TW200813876A - System and method for determining similarity of curves - Google Patents
System and method for determining similarity of curves Download PDFInfo
- Publication number
- TW200813876A TW200813876A TW95132532A TW95132532A TW200813876A TW 200813876 A TW200813876 A TW 200813876A TW 95132532 A TW95132532 A TW 95132532A TW 95132532 A TW95132532 A TW 95132532A TW 200813876 A TW200813876 A TW 200813876A
- Authority
- TW
- Taiwan
- Prior art keywords
- line
- feature
- characteristic
- shape
- control
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000004364 calculation method Methods 0.000 claims abstract description 25
- 239000000463 material Substances 0.000 claims description 15
- 230000000007 visual effect Effects 0.000 claims description 6
- 239000010931 gold Substances 0.000 claims description 3
- 229910052737 gold Inorganic materials 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 2
- 241000233866 Fungi Species 0.000 claims 1
- 208000027418 Wounds and injury Diseases 0.000 claims 1
- 230000006378 damage Effects 0.000 claims 1
- 208000014674 injury Diseases 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 claims 1
- 210000003205 muscle Anatomy 0.000 claims 1
- 230000006870 function Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 3
- 101100496114 Caenorhabditis elegans clc-2 gene Proteins 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004438 eyesight Effects 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000016776 visual perception Effects 0.000 description 2
- 241000255925 Diptera Species 0.000 description 1
- 229910052770 Uranium Inorganic materials 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- JFALSRSLKYAFGM-UHFFFAOYSA-N uranium(0) Chemical compound [U] JFALSRSLKYAFGM-UHFFFAOYSA-N 0.000 description 1
- PICXIOQBANWBIZ-UHFFFAOYSA-N zinc;1-oxidopyridine-2-thione Chemical class [Zn+2].[O-]N1C=CC=CC1=S.[O-]N1C=CC=CC1=S PICXIOQBANWBIZ-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
200813876 九、發明說明： 【發明所屬之技術領域】 本發明係有關於一種判斷線形相似度的技術領域，尤其係特 別有關於一種藉由選擇複數個特徵點、計算複數條特徵線之斜率 、進而判斷出二條曲線之線形相似度的技術領域。 【先前技術】 現今社會中，電腦及軟體已成為多數人經常所使用的工具， 而人們經常將各種數據應用電腦軟體繪製成各種圖表並進行比對 ，以獲得某種參考意義。一般來說，應用電腦來進行數據的比對 相當容易，但若要針對不規則的曲線來進行比對，判斷線形之間 的近似程度，由於曲線並未有明顯的規則變化，而是否相似又往 往取決於人類視覺的感官，因此往往十分困難。 判斷線形相似度的應用之一，例如在金融交易上，許多投資 人喜歡使用各種線圖來判斷股票、期貨或是外匯交易之趨勢，以 作為決定買賣交易之依據，例如經常聽到的V型、W底、頭肩形 等等。然而目前各式各樣的金融交易數量相當眾多，每一種金融 商品又可能具有多種線圖或技術指標圖，如走勢圖、K線圖、KD 圖、MACD圖、RSI圖等等，而各金融商品的線圖其線形也經常 變化多端，並非都是標準的線形，投資人往往很難直接從數以千 計的各種線圖中找出自己偏好之線形的金融商品。因此如果能夠 開發出一種工具，讓投資人可以挑選某一樣本線形，然後比對出 與該樣本線形相似度的線形及其對應金融商品，提供投資人參考 » ( » f ，將可讓投資人感到非常便利，快速地找到其所偏好的金融商品 ，可提高該投資人獲利的機會。 5 200813876 者可以彻工具及魏，讓使用 照線形的相似度，讓 、本線形與其他待篩選之對 要進行線形匈斷的應卜可#應用在金融商品交易或是其他需 數個待篩選之對照線 項功能更進-步時，可依據複200813876 IX. Description of the Invention: [Technical Field] The present invention relates to a technical field for judging linear similarity, in particular, a method for calculating a slope of a plurality of feature points by selecting a plurality of feature points, and further The technical field of determining the linear similarity of the two curves. [Prior Art] In today's society, computers and software have become the tools that most people often use. People often draw various data application computer software into various charts and compare them to obtain some reference significance. In general, it is quite easy to use a computer to compare data, but if you want to compare against irregular curves, judge the degree of approximation between the lines, because the curve does not have obvious rule changes, and whether it is similar or not It often depends on the senses of human vision, so it is often very difficult. One of the applications for judging linear similarity, for example, in financial transactions, many investors like to use various line graphs to judge the trend of stocks, futures or foreign exchange transactions, as a basis for determining trading transactions, such as the V type that is often heard. W bottom, head and shoulders and so on. However, the current number of various financial transactions is quite numerous. Each financial product may have a variety of line graphs or technical indicators, such as charts, K-line charts, KD charts, MACD charts, RSI charts, etc., and each financial The line drawings of goods are also often varied. They are not all standard linear shapes. It is often difficult for investors to find their own linear financial products directly from thousands of various line drawings. So if you can develop a tool that allows investors to pick a sample line shape and then compare the line shape that is similar to the sample line and its corresponding financial products, provide an investor reference » ( » f , which will allow investors It is very convenient to quickly find the financial products that he prefers, which can increase the chance of the investor's profit. 5 200813876 Those who can use the tool and Wei, let the use of the similarity of the line shape, the line shape and other to be screened For the application of the line-shaped hexagram, the application should be based on the financial commodity transaction or other functions that require several control line items to be screened.
照線形相對應的 資气泠^度的同低來進行排列，並提供與該對 關技術，因Ζ ，、本案如=者$知技術中，均未有出現類似相 的系統及方法，更進一提出種此夠自動判斷二條線形相似度 金融商品的報價、應^金融交易市場上’以輔助 動能、市場交易料1軟體，來協助《人判斷整體的市場 行金融商品的交易並獲取較高利益，助投資人進 發明内容】 法，可 m目r#…種_綱料的系統及方 數條線::選擇複數個特徵點，複 之線形相似度。 訂…、深办 本發狀另—目的係在於經由本發明所提出的㈣及方法， 、所判斷出該樣本線形和對照線形的相似度，符合人類視覺對相 似程度的觀感，大幅提高了使用者對判斷線形相似度的滿音度。 經由本發明系統及方法所計算出高近似度之線形，其外觀將^合 人類視覺上的相似。 ，本發明之另一目的係在於經由本發叼所提出的系統及穸法， 其更進-步應用在其他的應用軟體時，使用者只需經由簡單的操 作介面選擇某一樣本線形，即可將複數個對照線形與該樣本線形 200813876 進行相似度判斷，自動篩選出與該樣本線形相近令、 線圖，並提供與該篩選出來的對照線形相對應的=的對照線形， 考，讓使用者可快速地找到其所需要的資訊。、，代供使用者參 依據本發明之一特色，提出一種判斷線形相、 以比對一樣本線形和一對照線形以判斷其相似声X度之系統，用 包含有：一第一選擇模組，用以依據該樣本線形之其中，該系統 該樣本線形選擇出一特定數量之特徵點，再依护戶=位貝料，從 點來選擇出複數條特徵線；-第二選擇模組選擇出之特徵 形之數位資料及該第一選擇模組所選擇特徵點及二：：該對照線 數量，從該對照線形中選擇出其對應之對照特徵^、、、、之方式及 選擇出之對照特徵點來選擇出其對應之對照特徵線·w及依據所 =組，用以計算各特徵線及料應之對照特^__第—計算 #寸徵線及其對應之對照躲線的斜率域並取 ’且將各 特徵線與其對應之對照特徵線的差里产·以及一_ —，以作為各 用以將各特徵線與其對應之舰特徵二差二計算模級， 線之權重值，然後進行加總以產生_總差異^=乘上各特徵 ⑽|異仏獲得該樣本線形和 ·該線 ，用以二特色’提*另-種判斷線形相似度之♦ 選m "面模組’提供-操作介面，料 &擇-以_，以及於該樣 ί、ι用者 人進仃比對的該對照線形之數位 枳組’用以將 點及特徵線之方式& 、"义據該使用者所選擇 式及數*，選擇出其對應之對照特徵^擇特徵 β及仿 200813876 =選擇出之對簡徵點㈣擇#對應之對照 计鼻模m計算各特縣《對應之對昭特徵㈣^弟— 為各特徵線與其對應之對照特徵線的差 丨值，以作 包’用以將各特徵線與其對應之對照特徵^差以第二計算模 特徵線之權重值，然後進行加總以產料別乘上各 該線形總差異度叫得難本線形、、日j，將1減去 純徵狀權重___ 度，其中， • «本發明之另-特,，心 值不會超過1。 ，應用在金融商品資1 種判斷線形相似度之系統 他待比對金融商品之對㈣比f—金融商品之樣本線形和其 • -操作介面模組斷其相似度，其中，該系統包 擇-金融商品之樣本線形以及心以供-使用者選 '-選擇模組，用以依據…待比對金融商品之條件；-第 擇出-特定數量之特徵％ ^、㈣之數位資料’從該樣本線形選 數條特徵線；-第二選摆磁:依!所選擇出之特徵 點來選擇出複 • 金融商品之條件，從金s 以依據該使用者所選擇待比對 金融商品及其對照線形㈣商叩貝§fl中尋找出符合該條件之待比對 特徵線之方式及數量，彳〜依據該第—選擇模組所選擇特徵點及 對應之對照特徵點，^4比對金融商品之對照線形中選擇出其 對應之對照特徵線；一μ依據所選擇出之對照特徵點來選擇出其 應之對照特徵線的斜率第计异模組，用以計算各特徵線及其對 斜率相減並取絕對偉，^且將各特徵線及其對應之對照特徵線的 差異度；以及一第二卄三=為各特徵緣與其對應之對照，特徵線的 特徵線的差異度分別乘枳、、且，用以將各特徵線與其對應之對照 ^上各特徵線之權重值，然後進行加總以產 200813876 生線形總差異度，將1減去該線形總差異度以獲得該樣本線形和 對照線形之相似度，其中，各特徵線之權重值將使該線形總差異 度之數值不會超過i。 依據本發明之另一特色，本發明提出一種判斷線形相似度之 方法，用以比對一樣本線形和一對照線形以判斷其相似度，其中 ，該方法包含有以下步驟：依據該樣本線形之數位資料，從該樣 本線形選擇出一特定數量之特徵點，再依據所選擇出之特徵點來 選擇出複數條特徵線；依據該對照線形之數位資料及上述所選擇 特徵點及特徵線之方式及數量，從該對照線形中選擇出其對應之 對知、特徵點，以及依據所選擇出之對照特徵點來選擇出其對應之 對照特徵線；計算各特徵線及其對應之對照特徵線的斜率，且將 各特徵線及其對應之對照特徵線的斜率相減並取絕對值，以作為 各特徵線與其對應之對照特徵線的差異度；収將各特徵線與其 對應之對照特徵線的差異度分縣上各特徵線之權重值，然後進 行加總以產生線職差異度，將丨減去該_總差異度以獲得該 樣本線形和對照線形之相似度，其中，各特徵線之權重值將㈣ 線形總差異度之數值不會超過1。According to the same low level of the corresponding traits of the line shape, and providing the relevant technology, because of this, the case, such as = know the technology, there is no similar system and method, In order to automatically determine the quotation of two linear similarity financial products, the financial transaction market should use 'assisted kinetic energy and market trading materials 1 software to help people judge the overall market financial products and obtain higher prices. Benefits, help investors into the content of the invention] Law, can be m mesh r #... species _ the system of the system and the number of lines:: select a plurality of feature points, complex linear similarity. The purpose of ordering is to use the (4) and method proposed by the present invention to determine the similarity between the sample line shape and the control line shape, which is in line with the human visual perception of similarity, and greatly improves the use. The full scale of the line similarity is judged. The linear shape of high approximation is calculated by the system and method of the present invention, and its appearance will be similar to human visual similarity. Another object of the present invention is to provide a system and method according to the present invention. When it is further applied to other application software, the user only needs to select a certain sample line through a simple operation interface, that is, The plurality of control line shapes can be similarly judged with the sample line shape 200813876, and the line shape corresponding to the sample line is automatically screened, and the control line shape corresponding to the selected control line shape is provided, and the test is used. You can quickly find the information you need. According to one of the features of the present invention, a system for judging a linear phase, comparing the linear shape and a control linear shape to determine a similar sound X degree is provided, and includes: a first selection module According to the sample line shape, the sample line shape selects a specific number of feature points, and then selects a plurality of feature lines from the point according to the user's position; the second selection module selects The digital data of the characteristic shape and the selected feature points of the first selection module and the second: the number of the control lines, and the corresponding control features ^, , , and the manners selected from the control line shape and selected Corresponding to the feature points to select the corresponding control feature line ·w and according to the = group, to calculate the characteristic line and the material should be the control special ^__ first - calculation # inch line and its corresponding control line The slope domain is taken and the difference between each characteristic line and its corresponding control characteristic line is generated and a _ is used as the difference between the two characteristic lines and their corresponding ship features. The weight of the line is calculated. Value, then add up to produce _Total difference ^=Multiply each feature (10)|Different to obtain the sample line shape and the line, for the two features 'to mention the other - to judge the linear similarity ♦ select m " face module 'provide - operation interface , material & select - to _, and the number of the comparison line of the ί, ι user 仃 对 ' ' ' 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以 用以Select the type and number*, select the corresponding control feature and select the characteristic β and imitation 200813876 = select the pair of comparison points (four) select # corresponding to the comparison of the nose model m to calculate the corresponding characteristics of each county (4) ^ brother - for each feature line and its corresponding control feature line difference , value, as a package 'to use each feature line and its corresponding comparison feature ^ difference to the second calculation of the modulus feature line weight value, and then add In general, the total difference of the linearity of the product is called the difficulty of the line shape, and the date j, minus 1 is the pure symptom weight ___ degree, wherein, «The invention is another special, the heart value is not Will exceed 1. In the system of financial commodity capital, a system for judging linear similarity, he waits for the pair of financial products (4) to compare the sample line shape of f-financial goods with its • operation interface module, wherein the system selects - the sample line shape of the financial product and the heart for the user to select the '-selection module, based on the condition of the financial product to be compared; - the first choice - the specific number of features % ^, (4) the digital data 'from The sample line selects a number of characteristic lines; - the second selection of the magnetic field: according to the selected feature points to select the condition of the financial product, from the gold s according to the user's choice to compare the financial goods and The method and quantity of the characteristic line to be compared according to the condition are found in the comparison line (4) quotient §fl, 彳~ according to the feature point selected by the first selection module and the corresponding control feature point, ^4 comparison The corresponding control feature line is selected in the control line of the financial product; a μ selects the slope of the corresponding feature line according to the selected control feature point, and calculates each characteristic line and Subtract the slope and take For Wei, ^ and the difference degree of each feature line and its corresponding control feature line; and a second = = = for each feature edge and its corresponding contrast, the difference of the feature line of the feature line is multiplied by 枳, and , the weight value of each feature line corresponding to each characteristic line is compared, and then summed to produce the total difference of the line shape of 200813876, and the total difference of the line shape is subtracted by 1 to obtain the sample line shape and the control line shape. The similarity, wherein the weight value of each feature line will cause the total value of the line shape to not exceed i. According to another feature of the present invention, the present invention provides a method for judging linear similarity for comparing a linear shape and a control linear shape to determine the similarity thereof, wherein the method includes the following steps: according to the sample linear shape Digital data, selecting a specific number of feature points from the sample line shape, and then selecting a plurality of feature lines according to the selected feature points; according to the digital data of the control line shape and the manner of selecting the feature points and characteristic lines And the number, select the corresponding pair of knowledge points, feature points from the control line shape, and select corresponding control feature lines according to the selected control feature points; calculate each feature line and its corresponding control feature line Slope, and subtracting the slopes of the characteristic lines and their corresponding control feature lines and taking the absolute value as the difference between each feature line and its corresponding control feature line; receiving the feature lines and their corresponding control feature lines The difference degree is divided into the weight values of each characteristic line on the county, and then summed to generate the line job difference degree, and the _ total difference degree is subtracted to obtain Similarity of the sample line and control line, wherein the weight of each characteristic line of total weight value of the value of the difference (iv) not more than one line.
特徼點來選擇複數個Μι 一種判斷線形相似度之方法 對照線形以判斷其相似度，其中，該 一操作介面，以供一使用者選擇一樣Special points to select a plurality of Μι A method for judging the linear similarity. The line shape is compared with the line shape to determine the similarity, wherein the operation interface is selected by a user.
出其對應之對照特徵點， =上選擇蚊數量之特徵點，再依據該 出其對應之對照特徵線； 進行比對的該對照線形之數位 及特徽綠少古4 »去l θCorresponding to the corresponding control feature points, = select the feature points of the number of mosquitoes, and then according to the corresponding control feature lines; compare the digits of the control line and the special emblem green less ancient 4 » go l θ
畔异各特徵線及其對應之對照 照特徵線的 200813876 斜率，且將各特徵線及其對應之對照特徵線的斜率 值，以作為各籠線與其對應之韻特徵線的差取絕對 特徵線與其對應之對照特徵線的差異度分別乘上各及將各 值，然後進行加總以產生線形總差異度，將i減去心、、、之權重 度以獲得該樣本線形和對照線形之相似度，其中，總差異 重值將使該線形總差異度之數值不會超過卜、Q、徵線之權 依據本♦明之另-特色，提出另—種判 ，應用在金融商品資訊上，用以比對一金融商品之:：度之方法 他待比對金融商品之對照線形及判斷其相似度，1中K線形和其 含有以下步驟：提供-操作介面，以供—使用者選^亥方法包 之樣本線形以及其他待比對金融商品之條件；依據2商品 數位資料，從該樣本線形選擇出一特定數 X袭本線形之 選擇出之特徵點來選擇出複數條特徵線;：據:=再依據所 比對金融商品之條件，從金融商品資財尋/者所選擇待 比對金融商品及其對照線形，再依據上述所選擇^ 之待 之方式及數量’從待比對金融商品之對mu’錢#徵線 對照特徵點，以及依據所出之對照特_^=其對應之 對照特徵線，·計算夂特及.、 k擇出其對應之 夂w 4徵線及其對應之㈣特徵線的斜率，且將 口寸徵線及其對應之對照雜線的斜畅減絲絕 各特徵線與其對應之對照特徵線的差異度，·以及將夂 :、、、 對應之對料徵_差異度分別乘上各特徵線之權=徵線與其 行加總以產生線形總差異度，將1減去該線形總差I然f進 線形總差異度之數值不會超過！。 _值將使該 為使貴審查委員能進一步瞭解本發明之架構、特徵及功能 200813876 茲 附以圖式及其具體實施方式詳細說明如後 【實施方式】 本發明之目的、優點和特色由以下實施例之 當可更加明白’該實施例係作為實施方式之舉例說明：：, 為方料加說明本發明，而非仏限定本發明。本㈣ 對松本線开>以-對知線形，以判斷該樣本線形與 形的相似度，且說明書中所提及之樣本線形、對照線 線 在電腦之中依據數位資料所產生的數位線形，包括但不限&糸指The slope of 200813876 of each feature line and its corresponding control feature line, and the slope value of each feature line and its corresponding control feature line is taken as the absolute characteristic line of the difference between each cage line and its corresponding rhyme feature line. The difference between the corresponding control feature lines and the corresponding values are multiplied by the respective values, and then summed to generate the total linear difference, and the weight of the heart, and is subtracted from i to obtain the similarity between the sample line shape and the control line shape. Degree, wherein the total difference weight value will make the total value of the linear shape not exceed the weight of the Bu, Q, and the line. According to the other characteristics of this ♦, another type of judgment is proposed, which is applied to the financial commodity information. In comparison with a financial product:: degree method, he compares the comparison of financial products with the line shape and judges its similarity, 1 K-line and it contains the following steps: provide-operate interface for the user to choose ^hai The sample line shape of the method package and other conditions for comparing the financial products; according to the 2 product digital data, a specific number X is selected from the sample line shape to select a plurality of feature points to select a plurality of special features Line;: According to: = according to the conditions of the comparison of financial products, from the financial goods search / the choice of financial products and their control line shape, and then according to the above-mentioned choice of the way and the number of 'waiting Compare the financial product's pair of mu' money# levy control feature points, and according to the corresponding control feature _^= corresponding control characteristic line, calculate 夂 special and ., k select the corresponding 夂 w 4 sign The slope of the line and its corresponding (4) characteristic line, and the difference between the characteristic line of the mouth line and its corresponding control line and its corresponding control line, and the difference between the line and the corresponding control line, Corresponding to the sign of the material _ difference is multiplied by the weight of each characteristic line = the line is summed with its line to produce the total difference of the line shape, the total difference of the line is subtracted from the total difference of 1 and the value of the total difference of the line shape is not Will exceed! . The value of the present invention will enable the review board to further understand the structure, features and functions of the present invention. 200813876 is attached to the drawings and its detailed description. DETAILED DESCRIPTION [Embodiment] The objects, advantages and features of the present invention are as follows The embodiment is to be understood as an example of the embodiment: the invention is described by way of illustration and not by way of limitation. This (4) is to open the Matsumoto line > to - the line shape to determine the similarity between the line shape and the shape of the sample, and the sample line shape and the control line mentioned in the specification are digital lines generated by the digital data in the computer. , including but not limited to &
> JPG ^ JPE ^ JPEG > T1F ^ TIFF . BMP 位線形，亦可以是其他繪圖軟體所自訂的數位線形。 本發明之實施系統架構示意圖如第一圖所示，主人 一第-選擇模組⑽）、-第二選擇模組⑽）、 · (1〇6)和一第二計算模組（1〇8)，其中，夂 。异杈組 說明則詳細說明如下。 α、、、、 要功能及其特徵 料，模組(1G2)之主要魏，依據該樣本線形之數位資 本線形選擇出_特定數量之特㈣ 的複數個特徵點來選擇出複數 ’中選擇::選擇出 ，叫線形選擇出—特定數量之特徵徵點主 案中讀取該樣本綠报夕料祕〜 攸樣本線形之檔 徵，選擇可料^㈣:料，且依據該樣本線形之視覺特 樣本線形的_、㈣I^特㈣°例如’特徵點可以是 低點.....·）等，而折點（如最高點、最低點、次高點、次 擇該特徵點之數量越7 至少為3以上之整數。-般來說，選 越南，但相對的，其計 — U率度也就 序就自越稷4，所需要的時間也就 200813876 會越久。 該特徵點至少包含有爷栌 以及終點，而該對照特徵點：至:形之起點、N個重要轉折點、 個重要轉折點、以及終點；且夕包含有該對照線形之起點、N 例來說，如第二八圖所：，N為至少1以上之整數。舉 容易地選擇出該樣本線形的特徵點开i，其為簡單的曲線，可以报 的起點、最高點、最低點和終點f包含有心4’亦即該樣本線形 其為較複雜的曲線，可以代 而如第二B圖所示之樣本線形， 亦即該樣本線形的起點、線本線形的特徵點包含有bi〜b9， 當特徵點的數量較多時，將^折點b2〜b8。然而，一般來說， 需的時間會相當久，因此，^致後續的計算非常複雜，計算所 新選擇特徵點為cl〜e5,如第又土之實施方式’可採用簡化方式重 斯弟一 C圖所示。 —-ΜΡ 以攸，亥貝枓中讀取出該線形 可以直接讀取_資料二擇=特： 播格作介面所綠製的圖形及其自行定義的圖 可以二_: 式可供該第一選擇模組_可以讀取，即 了以攸錢據資料中來選擇出狀數量的特徵㉝ 其數量而ΓΓΓΓΓ為連接該複數個特徵點之其中二個特徵點，且 而成之俨徵二*:乂上之整數’但不超過所有特徵點所可以連接 成锊徵、、泉的數S。該特徵線主要是 覺特徵，其數量並不需與特徵狀數 乍為樣本柄之視 特徵線之數量越多，經過本發明計算但同樣的’選擇該 高，但相對的，其計算之程序合越〜目似度其精準度也就越. 越久。 "a啜雜，所需要的時間也就會 200813876 該特徵線至少包含有依序連接 2……N重要轉折點、以及到終點之連=、:形’從起點到第1' 亦至少包含有依序連接該對照線形、從及，該對照特徵線 要轉折點、以及到終點之連接線；且其中弟1、2 ···…N個重 數，且該特徵線最多為連接所有特徵點可:::少u上之整 的數量。 運接而成之特徵線 以第二A圖之樣本線形作為說明 個特徵點al〜a4,並可獲知該四個特徵7、、、形已經選擇了四 值，因此該第-選擇模組⑽)即可依據^二之數據值或座標 線，其選擇的方式和數量可依據實際需㈣^徵點來選擇特徵 線之數量最少為三條，如第二D圖所示可、、疋例如，該特徵 开祕線a2a3和線a3a4, 例如弟二E圖所示，除了上述三條特徵線之外 ^夕為A條， ala3、線ala4和線丑234。再以第二〔圖之样 ^包括有線 本線形已經選擇了五個特徵點el〜e4，° ^線形來說明，該樣 可選擇代表該線形變化的線cl 、：弟一選择模組(_ 線clc52等五條特徵線，如第二F圖所示。 線c4c5和 若是當該線形圖檔無法提供數據資料而只能士 該系統更進-步可具有一操作介面模組⑽，第_圖4科， 用以顯示該線形並提供讓使用麵作滑鼠或鍵盤 未頌不）’ 接在該線形上點選該使用者認為足以代表該線开:的具:二 入該線形之_、終點、特徵點的賴諸或是該線形在座= 的比射料，以取錄為該第—選師組⑽)所選擇特徵^數 據貧料。且再更進-步時，該操作介面模組⑽），亦可以讓 者操作滑鼠或鍵歸輸人卫具，直接在該樣本線形上點㈣複數 13 200813876 個特徵點之其中任二個來設定該特徵線，r 模組（102)所選擇特徵線的數據資料。、、，M取代做為該第一選擇 該第二選擇模組（104)之主要功能，= 料，依據該第-選擇触所選 :據鱗照線形之數位資 從該複數個對照線财選㈣其對應的之方式及數量， 以及依據所選擇出之對照特徵數1之對照特徵點， 例如’該第-選擇模組（1G2) 帛、對應的對照特徵線。> JPG ^ JPE ^ JPEG > T1F ^ TIFF . BMP bit line shape, can also be a custom line shape for other drawing software. The architecture of the implementation system of the present invention is as shown in the first figure, the owner-first selection module (10), the second selection module (10), the (1〇6) and the second computing module (1〇8). ), among them, 夂. The different group descriptions are described in detail below. The function of α, ,,, and its characteristics, the main Wei of the module (1G2), selects a plurality of feature points of the specific number of special (4) according to the digit line shape of the sample line to select the plural 'selection: : Select, call the line shape to select - the specific number of features in the main case to read the sample green note material secret ~ 攸 sample line shape of the sign, select the material ^ (four): material, and according to the sample line shape of the vision The _, (4) I^ special (four) ° of the special sample line, such as 'feature points can be low points ..... ·), etc., and the vertices (such as the highest point, the lowest point, the second highest point, the number of the feature points) The more 7 is at least an integer of 3 or more. In general, Vietnam is chosen, but the relative value of U-rate is from 4 to 4, and the time required for 200813876 will be longer. Contains a grandfather and an end point, and the control feature points: to: the starting point of the shape, the N important turning points, the important turning points, and the ending point; and the evening includes the starting point of the control line, in the case of N, for example, the second eight Figure: N is an integer of at least 1 or more. The feature point of the sample line is i, which is a simple curve. The start point, the highest point, the lowest point, and the end point f that can be reported include the heart 4', that is, the sample line shape is a more complicated curve, which can be replaced by a second The sample line shape shown in Figure B, that is, the starting point of the sample line shape, the feature points of the line-shaped line shape include bi~b9, and when the number of feature points is large, the points b2~b8 will be broken. However, in general, The time required will be quite long. Therefore, the subsequent calculations are very complicated. The newly selected feature points of the calculation are cl~e5. For example, the implementation method of the second soil can be simplified by the method of C. —ΜΡ ΜΡ 攸 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥 亥A selection module _ can be read, that is, the number of features 33 selected by the data in the data is selected as the number of features of the plurality of feature points, and the two feature points are formed. *: the integer on the 'but not more than all feature points can be connected The number S of the 锊 锊 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , High, but relative, the calculation of the program is more than ~ the degree of accuracy is more. The longer. "a noisy, the time required will also be 200813876 The feature line contains at least the sequential connection 2... ...N important turning point, and the connection to the end point =,: the shape 'from the starting point to the 1st' also includes at least the connection of the control line shape, the and the control feature line to the turning point, and the connection line to the end point; Among them, the brothers 1, 2 · · · ... N multiplicity, and the feature line is at most the connection of all feature points can be ::: less than the total number of u. The feature line of the transport is formed by using the sample line shape of the second A picture as the feature points a1 to a4, and it can be known that the four features 7, and the shape have selected four values, so the first selection module (10) ) can be based on the data value or coordinate line of ^2, the way and quantity of selection can be selected according to the actual needs (4) ^ points to select the number of feature lines at least three, as shown in the second D figure, for example, The feature opening line a2a3 and line a3a4, for example, as shown in the second figure E, except for the above three characteristic lines, are A, ala3, line ala4 and line ug. Then, in the second example, the figure includes the five feature points el~e4, ° ^ line shape, and the line cl can be selected to represent the line shape change: the brother selects the module ( _ Line clc52 and other five characteristic lines, as shown in the second F. Line c4c5 and if the line drawing file can not provide data data, but the system can be further advanced - can have an operation interface module (10), the first _ Figure 4, used to display the line shape and provide for the use of the face as a mouse or keyboard.) Connected to the line and click on the line that the user thinks is sufficient to represent the line: the second line into the line_ The end point, the point of the feature point or the specific shot of the line-shaped seat = to be selected as the feature selected by the first-selector group (10). And when further step-by-step, the operation interface module (10) can also allow the user to operate the mouse or the key to return to the human guard, directly on the sample line (4) plural 13 200813876 of any of the feature points To set the data line of the feature line selected by the feature line and the r module (102). And, M, as the first function of the first selection of the second selection module (104), = material, according to the first-selection touch selection: according to the scale of the line of the number of resources from the plurality of control lines Select (4) the corresponding mode and quantity, and the comparison feature points according to the selected control feature number 1, such as 'the first-selection module (1G2) 帛, the corresponding control feature line.
一且一、二:::=第線形= 組004)則可tt照其選擇特徵點及特徵線之 二擇4 形中將選擇五個對照特徵點且亦點、 里線 第一 rFI邮— . 〜、.、、、二個轉折點及終點，如 制線形之對簡徵_擇好之後，該第 Γ=104)則再依據先前樣本線形選擇到特徵線之方式和數 ==擇代表該韻線形的對照特徵線dld2、線伽、線伽、 線d4d5和線dld5等，如第二11圖所示。 該第-計算模組⑽)之主要功能，係計算各特徵線及其對應 之對照特徵線的斜率，且將各特徵線及其對應之對照特徵線的斜 率相減且取絕對值，以作為該特徵線與對照特徵線之差異度。 ^以前述之第二G圖和第二Η圖來舉例說明，該第一計算選擇 模、、且（106)可依據该線clc2、線c2c3、線c3c4、線ad和線“以 等五條特徵線之相關數據，計算出特徵線之斜率分別為Scl〜Sc5， 再依據該線dld2、線d2d3、線d3d4、線d4d5和線dld5等五條對 照特徵線之相關數據，計算出對照特徵線之斜率分別為s d t〜s d 5， 而特徵峰和對照特徵線之斜率相減取絕對值，亦即丨Scl_Sdl丨為代 表著線clc2和線dld2之間的差異度，而丨Sc2-Sd2|為代表著線c2c3 和線d2d3之間的差異度。 14 200813876 該第二計算模組⑽)之主要功能，係用以將各特徵線與其對 應之對照特徵線的差異度分縣上各特徵線之權重值，㈣進行 加總以產生線形總差異度’將i減切線雜差異度以獲得該樣 本線形和對照線形之相似度。其中，各特徵線之權重值將使該線 形總差異度之數值不會超過！，其主要目的，是用以調整各特徵線 之相似度在線形差異度中的比例，由於各特徵線對於線形相似度 影響並不完全相同，例如，特徵線可能會因為線長度較長、呈ς : 員著的斜二、位在特定的位置等因素’因而對整體的視覺效果 較大的影響’而某些特徵線則較不那麼重要，因此，經由、 線之權重值來調整各特徵線在線形差異度中的比例，將w = 之相似度的精準度更高。 ^形 各特徵線之權重值的計算方式，可以有多種實施方式， 定·當考量到各特徵線之線長度會影響各特徵線在線 似度中的重要性時’-般㈣，特徵線越長，其㈣One and one, two:::=first line = group 004) then tt according to its selection feature point and feature line of the second choice 4 will select five control feature points and also point, the first line rFI post - . , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , The rhythm-shaped contrast characteristic line dld2, the line gamma, the line gamma, the line d4d5, and the line dld5 are as shown in the second 11th. The main function of the first calculation module (10) is to calculate the slope of each characteristic line and its corresponding control characteristic line, and subtract the slope of each characteristic line and its corresponding comparison characteristic line and take an absolute value as The difference between the feature line and the control feature line. ^ exemplified by the foregoing second G map and second map, the first calculation selects the mode, and (106) can be based on the line clc2, the line c2c3, the line c3c4, the line ad and the line "to wait for five features According to the relevant data of the line, the slopes of the characteristic lines are calculated as Scl~Sc5, and the slope of the control characteristic line is calculated according to the correlation data of five control characteristic lines such as the line dld2, the line d2d3, the line d3d4, the line d4d5 and the line dld5. It is sdt~sd 5 respectively, and the slopes of the characteristic peak and the control characteristic line are subtracted from each other, that is, 丨Scl_Sdl丨 represents the degree of difference between the line clc2 and the line dld2, and 丨Sc2-Sd2| represents The difference between line c2c3 and line d2d3. 14 200813876 The main function of the second calculation module (10) is to divide the difference between each characteristic line and its corresponding control characteristic line into the weight values of each characteristic line on the county. (4) performing summation to generate a total linear difference degree', i is reduced by the tangent difference degree to obtain the similarity between the sample line shape and the control line shape, wherein the weight value of each characteristic line will make the value of the total line difference of the line not More than! The main purpose is to use The ratio of the similarities of the entire characteristic lines in the linear shape difference degree, because the influence of each characteristic line on the linear similarity is not completely the same, for example, the characteristic line may be long due to the length of the line, and the ς: Factors such as location at a particular location 'and thus a greater impact on the overall visual effect' and some feature lines are less important, therefore, the ratio of the linearity of each feature line is adjusted by the weight value of the line The accuracy of the similarity of w = is higher. ^The calculation method of the weight value of each characteristic line can be implemented in various ways. When considering the line length of each characteristic line, it will affect the linearity of each characteristic line. The importance of 'in general (four), the longer the characteristic line, its (four)
因此可依據各特徵線之長度比例來決定其權重值，例如，H 線之分子，然後將分子除以分權重分;特徵 值。另一種計算方式，亦可將各特徵線相對於整开線=重 別在X'γ方向上之比例來作為權重值 2上’分 Χ、γ方向上之比例的考量，還可將所有特徵線之^方:是欲综合 方向上的長度加總以作為權重分母，而各特徵線之向上和在Υ Υ方向上的長度相加作為各特徵線之分子，然後之1方向上和在 重分母即可獲得各特徵線之權重偉。此外，除以分權 重值之後，還可再依據各特徵線是否具有顯著的余二寸徵線之權 或是平緩）、是否位在特定的位置（如位在中央立4率(如特別陡峭 、位置)等因素，而對 15 200813876 權重值來進行修正調整。 本發明之线既適用於—樣本線形和—對照線形之 度’亦同樣義於觸該樣本線形和複數個對照線形之線 度’只需重覆該第二選擇模組、第—計算模組和第二計算 別對複數個線形的及計算㈣，就可崎得該樣本線形2 對照線形特徵線之間的相似度，並可進而依據該判斷結果，依： 該相似度高低，進行排序及顯示。 μTherefore, the weight value of each characteristic line can be determined according to the ratio of the length of each characteristic line, for example, the numerator of the H line, and then the numerator is divided by the weight division; the eigenvalue. In another calculation method, the ratio of each characteristic line to the entire open line = the difference in the X' γ direction may be taken as the consideration of the ratio of the 'minutes' and the γ directions on the weight value 2, and all the features may be used. The square of the line is the sum of the lengths in the combined direction to be used as the weight denominator, and the length of each characteristic line and the length in the direction of the 相 are added as the numerator of each characteristic line, and then in the 1 direction and in the weight The denominator can obtain the weight of each feature line. In addition, after dividing by the weighted value, it can be based on whether each characteristic line has a significant remaining two-inch line of the right or flat), whether it is in a specific position (such as the position in the central vertical rate (such as particularly steep , position) and other factors, and the weight adjustment value of 15 200813876 is adjusted and adjusted. The line of the invention is applicable to both the sample line shape and the degree of the control line shape, and is also suitable for touching the sample line shape and the line of the plurality of control lines. 'Repetitively repeating the second selection module, the first calculation module and the second calculation for the plurality of linear shapes and calculations (4), the similarity between the sample line shape 2 and the line characteristic line can be obtained, and According to the judgment result, according to the similarity level, the sorting and display are performed.
…當本^明可以與其他的軟體結合應用，例如，制在金融商 =易上蚪例如，目鈾許多的網頁及應用軟體均可將各股票戋 期貨等金融商品之技術線形予以顯示，本發明之系統將可辅: 用者來篩選及判斷數以千計的股票、期貨技術線圖，其線形是否 μ使用者所捕的樣本線料似。本發❹統與金融商品之 f體結合時’由於該應陳體可接收㈣金融商品之數據資料並 猎以產生對應的技術線形，例如κ線圖、KD圖、脱圖等，且 數據資料係可經由本發”辭m " 用以Γ Γ」3該判斷線形相似度系統應用在金融商品資訊上時， 、孟融商品之樣本線形和其他複數個待比對全 =r:r本線形與該對照線形的相似度=商統= :Γ:二同之處主要在於:增加-操作恤^ 以棱供一刼作介面，以供一 八 用 及其他複數個待㈣全糾商品之樣本線形以...when this can be combined with other software applications, for example, in the financial business = easy to upload, for example, many web pages and application software of the uranium can display the technical line of financial products such as stocks and futures. The system of invention will be supplemented by: users to screen and judge thousands of stock and futures technical line graphs, whether the line shape is similar to the sample line captured by the user. When the hair system is combined with the financial product's f body, 'because the person should receive the data of the financial product and hunter to generate the corresponding technical line, such as κ line graph, KD map, off map, etc., and the data The system can use the word "m " for Γ Γ 3 3 该 该 该 该 该 该 该 该 该 该 该 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线 线The similarity between the line shape and the control line shape = quotient = : Γ: The two main points are: increase - operation shirt ^ 棱 for one interface, for one-eight use and other multiple (four) full correction products Sample line shape
于至嘁商口口之條件；以及，該 W 該樣本線形選擇出—特定數量之線 r選，可依據該使用者所選擇待比對金融; ，再依據該第-選擇模cr待比對金融商品及其對照線形 、、、所忠擇特徵點及特徵線之方式及數量，The condition of the mouth of the merchant; and, the selection of the sample line--the specific number of lines r can be selected according to the user's choice to be compared to the financial; and then according to the first-choice mode cr to be compared The manner and quantity of financial products and their control linear shapes, and the selected feature points and characteristic lines,
A 16 200813876 從所尋找出之對照_，再㈣選擇出其對應 特徵點及對應的對轉徵線。 、讀ϊ之對照 此外’該操作介面可經由提 用者來選擇金融商品’以供使 未减示）來儲存❹相聊金圖中 ，以供該使用、， 门^及其相關資料 者下夂選擇樣本線形時可以直接選用。n 還可經由提供多“用L乍介面 的金融商品的種以供使用者來設定其所偏好 特定金融商品等二指標、以及事先所儲存之 、 等以作為待比對金融商品之條件。 品對二二計算模組還可將該複數個待筛選金融商 尺心之相似度的判斷結果，經由一顯介 巇陶 度高低來進行换… 依照該相似 其中—全^序及料，且提供該制者從其_結果中選擇 供給使用2::然後將該金融商品相對應的其他相關 以比為本發明所提出第—種麟線形相似度之方法，用 ,.^ κ本線形和一對照線形以判斷其相似度，复由 包含有以下步驟： ，、中，該方法 :驟S302,依據該樣本線形之數位資料， 出一特定數量之p n樣本線形選擇 條特徵線； 再依據闕擇出之特徵點來選擇出複數 及特Hs=，rr對照線形之數位㈣及上14所選擇特徵點 徵點從㈣照線形中選擇㈣心之對_ 徵線；㈣所選擇出之對照特徵點來選擇出其對應之對= ^驟S3〇6,計算各特徵線及其對應之對照特 、瓦斜率，曰A 16 200813876 From the comparison _ sought, and then (4) select its corresponding feature points and corresponding pairs of transition lines. In addition, the reading interface can be stored in the 聊 聊 金 , , ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' 该 该 该 该 该 该 该 该 该 该可以 You can choose directly when selecting the sample line shape. n can also be used as a condition for the financial products to be compared by providing a plurality of "financial products using the L" interface for the user to set two indicators such as the specific financial products preferred by the user, and pre-stored. The second and second calculation modules can also change the judgment result of the similarity of the plurality of financial quotients to be screened through a display 巇 巇 高 依照 依照 依照 依照 依照 依照 依照 依照Providing the method that the producer selects the supply use 2:: and then the financial product corresponding to the other method according to the first lining-like similarity proposed by the present invention, using the .^ κ line shape and a control line shape to determine the similarity, the complex includes the following steps: , , , the method: step S302, according to the digital data of the sample line shape, a specific number of pn sample line shape selection bar feature line; Select the characteristic points to select the complex number and the special Hs=, rr control the linear digits (4) and the top 14 selected feature points. Select (4) the line of the line (4) the pair of hearts _ the line; (4) the selected control features Point to choose Which corresponds to the step S3〇6 = ^, and the control was calculated for each characteristic line corresponding to the Patent, the slope of the tile, said
17 200813876 2特徵線及其對應之_錄_料相減絲絕對值 ”、、口特徵線與其對應之對照特徵線的差異度；以及 步驟讀，將各特徵線與其對應之對照特徵線的差 “上各特徵線之權重值，然後進行加總以產、又刀別 :去:線形總差異度以獲得該樣本線形和對照線形2^，將复1 ，'特徵線之權重值將使該線形總差異度之數值不會超二。、 該步驟S302中，該特徵點係從樣本 。 整數轉折點專之中來進行選擇，且該特定數量係為至少3以上= 少口亥步驟S302中’從該樣本線形選擇出一特旦士/ ，係從樣本線形之相關數據中 、里之特徵點 據該樣本線形之視覺特徵m:之數據資料，且依 …，該特少包含;徵點 乂及終點’而該對照特徵點 重要轉折 ，重要轉折™…其二有:= 個特徵點該:::線係，接該複數個特徵點中任二 不超過所t Λ ^之數量係為至少3以上之整數 該= 特徵點所可以連接而成之特徵線的數量。 依據該樣本線形之损誓脖 出斗寸 線形之特徵線；以及，該特;^心選擇可以表示出該樣本 中、從起點到第i、2 .·....N個重要:包含有依序連接該樣本線形 以及，該對照特徵線亦 1點、以及到終點之直線； 點到第h·.....N個重要轉^有依序連接該對照線形中、從起 衿折點、以及到終點之直線丨且其中，17 200813876 2 feature line and its corresponding _ record _ material phase minus wire absolute value", the difference between the mouth feature line and its corresponding control feature line; and step reading, the difference between each feature line and its corresponding control feature line "The weight value of each feature line, then add up to produce, and cut the knife: go: the total difference of the line shape to obtain the sample line shape and the control line shape 2^, will be 1 , 'the weight value of the feature line will make this The total difference in linearity will not exceed two. In step S302, the feature point is from the sample. The integer turning point is selected to be selected, and the specific number is at least 3 or more = less than the mouth step S302, 'selecting a special dan from the sample line shape, and selecting the data from the sample line shape According to the visual characteristics of the sample line shape m: the data, and according to ..., the special contains; the point 乂 and the end point 'and the important points of the control feature point, the important transition TM ... the second: = feature points ::: Line system, the number of any two of the plurality of feature points not exceeding t Λ ^ is an integer of at least 3 or more = the number of feature lines that the feature points can be connected. According to the line shape of the sample, the characteristic line of the stroke line shape; and, the special selection; can be expressed in the sample, from the starting point to the i, 2, ..... N important: including The sample line shape is sequentially connected, and the control feature line is also 1 point, and a straight line to the end point; the point to the h·.....N important turns ^ are sequentially connected to the control line shape, and are folded from the same Point, and the straight line to the end point, and
18 200813876 N為至少m之魏’且職徵線最多為連接所有特 以連接而成之特徵線的數量。 - 該步驟S306中，各特徵線之權重值係用以調整各特徵線之 差異度在整體線形之總差異度中的比例，其係、依據各特徵線對於 該線形相似度之影響程度來決定；且各特徵線係因線長度較長、 具有顯著的斜率、位在特定的位置等因素，而調整該各特徵線之 權重值。 該第一種判斷線形相似度之方法中，更進一步時，更可用以 瞻判斷該樣本線形和複數個對照線形之線形相似度；且其中，該方 法更包含有以下步驟：分別依據每一個對照線形之數位資料，來 選擇該複數個對照線形之對照特徵點及其對照特徵線；分別計算 每一個對照線形之對照特徵線的斜率，且針對每一個對照線形， 將各特徵線及其對應之對照特徵線的斜率相減並取絕對值，以計 异該樣本線形分別與該複數個對照線形之間，各特徵線與其對應 之對照特徵線的差異度；以及分別將該樣本線形與每一個對照線 形之間，將各特徵線與其對應之對照特徵線的差異度分別乘上各 ® 特徵線之權重值，然後進行加總以產生線形總差異度，將1減去 該線形總差異度以獲得該樣本線形和每一個對照線形之相似度， 其中’各特徵線之權重值將使該線形總差異度之數值不會超過1 〇 第四圖係為本發明所提出第二種判斷線形相似度之方法，用 以比對一樣本線形和一對照線形以判斷其相似度，其中，該方法 包含有以下步驟： r ί 步驟S402，提供一操作介面，以供一使用者選擇一樣本線形 以及於該樣本線形上選擇特定數量之特徵點，再依據該特徵點 19 200813876 來選擇複數個特徵線· 步驟S404，蔣外、知> ^ 使用者所選擇人m靖照線形之數位資料，依據該 昭特：Γ:及特徵線之方式及數量，選擇出其對應之對 照特徵線； 、擇 卞，、、為徵點來選擇出其對應之對 步驟S406,計算各特徵線及其對應之 將各特徵歧料紅對照特徵 ^禮_斜率，且 為夂胪Μ ^ Λ 4辜相減並取絕對值，以作 :、、、口寸徵線與其對應之對照特徵線的差異声·以及 fy，將各特徵線與其對應之對H線的差異度分別 2徵線之權重值，然後進行加總以產生線形總差異度，將】 =去该線縣差異度以獲得雜本_和對照線形之相似度，盆 ，各特徵線之《㈣使該_總差^之數料會超過】 5亥步驟S402中，該操作介面孫担y 該使用者«該樣本線形，以及圖形使用者介面以供 來選擇該魏點和特徵線。^ ^工具以供制者操作 該第二種判斷線形相似度之方法中，更進—步時，更可用以 判斷該樣本線形和減㈣照線形之線狗目似度；且其中’該方 2包含步驟：㈣依據每—個對照線形之數位資料，來 讀該複數個對照線形之對照特徵點及其對照特徵線；分別計算 ^個對照線形之對照特徵_斜率，且針對每—個對照線形， ，各特徵線及其對應之對簡徵線的斜率相減並取絕對值，” 异該樣本線形分別與該複數個對照線形之間，各特徵線與並對岸 之對照特徵_差異度；以及分㈣該樣本線形與每—個對㈣ 形之間，將各特徵線與其對應之對照特徵_差異度分別乘上夂 特徵線之權重值，，然後進行加㈣產生線形總差異度，將t 20 200813876 該線形總差異度以獲得該樣 、 其中，各特徵線之權重值將::如母一個對照線形之相似度， 。 值將使該線形·異度之數值不會超過1 用在全\Γ，為本《明所提出第三種绩線形她度之方去a 用在金箱品賢訊上，用以比對 又之方法，應 比對金融商品之對照線形及判_相=二t和其他待 以下步驟： 又，、中，该方法包含有 步驟S502,提供一操作 之樣本線料及其他待轉麵商麵择—金融商品 步驟S504，依據該樣本 二， 出-特定數量之特徵點，再/ 位貝料’從該樣本線形選擇 條特徵線； 、細選擇出之特徵點來選擇出複數 步驟S506，依據該使用 金融商品資訊中尋找出符人^畢待比對金融商品之條件，從 形’再依據上述所選擇特;二錢商品及其對照線 金融商品之_線形巾選 +、、之方式及數量，從待比對 選擇出之對照特徵點來選擇出其對以及依據所 將各=;，：^^:之對_線的斜率，且 為各特徵線與其對應之對照特徵線目減亚取絕對值，以作 步驟S5H)，將各特徵線與 ^，以及 乘上各特徵線之權重值〜龍特徵線的差異度分別 減去該線形總差異度以獲得 H、4 W產生線形總差異度，將1 中，各特徵線之權重值將使該之t夜，其 21 200813876 項，以供使用者來選擇金融商品之樣本線形；以及，經由提供多 種選擇條件之選項，以供使用者來設定其所欲進行比對之金融商 品的種類、價位、是否為核心指標、以及事先所儲存之特定金融 商品等，以作為待比對金融商品之條件。。 該第二種判斷線形相似度之方法中，更進一步時，更可儲存 使用者所選擇金融商品之樣本線形及其相關資料，以供該使用者 下次選擇樣本線形時可以直接選用；以及，將該複數個待比對金 融商品對照線形之相似度的判斷結果，經由一顯示介面，依照該 相似度高低來進行排序及顯示，且其中，該顯示介面更包含用以 提供該使用者從其判斷結果中選擇其中一金融商品時，可將該金 融商品相對應的其他相關資訊，提供給使用者參考。 經由本發明的判斷線形相似度之系統及方法，可經由從樣本 線形和對照線形中選擇複數個特徵點、計算複數條特徵線之斜率 等技術步驟，來判斷出該樣本線形和對照線形之線形相似度，其 符合人類視覺對相似程度的觀感，大幅提高了使用者對判斷線形 相似度的滿意度，且可更進一步應用在其他的應用軟體上，以幫 助使用者自動篩選出與該樣本線形相近似的對照線形，提供與該 篩選出來的對照線形相對應的資訊供使用者參考，實為以往習知 技術中所無法達到的功能。 綜上所述，藉由本發明所提出的系統及方法，具有習知技術 中所缺乏的優點，無論就目的、手段及功效，在在均顯示其優於 習知技術，懇請貴審查委員明察，早曰賜准專利，俾嘉惠社會 ，實感德像。惟應注意的是，上述實施例僅為了便於說明之舉例 而已，本發明所主張之權利範圍自應以申請專利範圍所述為準， 而非僅限於上述實施例，大凡依據本發明所為之各種修飾與變化 22 200813876 ，或將本發明之各項模組或步驟進行整合或分割，均仍應包含於 本案之申請專利範圍内。 【圖式簡單說明】 第一圖：本發明系統之架構圖。 第二A〜Η圖：本發明舉例說明之各種線形圖。 第三圖 :本發明第一實施方法之流程圖。 第四圖 :本發明第二實施方法之流程圖。 m 第五圖 :本發明第三實施方法之流程圖。 【主要元件符號說明】 10 本發明系統 102 第一選擇模組 104 第二選擇模組 106 第一計算模組 108 第二計算模組18 200813876 N is at least m's and the number of lines is the maximum number of characteristic lines connecting all the special connections. - In step S306, the weight value of each characteristic line is used to adjust the proportion of the difference degree of each characteristic line in the total difference degree of the overall line shape, which is determined according to the influence degree of each characteristic line on the linear similarity degree. And each characteristic line adjusts the weight value of each characteristic line due to factors such as a long line length, a significant slope, and a specific position. In the first method for judging the linear similarity degree, further, the linear similarity between the sample linear shape and the plurality of control linear shapes may be further determined; and wherein the method further comprises the following steps: respectively according to each comparison The digital data of the line is selected to select the control feature points of the plurality of control lines and the control feature lines thereof; the slopes of the control feature lines of each of the control lines are respectively calculated, and for each of the control lines, each feature line and its corresponding Comparing the slopes of the characteristic lines and taking the absolute values to calculate the difference between the line shape of the sample and the plurality of control lines, the difference between each characteristic line and its corresponding control line; and respectively shaping the sample line with each Between the control line shapes, the difference between each feature line and its corresponding control feature line is multiplied by the weight value of each of the feature lines, and then summed to generate a total line difference, and 1 is subtracted from the total line difference. Obtaining the similarity between the sample line shape and each of the control line shapes, where 'the weight value of each feature line will make the total difference degree of the line shape The value is not more than 1 〇 The fourth figure is the second method for judging the linear similarity of the present invention, which is used to compare the linear shape and a control linear shape to determine the similarity. The method includes the following steps. Step s402, providing an operation interface for a user to select the same line shape and selecting a specific number of feature points on the sample line shape, and then selecting a plurality of feature lines according to the feature point 19 200813876, step S404, Jiang Wai, Zhi > ^ The digital data of the user selected by the user, according to the method and quantity of the Zhaot: and the characteristic line, select the corresponding control feature line; For the locating point, the corresponding step S406 is selected, and each characteristic line and its corresponding characteristic viscous red comparison feature 斜率 _ slope is calculated, and 夂胪Μ ^ Λ 4 辜 is subtracted and taken as an absolute value, For the difference between the sound and the corresponding characteristic line of the mouth line and the corresponding characteristic line, and fy, the difference between each characteristic line and its corresponding pair of H lines is the weight value of the 2 line, and then summed to generate Total line difference Degree, will] = go to the line county difference to obtain the similarity of the hybrid _ and the control line shape, the basin, the characteristic line "(4) make the number of the _ total difference ^ will exceed] 5 Hai step S402, the The operation interface Sun Dan y the user « the sample line shape, and the graphical user interface to select the Wei point and the feature line. ^ ^ The tool is used by the controller to operate the second method for judging the linear similarity. When the step is further advanced, the tool can be used to determine the line shape of the sample and the line of the line (4); and wherein the party 2 Including the steps: (4) reading the control feature points of the plurality of control lines and their control feature lines according to the digital data of each control line; respectively calculating the control features _ slope of the control lines, and for each control The linear shape, the characteristic lines and their corresponding slopes of the simplification line are subtracted and take the absolute value," the difference between the characteristic line and the pair of control lines, and the characteristic characteristics of the characteristic line and the opposite shore And sub-(4) between the sample line shape and each pair (four) shape, multiply each feature line and its corresponding comparison feature _ difference degree by the weight value of the 夂 characteristic line, and then add (4) to generate a total line difference, Will t 20 200813876 the total difference of the line shape to obtain the sample, wherein the weight value of each feature line will be: the similarity of the parent line to the line shape, the value will make the value of the line shape and the difference degree not exceed 1 In full \Γ, for the third line of the Ming Dynasty, the third line of performance is to be used in the gold box product, for comparison, the method should be compared with the financial product and the line shape and judgment _ phase = The second step and the other steps are as follows: In addition, the method includes the step S502, providing an operation sample line material and other to-be-transformed face-financial product step S504, according to the sample two, out-specific quantity The feature point, the re-bit material 'selects the feature line from the sample line shape; selects the feature point to select the complex step S506, and finds the person to be compared according to the financial product information. The conditions are selected from the shape of 'the other ones according to the above; the two money products and their control line financial products _ linear towel selection +, the way and quantity, select the corresponding feature points from the comparison to select the right And according to the slope of the pair of _ lines of each of the =;,:^^:, and the absolute value of each feature line and its corresponding control feature line minus the sub-absence, as step S5H), the feature lines and ^, And multiply the weight of each characteristic line ~ the difference of the dragon characteristic line The degree of difference is subtracted from the total difference of the line to obtain the total difference of the line shape of H and 4 W. In the 1st, the weight value of each feature line will make the t night, and the 21 200813876 item is selected by the user. a sample of financial products; and, by providing options for a variety of selection conditions, for the user to set the type, price, financial indicators, and specific financial products stored in advance, etc. In order to compare the condition of the financial product, the second method for judging the linear similarity may further store the sample line shape of the selected financial product of the user and related materials for the user. The next time the sample line shape is selected, the result can be directly selected; and the determination result of the plurality of similarity comparisons of the financial products against the line shape is sorted and displayed according to the similarity level through a display interface, and wherein The display interface further includes other information corresponding to the financial product when the user selects one of the financial products from the judgment result thereof. Off information provided to the user's reference. Through the system and method for determining the linear similarity of the present invention, the linear shape of the sample line shape and the control line shape can be determined by selecting a plurality of feature points from the sample line shape and the control line shape, calculating the slope of the plurality of feature lines, and the like. Similarity, which conforms to the human visual perception of similarity, greatly improves the user's satisfaction with judging the linear similarity, and can be further applied to other application software to help the user automatically screen out the sample and the sample. The similar control line shape provides information corresponding to the selected control line shape for the user's reference, which is a function that cannot be achieved in the prior art. In summary, the system and method proposed by the present invention have the advantages that are lacking in the prior art, and are superior to the prior art in terms of purpose, means, and efficacy. As early as the grant of patents, 俾 Jiahui society, real sense of virtue. It should be noted that the above-mentioned embodiments are merely examples for convenience of description, and the scope of the claims should be based on the scope of the patent application, and not limited to the above embodiments, and various Modifications and variations 22 200813876, or the integration or division of the various modules or steps of the present invention, should still be included in the scope of the patent application of the present application. [Simple description of the diagram] First figure: The architecture diagram of the system of the present invention. Second A~Η: Various line graphs exemplified by the present invention. Third Figure: Flow chart of the first embodiment of the present invention. Fourth Figure: Flow chart of a second embodiment of the present invention. m Fifth Figure: Flow chart of the third embodiment of the present invention. [Main component symbol description] 10 The present invention system 102 First selection module 104 Second selection module 106 First calculation module 108 Second calculation module
Claims (1)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW95132532A TW200813876A (en) | 2006-09-01 | 2006-09-01 | System and method for determining similarity of curves |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW95132532A TW200813876A (en) | 2006-09-01 | 2006-09-01 | System and method for determining similarity of curves |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TW200813876A true TW200813876A (en) | 2008-03-16 |
Family
ID=44768431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW95132532A TW200813876A (en) | 2006-09-01 | 2006-09-01 | System and method for determining similarity of curves |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TW200813876A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103729546A (en) * | 2013-12-06 | 2014-04-16 | 杭州安恒信息技术有限公司 | Method for contrastively analyzing level of similarity through broken line deviation variance accumulation |
| TWI790466B (en) * | 2020-08-05 | 2023-01-21 | 國立高雄科技大學 | Graphical investment decision-making system based on price pattern of securities |
| CN118798658A (en) * | 2024-09-11 | 2024-10-18 | 奥德科技有限公司 | Gas user behavior prediction method based on big data analysis |
-
2006
- 2006-09-01 TW TW95132532A patent/TW200813876A/en unknown
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103729546A (en) * | 2013-12-06 | 2014-04-16 | 杭州安恒信息技术有限公司 | Method for contrastively analyzing level of similarity through broken line deviation variance accumulation |
| CN103729546B (en) * | 2013-12-06 | 2017-02-22 | 杭州安恒信息技术有限公司 | Method for contrastively analyzing level of similarity through broken line deviation variance accumulation |
| TWI790466B (en) * | 2020-08-05 | 2023-01-21 | 國立高雄科技大學 | Graphical investment decision-making system based on price pattern of securities |
| CN118798658A (en) * | 2024-09-11 | 2024-10-18 | 奥德科技有限公司 | Gas user behavior prediction method based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063945B (en) | A method for constructing 360-degree customer portraits of electricity sales companies based on value evaluation system | |
| US8099346B2 (en) | Reference price framework | |
| CN107133865B (en) | A method and device for obtaining credit score and outputting feature vector value | |
| EP2074564A2 (en) | Diamond valuation method, apparatus and computer readable medium product | |
| CN101308560A (en) | Store management system and program | |
| Bu et al. | Estimating option implied risk‐neutral densities using spline and hypergeometric functions | |
| JP2018169986A (en) | Enterprise information providing system and program | |
| Hwang | Forecasting credit ratings with the varying-coefficient model | |
| JP2023018101A (en) | Learning device, machine learning model and learning method | |
| Neuhierl et al. | Data snooping and market-timing rule performance | |
| Chao et al. | Risk‐minimizing approach to bid‐cutting limit determination | |
| US7689490B2 (en) | Matching resources of a securities research department to accounts of the department | |
| CN110574065A (en) | Forecast management system and method | |
| AU2013101756A4 (en) | Method and system of optimizing a marketing campaign of a saleable commodity | |
| TW200813876A (en) | System and method for determining similarity of curves | |
| JP2004046454A (en) | Stock investment rating evaluation method and system, program for making computer execute the method, and storage medium with the program stored therein | |
| AU2010200243B2 (en) | Personal point of sale commission rate benchmarking tool | |
| Forster et al. | Artwork characteristics and prices in the New Zealand secondary art market, 1988–2011 | |
| US20230342792A1 (en) | Systems and Methods for Determining Entity Characteristics | |
| TW201013572A (en) | System and method for providing financial products with comprehensive evaluation data | |
| JP5542035B2 (en) | Company information management system, company information management method, company information management program | |
| JPWO2022259512A5 (en) | ||
| CN119558941B (en) | Commodity module ordering method and device, electronic equipment and storage medium | |
| Sharma et al. | Customer criterion for selecting a bank: an exploratory factor Analysis from an Indian perspective | |
| CN111210245B (en) | Evaluation method and evaluation device for insurance product, intelligent system and storage device |