbed from network and jump to step a105；

Step a305: then this application grabbed is added to legal application references list (with above-mentioned steps a105).

Wherein, also can add after described step a302, step a303 step the application grabbed is carried out the step of signature detection.

Refer to shown in Fig. 4, its be android application searches of the present invention mountain vallage application elimination method in the flow chart of icon template matching process, in above-mentioned steps a303, described in the icon that grabs and the icon in the authentication database process that carries out template be:

Step a41, unifies size by the icon of the icon in authentication database with the application grabbed；Depending on its size is according to numerical value set in advance；

Icon in step a42, relatively described authentication database and the average color of the icon of the application grabbed, if average color difference is less than a preset value k (k ＞ 0), then jumps to following step a43, otherwise, jump to above-mentioned steps a304；In the present embodiment, this preset value k is 10 degree；

Icon in step a43, relatively described authentication database is the template matching of multiplication based on standard relevant matches operation with the icon of the application grabbed, and calculates matching degree R₁If, R₁More than predetermined threshold value K₁, then following step a44 is performed；If it is not, then jump to step a46；This threshold k in the present embodiment₁(K₁>=0) value is 0.99；

Above-mentioned calculating matching degree R₁Carry out according to following formula (1),

$R_1(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T(x^{\′},y^{\′})*I(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}T{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}I{(x+x^{\′},y+y^{\′})}^2}}---\left(1\right)$

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, x ', y ' represents the transverse and longitudinal coordinate figure of the icon in described authentication database respectively, and T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed.

The template matching of the multiplication operation of the icon relative value to average based on standard relevant matches with the icon of the application grabbed in step a44, relatively described authentication database, calculates matching degree R₂If, R₂More than predetermined threshold value K₂(K₂>=0), then following step a45 is performed；If it is not, then jump to step a46；This threshold k in the present embodiment₂Value is 0.95；

Above-mentioned calculating matching degree R₂Carry out according to following formula (2),

$R_2(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T^{\′}(x^{\′},y^{\′})*I^{\′}(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}{T}^{\′}{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}{I}^{\′}{(x+x^{\′},y+y^{\′})}^2}}---\left(2\right)$

Wherein, T ' and I ' is defined by following formula (3), (4),

T ' (x ', y ')=T (x ', y ')-1/ (w h) ∑_{x ", y ″}T (x ", y ") (3)

I ' (x+x ', y+y ')=I (x+x ', y+y ')-1/ (w h) ∑_{x ", y ″}T (x+x ", y+y ") (4)

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；X "; y " representing the transverse and longitudinal coordinate figure of the truncated picture of the icon of the application grabbed respectively, w represents the width of truncated picture, and h represents the height of truncated picture, T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed.

The template matching that icon in step a45, relatively described authentication database mates based on standard deviation with the icon of the application grabbed, calculates matching degree R₃If, R₃More than predetermined threshold value K₃(K₃>=0), then device above-mentioned steps a305 is jumped；If it is not, then jump to step a46；This threshold k in the present embodiment₃Value is 0.01；

Above-mentioned calculating matching degree R₃Carry out according to following formula (5),

$R_3(x,y)=\frac{{\mathrm{\Σ}}_{x^{\′},y^{\′}}(T(x^{\′},y^{\′})-I(x+x^{\′},y+y^{\′}))}{\sqrt{{\mathrm{\Σ}}_{x^{\′},y^{\′}}T{(x^{\′},y^{\′})}^2*{\mathrm{\Σ}}_{x^{\′},y^{\′}}I{(x+x^{\′},y+y^{\′})}^2}}---\left(5\right)$

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, x ', y ' represents the transverse and longitudinal coordinate figure of the icon in described authentication database respectively, and T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed.

Step a46, jumps to above-mentioned steps a304；

Step a47, jumps to above-mentioned steps a305.

After the described icon grabbed carries out above-mentioned template matching with the icon in authentication database, then can conclude that whether this application program grabbed is brand-new application program.

Referring to shown in Fig. 5, it is the flow chart of signature comparison implementation method of mountain vallage application elimination method of android application searches of the present invention.

Step a400: the application grabbed judges via authentication database, exists and has identical bag name with the application of institute's certification storage in authentication database；

Step a401: if judged result exists the application of identical bag name for detecting in authentication database, then capture the signature of this application grabbed；

Step a402: the signing messages that described step 401 obtains is sent to the censorship of authentication signature storehouse；

Step a403: the signature corresponding to bag name identical in the authentication database detected in positioning step 102 in authentication signature storehouse；

Step a404: the signature detecting the censorship of described step 402 is the most identical with the authentication signature of location in described step 403；If described step 402 testing result is different, authentication database then exists authentication application there is from the application grabbed identical bag name and different signatures, can assert, the application grabbed belongs to inherits the method that original bag name carries out mountain vallage to legal copy application, owing to it cannot replicate the signature of original author, so being identified by this kind of method, if testing result identical, the signature difference that is bag name, and jump to step a405；If step 402 testing result is identical, then authentication database exists authentication application and the application grabbed there is identical bag name and signature, then can determine that grabbed application is legal copy, and jump to step a406；

Step a405: jump to following step a105；

Step a406: jump to following step a106；

Wherein, also can add after above-mentioned steps a402, a403, a404 the application grabbed is carried out the step of icon detection.

In the present invention, the execution program of the judge process of above-mentioned bag name, signature and icon is stored in a contrast device, and described contrast device obtains treating in reptile network respectively and captures application and the bag name of application in authentication database, signature, icon information；And by result transmission a to index data base, forming described legal application references list, collated rear transmission is to front end.

The foregoing is only presently preferred embodiments of the present invention, be merely illustrative for invention, and nonrestrictive.Those skilled in the art understands, it can be carried out many changes, amendment, even equivalence, but fall within protection scope of the present invention in the spirit and scope that invention claim is limited.

Claims (9)

1. the mountain vallage application elimination method of an android application searches, it is characterised in that it comprises the following steps:

Step a, is captured qualified application by reptile network；

Step b, it is judged that the qualified application grabbed in above-mentioned steps a whether with authentication database present in application there is identical bag name；If application exists identical bag name present in the qualified application grabbed and authentication database, then jump to following step d；If not existing, then jump to following step c；

Step c, judge whether this qualified application grabbed exists same icon with the application of storage in authentication database, if the application existed in authentication database and grab bag name is different but application that icon is identical, can be judged as that mountain vallage is applied, then be jumped to following step e；If not existing, then this qualified application belongs to new application, jumps to following step f；

Step d, it is judged that whether this qualified application grabbed has identical signature with this application in authentication database；If the application of storage is the most identical with the application Bao Mingyu signature grabbed in authentication database, legal copy can be judged as, jumped to following step f；If judged result is no, it is judged as that mountain vallage is applied, is jumped to following step e；

Step e, rejects this application grabbed from network, jumps to following step h；

Step f, adds this application grabbed to legal application references list；

Step g, the legal application references list that will obtain in above-mentioned steps f, collated rear transmission presents to user to front end；

Step h, terminates；

The forming process of described authentication database is:

Step b1, it would be desirable to be included the legal application into authentication database and send certification request to authentication database；

Step b2, the legal application filed a request is authenticated by described authentication database；

Step b3, will be resolved by the legal application of above-mentioned steps b2 certification, obtains legal author's signature；

Step b4, obtains legal application icon；

Step b5, obtains legal copy application bag name from certification legal copy is applied and is stored in authentication signature storehouse；

Step b6, becomes long string of sequence code by legal copy author signature through hash function computing and for storage and compares, digital signature be stored in authentication signature storehouse with the form of above-mentioned long string sequence code；

The bag name of all of legal application is stored to authentication signature storehouse by step b7；

Step b8, stores the icon of all of legal application to application icon storehouse；

Step b9, the application icon stored to described application icon storehouse through above-mentioned steps b8 forms described application icon storehouse；

Step b10, the legal bag name stored to the authentication signature in described authentication signature storehouse and the storage of described step b7 through above-mentioned steps b6 forms described authentication signature storehouse；

Step b11, described authentication signature storehouse and application icon storehouse form described authentication database.

The mountain vallage application elimination method of android application searches the most according to claim 1, it is characterised in that in described step c, the process of icon contrast is:

Apply present in step c1, the application grabbed and authentication database and there is not identical bag name；

Step c2, obtains the application icon grabbed；

Step c3, application icon above-mentioned steps c2 grabbed is sent to application icon storehouse and detects；

Step c4, detects the icon in all application icon storehouses；The icon that if the application icon existed in authentication database and grab is identical, bag name is different, then jump to step c5；If not existing, then jump to step c6；

Step c5, jumps to above-mentioned steps e；

Step c6, jumps to above-mentioned steps f.

The mountain vallage application elimination method of android application searches the most according to claim 1 and 2, it is characterised in that in described step d, the process of signature comparison is:

Present in step d1, the application searched out and authentication database there is identical bag name in application；

Step d2, captures the signature of this application grabbed；

Step d3, sends to the censorship of authentication signature storehouse by the signing messages that described step d2 obtains；

Step d4, positions the signature corresponding to bag name identical in the authentication database detected in above-mentioned steps d2 in authentication signature storehouse；

Step d5, the signature detecting the censorship of described step d3 institute is the most identical with the authentication signature of location in described step d4；If identical, then jump to following step d6；If it is different, then jump to following step d7；

Step d6, jumps to above-mentioned steps f；

Step d7, jumps to above-mentioned steps e.

The mountain vallage application delet method of android application searches the most according to claim 1, it is characterized in that, in above-mentioned steps a, key word is inputted in front end by user, the key word that described reptile network is inputted according to user scans in network-wide basis, and crawl meets the android of keyword condition and applies and form Preliminary List.

The mountain vallage application delet method of android application searches the most according to claim 1, it is characterised in that the process in above-mentioned steps c contrasted the icon in all application icon storehouses, uses the method for template matching to compare, and its process is:

Step c11, unifies size by the icon of the icon in authentication database with the application grabbed；

Icon in step c12, relatively described authentication database and the average color of the icon of the application grabbed, if average color difference is less than a preset value k (k ＞ 0), then jumps to following step c13, otherwise, jump to above-mentioned steps e；

Icon in step c13, relatively described authentication database is the template matching of multiplication based on standard relevant matches operation with the icon of the application grabbed, and calculates matching degree R₁If, R₁More than predetermined threshold value K₁(K₁>=0), then following step c14 is performed；If it is not, then jump to above-mentioned steps e；

The template matching of the multiplication operation of the icon relative value to average based on standard relevant matches with the icon of the application grabbed in step c14, relatively described authentication database, calculates matching degree R₂If, R₂More than predetermined threshold value K₂(K₂>=0), then following step c15 is performed；If it is not, then jump to above-mentioned steps e；

The template matching that icon in step c15, relatively described authentication database mates based on standard deviation with the icon of the application grabbed, calculates matching degree R₃If, R₃More than predetermined threshold value K₃(K₃>=0), then above-mentioned steps f is jumped to；If it is not, then jump to above-mentioned steps e.

The mountain vallage application delet method of android application searches the most according to claim 5, it is characterised in that matching degree R in above-mentioned steps c13₁, calculate according to following formula (1),

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed；

Matching degree R in above-mentioned steps c14₂, calculate according to following formula (2),

Wherein, T ' and I ' is defined by following formula (3), (4),

T ' (x ', y ')=T (x ', y ')-1/ (w h) ∑_{x ", y ″}T (x ", y ") (3)

I ' (x+x ', y+y ')=I (x+x ', y+y ')-1/ (w h) ∑_{x ", y ″}T (x+x ", y+y ") (4)

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；X "; y " representing the transverse and longitudinal coordinate figure of the truncated picture of the icon of the application grabbed respectively, w represents the width of truncated picture, and h represents the height of truncated picture, T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed；

Matching degree R in above-mentioned steps c15₃, calculate according to following formula (5),

In formula, x, y represent the transverse and longitudinal coordinate figure of the icon of the application grabbed respectively, and x ', y ' represent the transverse and longitudinal coordinate figure of the icon in described authentication database respectively；T represents the pixel computing of the icon in authentication database, and I represents the pixel computing of the icon of the application grabbed.

The mountain vallage application elimination method of android application searches the most according to claim 5, it is characterised in that described preset value k is 10 degree；Described threshold k₁Value is 0.99；Described threshold k₂Value is 0.95；Described threshold k₃Value is 0.01.

The mountain vallage application elimination method of android application searches the most according to claim 1, it is characterised in that described authentication database includes an authentication signature storehouse and an application icon storehouse, the bag name in described authentication signature storehouse and signature one_to_one corresponding.

The mountain vallage application elimination method of android application searches the most according to claim 1, it is characterized in that, the execution program of the judge process of above-mentioned bag name, signature and icon is stored in a contrast device, and described contrast device obtains treating in reptile network respectively and captures application and the bag name of application in authentication database, signature, icon information；And by result transmission a to index data base, forming described legal application references list, collated rear transmission is to front end.