M410287 五、新型說明： 【新型所屬之技術領域】 [0001] 本創作是有關於一種自動影像比對系統，特別是有關於 一種可自動、半自動及手動之方式辨識動態或靜態影像 之自動影像比對系統。 【先前技術】M410287 V. New Description: [New Technology Field] [0001] This creation is about an automatic image comparison system, especially for an automatic image ratio that can automatically, semi-automatically and manually identify dynamic or static images. For the system. [Prior Art]
[0002] 隨著科技時代的來臨，智慧型相關系統的研究也隨著電 腦而蓬勃發展，其應用已在電腦的領域發揮得淋漓盡致 ，可說是有電腦存在的地方，就有智慧型相關系統的存 在；無論何處需要自動化系統，何處就在應用人工智慧 如-> ·/ - v 1 的理論、方法和技術。而影政處熬#智慧型系統 一項重要的環節，尤其是圖；¥辨鐵S目'關妁，應，角如人臉辨 識、指紋辨識及車牌辨識等，皆需要影像處理技術才可 完成，使得近年來，越來越多人投身於數位影像的相關 研究，讓影像處理技術領域有.了肤速的發展。例如在指 紋辨識方面，為了讓指紋紋路更;Η碉顯就必須做影像前 處理、二值化及細線化等箏，Ά可使細部特徵抽取更加 明顯，辨識率才會相對提升。 [0003] 目前，大部份的影像比對辨識系統，都僅能對靜態影像（ 如照片、圖片等）以人工或半自動的方式進行辨識比對動 作，而無法對動態影像（影片）自動解析以進行辨識比對 。且目前的系統在辨識完影像後，並無法自動的分辨其 辨識比對結果以加以分類儲存，使得辨識完後，僅能較 快速的找到比對成功之影像，倘若使用者想要再進一步 的復查未辨識成功或未比對成功之影像時，在尋找檔案 表單編號Α0101 第3頁/共22頁 影像時，就必須先花費一段時間了。 [0004] 因此，本創作便設立一套自動影像辨識系統，使得動態 影像可自動且快速地被辨識比對，並且在辨識比對後， 系統會自動將其分類，使得要重新復查比對影像時，可 快速的找尋到檔案。 【新型内容】 [0005] 有鑑於上述習知技藝之問題，本創作之其中一目的就是 在提供一種自動影像比對系統，以解決習知技術無法自 動辨識動態影像，以及辨識比對完影像後無法自動的分 類儲存之問題。 [0006] 根據本創作之另一目的，提出一種自0½像比對系統， 、 .〆 -e., ·-' <·" 其包含一比對資料庫、一接收模組、一影像處理模組、 一辨識模組、一比對模組及一提示模組。比對資料庫係 儲存至少一比對影像。接收模組係選擇性地開啟以接收 一影像資料。影像處理模組係連接接收模組，以對影像 資料進行影像處理。辨識模組係連接影像處理模組，以 辨識影像資料所要進行比對之部份。比對模組係連接辨 識模組及比對資料庫，若辨識模組對影像資料辨識成功 ，比對模組則根據已辨識之影像資料與各比對影像進行 比對，且判別已辨識之影像資料是否對應其中一比對影 像。提示模組係連接比對模組，若已辨識之影像資料對 應其中一比對影像時，提示模組則發出提示信號。 [0007] 其中，本創作所述之自動影像比對系統，更包含一區域 限定模組，係連接辨識模組，以提供辨識模組對影像資 表單編號A0101 第4頁/共22頁 料之一限定區域進行辨識處理，以提高辨識率。 [0008] 其中，比對影像係為一靜態影像。 [0009] 其中，影像資料係為一動態影像，動態影像係由複數個 靜態影像組成而成。辨識模組係對動態影像之各靜態影 像進行辨識動作，且比對模組係對動態影像中已辨識出 之靜態影像，分別與各比對影像進行比對。本創作之自 動影像比對系統更包含一儲存模組，係連接辨識模組及 比對模組，儲存模組係包含複數個資料夾，以分別儲存 已辨識成功、未辨識成功及比對成功之動態影像中之靜 態影像。 . >>· -·< · ν' [0010] 其中本創作所述之自動影像比1對’系遽m‘更包:含一儲存模 組，係包含至少一資料夾，α德複彳ϊ#-態影像。且 自動影像比對系統更包含一選取模組，係連接儲存模組 及影像處理模組，以提供一使用者選取其中一資料夾後 ，經由影像處理模組對所遠踉之赛科:爽芝4數個靜態影 像進行影像處理。儲存模更φ接模組及比對模組 /" » ^ t ，且儲存模組更包含一已旖嘁資、未辦識資料夾及 一比對成功資料夾，以分別儲存已辨識成功、未辨識成 功及比對成功之靜態影像。其中，選取模組亦可提供使 用者選取其中一資料夾之其中一靜態影像，及提供使用 者選取比對資料庫之其中一比對影像，以進行人工比對 〇 [0011] 其中，影像資料係為一靜態影像，且本創作所述之自動 影像比對系統，更包含一儲存模組，係連接辨識模組及 表單編號A0101 第5頁/共22頁 M410287 比對模組，以將靜態影像儲存於一已辨識資料夾、未辦 識資料夾或一比對成功資料夾中。 [0012] 其中，辨識模組辨識影像資料所要進行比對之部份，係 為文字、數字、圖形或其組合。 [0013] 其中，接收模組係為一攝像裝置，以擷取影像資料。 [0014] 其中，影像處理模組對影像資料進行影像處理之動作包 含二值化、滤波或背景去除。 [0015] 其中，提示模組係以燈號、聲音或其組合之方式發出提 示信號。 [0016] 其中，本創作所述之自動影像比對系統，更包含一傳輸 模組，以提供一遠端操作端^結i：遠端控:制自動影像比 對系統。 [0017] 承上所述，因依本創作之自動影像比對系統，其可具有 一或多個下述優點： [0018] (1)此自動影像比對系統可自動的辨識比對所擷取到之 動態影像，意即攝影鏡頭在做連續攝取動作時，便可同 時同步的進行辨別比對動作，而不再需要將動態影像儲 存為多個靜態影像時再——的進行人工辨識。 [0019] (2)此自動影像比對系統在對影像進行完辨識比對動作 之後，系統便會自動的將已辨識成功、未辨識成功及比 對成功之影像分別的儲存起來，使得使用者欲進行復查 影像時，可快速的找尋到檔案。 [0020] (3)此自動影像比對系統更結合了網路連結之功能，使 表單編號A0101 第6頁/共22頁 得使用者可經由遠端操控此系統。 [0021] (4)此自動影像比對系統除了提供了自動辨識比對影像 功能外，更提供有半自動及人工辨識靜態影像之功能， 讓使用者可進一步的確認辨識比對影像之結果。 【實施方式】 [0022] 以下將參照相關圖式，說明依本創作之自動影像比對系 統之實施例，為使便於理解，下述實施例中之相同元件 係以相同之符號標示來說明。 [0023] 請參閱第1圖，其係為本創作之自動影像比對系統之第一 實施例之方塊圖。第1圖中具帝f自動:影象比對系統100 及一遠端操作端200。自動影::像比對*‘系^統10备:包含比對資 料庫101、接收模組102、影;橡處發模紐?1^3、辨識模組 104、比對模組105、提示模組106、區域限定模組107、 選取模組108、儲存模組109及傳輸模組110。自動影像 比對系統100中之各模組之连接湖孤.，測〔如圖式所示。 [0024] 第一實施例中，本創作之自-動；f像比對系統100，主要可 分為自動動態影像辨識比對、半自動靜態影像辨識比對 以及手動靜態影像辨識比對三種辨識比對模式。 [0025] 自動動態影像辨識比對模式：自動動態影像辨識比對之 方式，係為接收模組102(攝像裝置）接收擷取到動態影像 時，影像處理模組103便會開始對動態影像進行二值化、 濾波或是去除背景等影像處理動作，此動作係讓辨識模 組104可更容易對影像進行辨識。影像處理動作完成後， 辨識模組104便會開始對動態影像進行辨識動作，以擷取 表單編號A0101 第7頁/共22頁 M410287 出影像中將進行比對之部份（可為數字、文字 '圖片戈其 組合）。細識模組1G4對影像成功的操取辨識出來 ，便會經由比對模組105將影像來與比對資料庫1〇1中<所 儲存的各㈣影像進行比對。若比對後，影像對應其^ 一比對影像時，提示模組106便會以燈號、聲音或其組八 之方式發出提示信號。其中，動態影像可視為由一連举。 快速跑動的靜態影像組合而成之料，而影像處理模組 1〇3、辨識模組m及比對模組105，便是針對動態影像 每個跑過的㈣影像畫面進行影像處理、_及比對。 因此’在接收模組1G2不斷的掏取動態影像時系統便已 不斷的在進行影像處理、_及比對之轉（擷取動態影 像之動作仍是連續不中斷的>，兩並_動態影像先--儲存成靜態影像畫面時才進行辨遍比對動作。 [0026] 上述中，動㈣像中若有無法辨識之靜態影像時，比對 模組m便無純對此張靜態影像畫m減會將此 張靜態影像畫面儲存於儲存模組1G9中的未辨識資料炎（ 資料夾名稱可由使用者自訂）中。A 4 而右動態影像中的靜態 影像被成功辨識時，除了再繼續後換 只设續之比對動作外，系 統亦會同時將此張靜態影像畫面鍅 辟存於儲存模組109中的 已辨識資料夾中。而經由比對根知 $4105成功比對之影像， 便會儲存於儲存模組109中的比對由丄 τ成功資料夾中。此分類 儲存的動作，是為了方便當使用者想要復查重新比對影 像時，可讓使用者可快速的找到影像檔。上述儲存的動 Γ皆是將動態影像—解析為靜態影像再加以分類儲 存》 表單編號Α0101 第8頁/共22頁 [0027] 半自動靜態影像辨識比對模式：半自動靜態影像辨識比 對之方々 〆 : 係先將複數個靜態影像儲存在儲存模组109中 :I料夹中’儲存模組109另外更有未辨識資料夾、已 /貝料失及比對成功資料夾。使用者使时自動靜態 辨識比對模式時，可先藉由選取模組H點選健存模 虎中餘存有靜態影像之其中之-個資料夾。接著影像 處理模組103、辨識模組1〇4及比對模組1〇5，便會針對 被點選資料夾_所有靜Μ像--的進行影像處理、 δ ,及比對動作（與比對資料庫1〇1内的各比對影像進行 中對^比對成功時，提示模組變會發出提示訊息。其 半自動靜態影像辨識比響^ 都會將無法華，會比對成功 之靜態影像’㈣儲存於鑛義識資料夹 及比對成功資料夾中。 [0028] [0029] 手動靜態料韻比龍《：手動靜㈣像辨識比對方 式:則是使用者藉由選取賴組1〇9之任一 資料夹中’讀取一張靜態影,V學身選取比對資料庫 1中的比對影像，以進行人工手動辨識比對。 在第-實_巾’ 11域限錢組107的㈣，是用來對影 像進賴域限定之用，如此，辨識模組m便會只針㈣ 像中限&區域進行辨識，藉以提高辨識模組104對影像 的成功辨識率。傳輪模組llGUeb、KP/lm則是；來 讓遠端細作端2GG可連結至自動影像比對系統⑽，如此 來使用者便可藉由遠端操作端200遠端地控制自動影 像比對系統1〇〇。 〜 第9頁/共22頁 表單編號A0101 M410287 [0030] 請參閱第2及3圖，其係為本創作之自動影像比對系統之 自動動態影像辨識比對模式之第一示意圖及第二示意圖 。第2圖係擷取辨識影像中的圖像，而第3圖係擷取辨識 影像中的數字。由第2及3圖中可以看到有兩個晝面框可 以作比較，左邊的畫面框經由攝影鏡頭擷取而來的影像 ，右邊的則是經由影像處理過後之影像。右邊影像當中 有一鎖定框架，此即為ROI鎖定之功能（區域限定）。兩示 意圖中，有一按钮名稱為Open Character Set File， 此按鈕動作後可以讓使用者選擇對照的特性資料。左下 方有兩個陣列畫面，一為儲存已辨識影像之資料庫，另 一為儲存未辨識影像之資料庫，雨資料庫·都是以影像檔 (.bmp)儲存於相對應資料爽的動作1為圖ί有一橢圓狀 , . .- <ι· '- 之物件，其係為本創作所述之提示模組，·係比對成功後 發出提示信號之用。 [0031] 請參閱第4圖，其係為本創作之自動影像比對系統之半自 動靜態影像辨識比對模式之示意圖、半自動靜態影像比 對之操作是將已經儲存好的影像，再——從資料夾中取 出做比對，並設有和資料庫比對成功時，發出提示動作 的功能。當使用者想要一邊執行半自動靜態影像比對功 能，一邊監控時，可調整系統讀取時間放慢或加快。 [0032] 請參閱第5圖，其係為本創作之自動影像比對系統之手動 靜態影像辨識比對模式之示意圖。靜態影像比對是將已 經儲存起來之影像檔取出來，經由人工的方式比對其内 容。此模組則無比對成功提示之功能。 [0033] 第參閱第6圖，其係為本創作之自動影像比對系統之第二 表單編號Α0101 第10頁/共22頁 實施例之第一流程圖。此為自動動態影像辨識比對之流 程圖，其步驟為：S61 :利用攝像機接收動態影像；S62 :對影像進行影像處理；S63 :對影像中之圖像或數字進 行擷取辨識；S64 ;是否辨識成功，若不成功至步驟S65 ，若成功則至步驟S66及S67 ; S65 :儲存無法辨識之影 像於未辨識資料夾；S66 :儲存辨識成功之影像於已辨識 資料夾；S67 :對辨識成功之影像進行資料庫比對動作； 568 :是否比對成功’若否至步驟S67，若是則至步驟 569 ; S69 :將比對成功之影像儲存於比對成功資料夾中 ，並發出提示訊號。 [〇〇34]請參閱第7圖’其係為本創作挲;1¾¼¾¾.系統之第二 實施例之第二流程圖。此為，i自:丨影廣^辨識比對之 流程圖，其步驟為：S71 :系統提供使用者讀取其中一資 料爽中所有的靜態影像；S72 :對影像進行影像處理； S73 :對影像中之圖像或數:字進行擁取辨識；S74 ;是否 辨識成功，若不成功至步_公7^〔若動至步驟S76及 . * / ： ··' ·' S77 ; S75 :儲存無法辨識之影#於識資料夾；S76 net:: :儲存辨識成功之影像於已辨識資料夾；S77 :對辨識成 功之影像進行資料庫比對動作；S78 :是否比對成功，若 否至步驟S77，若是則至步驟S79 ; S79 :將比對成功之 影像儲存於比對成功資料夾中，並發出提示訊號。 [0035] 請參閱第8圖，其係為本創作之自動影像比對系統之第二 實施例之第三流程圖。此為手動靜態影像辨識比對之流 程圖，其步驟為：S81 :系統提供使用者讀取其中一資料 夾之其中之一個靜態影像；S82 :使用者開啟比對資料庫 表單編號A0101 第1丨頁/共22頁 M410287 中的比對影像；S83 :開始進行人工比對。 [0036] 請參閱第9圖，其係為本創作之自動影像比對系統之第三 實施例之示意圖。此示意圖係要更加明述本創作之自動 影像比對系統可置於何種用途。本自動影像比對系統300 可架置於道路旁，以對行進間的車輛進行車牌辨識之用 。當藉由攝影機照射到車牌時，自動影像比對系統300便 會對所擷取到的影像進行影像處理，再加以進行資料庫 辨識比對。如此，不論是對於要記錄違反交通規則之車 輛，或是要用來查尋贓車，都有很大的效用。當然本創 作不僅只可辨識數字，亦可辨識圖像、文字等等，在此 僅係以數字辨識做一說明，但不侷限#此。 -. 人 ' · [0037] 综合上述，經由本創作之自動影像比對系統，可自動且 快速的辨識比對動態影像，且在辨識及比動的過程中， 對處理過的影像進行分類儲存，讓使用者在使用操作上 非常的便利。 [0038] 以上所述僅為舉例性，而非為限制性者。任何未脫離本 創作之精神與範疇，而對其進行之等效修改或變吏，均 應包含於後附之申請專利範圍中。 【圖式簡單說明】 [0039] 第1圖係為本創作之自動影像比對系統之第一實施例之方 塊圖； 第2圖係為本創作之自動影像比對系統之自動動態影像辨 識比對模式之第一示意圖； 第3圖係為本創作之自動影像比對系統之自動動態影像辨 表單編號A0101 第12頁/共22頁 M410287 識比對模式之第二示意圖； 第4圖係為本創作之自動影像比對系統之半自動靜態影像 辨識比對模式之示意圖； 第5圖係為本創作之自動影像比對系統之手動靜態影像辨 識比對模式之示意圖； 第6圖係為本創作之自動影像比對系統之第二實施例之第 一流程圖； 第7圖係為本創作之自動影像比對系統之第二實施例之第 二流程圖； • 第8圖係為本創作之自動影像比對系統之第二實施例之第 三流程圖；以及 第9圖係為本創作之自動影像比對系統之第.三實施例之示 意圖。 ^ 【主要元件符號說明】 [0040] 100、300 :自動影像比對系統 101 :比對資料庫 102 :接收模組 103 :影像處理模組 104 :辨識模組 105 :比對模組 106 :提示模組 * 107 :區域限定模組 108 :選取模組 10 9 :儲存模組 110 :傳輸模組 表單編號A0101 第13頁/共22頁 M410287 200 :遠端操作端 S6卜S69、S7卜S79、S8卜S83 :步驟 表單编號A0101 第14頁/共22頁[0002] With the advent of the era of science and technology, the research on intelligent related systems has flourished with the development of computers. Its application has been fully realized in the field of computers. It can be said that there are places where computers exist, and there are intelligent related systems. The existence; wherever automation systems are needed, where is the application of artificial intelligence such as -> ·/ - v 1 theory, methods and techniques. And the Department of Shadows 熬# an important part of the intelligent system, especially the map; ¥ 铁 S S ' 妁 妁 妁 妁 妁 妁 妁 妁 妁 妁 妁 妁 妁 ¥ ¥ ¥ ¥ ¥ 人 人 人 人 人 人 人 人 人 人 人 人 人 人 人 人 人Completion, in recent years, more and more people have devoted themselves to the research of digital images, which has led to the development of skin speed in the field of image processing technology. For example, in the fingerprint identification, in order to make the fingerprint lines more; the display must be pre-image processing, binarization and thinning, etc., so that the detailed feature extraction can be more obvious, and the recognition rate will be relatively improved. [0003] At present, most of the image comparison systems can only recognize and compare static images (such as photos, pictures, etc.) in a manual or semi-automatic manner, but cannot automatically analyze moving images (movies). For identification comparison. Moreover, after the current system recognizes the image, it cannot automatically distinguish the identification result and classify and store it, so that after the identification, only the successful image can be found relatively quickly, if the user wants to further When reviewing images that have not been successfully identified or successfully compared, it is necessary to spend some time searching for the image of the file form number Α0101, page 3 of 22 pages. [0004] Therefore, this creation establishes an automatic image recognition system, so that the dynamic images can be automatically and quickly identified and compared, and after the identification is compared, the system automatically classifies them so that the comparison images are to be re-examined. When you can find the file quickly. [New Content] [0005] In view of the above-mentioned problems of the prior art, one of the purposes of the present invention is to provide an automatic image comparison system to solve the problem that the prior art cannot automatically recognize the motion image and recognize the image after the comparison. Unable to automatically sort storage issues. [0006] According to another object of the present invention, a self-alignment system is proposed, which includes a comparison database, a receiving module, and an image. The processing module, an identification module, a comparison module and a prompt module. The comparison database stores at least one alignment image. The receiving module is selectively turned on to receive an image data. The image processing module is connected to the receiving module for image processing of the image data. The identification module is connected to the image processing module to identify the portion of the image data to be compared. The comparison module is connected to the identification module and the comparison database. If the identification module successfully identifies the image data, the comparison module compares the identified image data with each comparison image, and discriminates the identified image. Whether the image data corresponds to one of the comparison images. The prompting module is connected to the comparison module. If the recognized image data corresponds to one of the matching images, the prompting module sends a prompt signal. [0007] The automatic image comparison system of the present invention further comprises an area defining module, which is connected to the identification module, to provide the identification module to the image form number A0101, page 4 / total 22 pages. A limited area is subjected to identification processing to improve the recognition rate. [0008] wherein the comparison image is a still image. [0009] The image data is a dynamic image, and the dynamic image is composed of a plurality of static images. The identification module identifies each static image of the motion image, and the comparison module compares the static images identified in the motion image with the comparison images. The automatic image comparison system of the present invention further comprises a storage module, which is a connection identification module and a comparison module. The storage module includes a plurality of data folders to respectively store the successfully identified, unidentified success and successful comparison. A still image in a motion picture. >>···< · ν' [0010] The automatic image described in this creation is more packaged than the pair of 'systems': it includes a storage module, which contains at least one folder, α德Reclamation #-state image. And the automatic image comparison system further comprises a selection module, which is connected to the storage module and the image processing module, so as to provide a user to select one of the folders and then use the image processing module to view the game: Shiba 4 several still images for image processing. The storage module is further connected to the module and the comparison module /" » ^ t , and the storage module further includes a funded, undocumented folder and a successful folder for storing the identified successfully. Static images that have not been identified successfully and compared successfully. The selection module can also provide the user to select one of the static images of one of the folders, and provide the user to select one of the comparison images for manual comparison [0011] wherein the image data The system is a static image, and the automatic image comparison system described in the present invention further comprises a storage module, which is connected to the identification module and the form number A0101 page 5/22 pages M410287 comparison module to be static The images are stored in an identified folder, an unidentified folder, or a matching success folder. [0012] wherein the identification module identifies the portion of the image data to be compared, which is a text, a number, a graphic, or a combination thereof. [0013] wherein, the receiving module is an imaging device to capture image data. [0014] wherein the image processing module performs image processing on the image data, including binarization, filtering, or background removal. [0015] wherein the prompting module sends a prompt signal in the manner of a light number, a sound, or a combination thereof. [0016] wherein, the automatic image comparison system of the present invention further comprises a transmission module to provide a remote operation terminal: i: remote control: automatic image comparison system. [0017] As described above, the automatic image matching system according to the present invention may have one or more of the following advantages: [0018] (1) The automatic image comparison system can automatically identify and compare the images. The captured dynamic image means that when the photographic lens is continuously ingested, the simultaneous comparison can be performed simultaneously, and the dynamic image is no longer needed to be manually recognized when stored as a plurality of still images. [0019] (2) After the automatic image comparison system performs the recognition and comparison operation on the image, the system automatically stores the images that have been successfully recognized, unidentified, and successfully compared, so that the user When you want to review the image, you can quickly find the file. [0020] (3) The automatic image comparison system further integrates the function of the network connection, so that the form number A0101 can be controlled by the user via the remote control. [0021] (4) In addition to providing an automatic recognition comparison image function, the automatic image comparison system provides a semi-automatic and manual recognition static image function, so that the user can further confirm the result of the recognition comparison image. [Embodiment] The embodiments of the automatic image comparison system according to the present invention will be described below with reference to the related drawings. For the sake of understanding, the same components in the following embodiments are denoted by the same reference numerals. [0023] Please refer to FIG. 1 , which is a block diagram of a first embodiment of the automatic image comparison system of the present invention. In Fig. 1, there is an automatic f: image comparison system 100 and a remote operation terminal 200. Automatic shadow:: like comparison * 'system 10: contains comparison database 101, receiving module 102, shadow; rubber hair module? 1 ^ 3, identification module 104, comparison module 105, The prompt module 106, the area defining module 107, the selecting module 108, the storage module 109, and the transmission module 110. The automatic image comparison system is connected to each module in the system 100. The measurement is shown in the figure. [0024] In the first embodiment, the self-motion and f-image comparison system 100 of the present invention can be mainly divided into three types: automatic motion image recognition comparison, semi-automatic still image recognition comparison, and manual static image recognition comparison. On the mode. [0025] The automatic motion image recognition comparison mode: the automatic motion image recognition comparison method is that when the receiving module 102 (camera device) receives the captured motion image, the image processing module 103 starts to perform motion image processing. Binaryization, filtering, or image removal operations such as background removal allow the recognition module 104 to more easily identify the image. After the image processing operation is completed, the recognition module 104 will start to recognize the motion image to capture the part number A0101, page 7 of 22, M410287, which will be compared (can be digital, text) 'Image Geqi combination'. After the module 1G4 recognizes the successful operation of the image, the comparison module 105 compares the image with the (4) images stored in the comparison database 1〇1. If the image corresponds to the image after the comparison, the prompt module 106 will issue a prompt signal in the form of a light number, a sound or a group of eight. Among them, the motion picture can be seen as a series. The fast-moving static image is combined, and the image processing module 1〇3, the identification module m and the comparison module 105 perform image processing for each (four) image frame that the moving image runs. And comparison. Therefore, when the receiving module 1G2 continuously captures the motion picture, the system has been continuously performing image processing, _ and comparison (the action of capturing the motion picture is still continuous and uninterrupted). When the image is first stored in the still image, the comparison operation is performed. [0026] In the above, if there is an unrecognizable still image in the moving (four) image, the matching module m is not pure. The m minus will store the still image in the unidentified data in the storage module 1G9 (the folder name can be customized by the user). A 4 and the still image in the right motion image is successfully recognized, except After the continuation, the system will also save the still image in the identified folder in the storage module 109. The comparison is successful through the comparison of $4105. The image is stored in the storage module 109 by the 丄τ success folder. The action stored in this category is for the convenience of the user when the user wants to review and re-align the image. Find the image file. The stored images are analyzed by moving images into still images and then sorted and stored. Form No. 1010101 Page 8 of 22 [0027] Semi-automatic still image recognition comparison mode: semi-automatic static image recognition comparison: Firstly, a plurality of static images are first stored in the storage module 109: in the I-clip, the storage module 109 additionally has an unidentified folder, and the data has been lost and compared to the successful folder. The user automatically statically makes the time. When the comparison mode is identified, one of the remaining images of the static image can be selected by selecting the module H. Then the image processing module 103, the identification module 1〇4 and the comparison are performed. Module 1〇5 will perform image processing, δ, and comparison operations on the selected folder _ all static images--with respect to the comparison images in the comparison database 〇1 ^ When the comparison is successful, the prompt module will send a prompt message. The semi-automatic static image recognition ratio will be unsuccessful, and the successful static image will be compared. (4) Stored in the mineral knowledge folder and the comparison success folder [0029] [0029] Manual static rhyme Bilong": Manual static (four) image recognition comparison method: the user selects a static image by selecting one of the folders of the Lai group 1〇9, and the V body selects the comparison database. Compare the images for manual manual identification. In the first-real _ towel '11 domain limit group 107 (four), is used to limit the image into the domain, so the identification module m will only (4) Identifying the middle limit & area to improve the recognition rate of the image by the recognition module 104. The transfer module llGUeb, KP/lm is; let the remote fine end 2GG can be connected to the automatic image comparison system (10) Thus, the user can remotely control the automatic image comparison system 1 by the remote operation terminal 200. ~ Page 9 of 22 Form No. A0101 M410287 [0030] Please refer to Figures 2 and 3, which are the first schematic diagram and the second schematic diagram of the automatic motion image recognition comparison mode of the automatic image comparison system of the present invention. . Figure 2 captures the image in the identified image, while Figure 3 captures the number in the identified image. It can be seen from Figures 2 and 3 that there are two face frames that can be compared. The image frame on the left is captured by the camera lens, and the image on the right is the image processed through the image. There is a locking frame in the image on the right, which is the function of ROI locking (area definition). In the two intents, there is a button named Open Character Set File, which allows the user to select the matching profile data. There are two array screens on the lower left. One is a database for storing recognized images, and the other is a database for storing unidentified images. The rain database is stored in the corresponding file (.bmp). 1 is an ellipse, . .- <ι· '- object, which is the prompt module described in the creation, and is used to send a prompt signal after successful comparison. [0031] Please refer to FIG. 4, which is a schematic diagram of a semi-automatic still image recognition comparison mode of the automatic image comparison system of the present invention, and the semi-automatic static image comparison operation is to store the already stored image, and then The folder is taken out for comparison, and the function of prompting the action is performed when the comparison with the database is successful. When the user wants to perform the semi-automatic still image comparison function while monitoring, the system reading time can be adjusted to slow down or speed up. [0032] Please refer to FIG. 5, which is a schematic diagram of a manual static image recognition comparison mode of the automatic image comparison system of the present invention. The static image comparison is to take out the stored image file and compare it manually. This module is incomparable to the function of successful prompts. [0033] Referring to FIG. 6, which is the second form number of the automatic image matching system of the present invention, the first form number is 。0101, page 10/22. The flow chart of the automatic motion image recognition comparison is as follows: S61: receiving a motion image by using a camera; S62: performing image processing on the image; S63: capturing an image or a number in the image; S64; If the identification is successful, if it is not successful to step S65, if it succeeds, it will go to steps S66 and S67; S65: store the unrecognizable image in the unidentified folder; S66: store the successfully identified image in the identified folder; S67: the identification succeeds The image is compared with the database; 568: Whether the comparison is successful 'If not to step S67, if yes to step 569; S69: The successful image is stored in the comparison success folder, and a prompt signal is issued. [〇〇34] Please refer to Fig. 7' which is the second flow chart of the second embodiment of the system; 13⁄41⁄43⁄43. This is, i from: 丨影广^ identification comparison flow chart, the steps are: S71: the system provides the user to read all the static images of one of the data cool; S72: image processing of the image; S73: right Image or number in the image: word for recognition; S74; whether the recognition is successful, if it is not successful, step _ public 7^ [If you move to step S76 and. * / : ··· · · S77 ; S75 : Save Unrecognizable shadow #知识夹; S76 net:: : store the image of the successful identification in the identified folder; S77: compare the image of the successfully identified image; S78: whether the comparison is successful, if not Step S77, if yes, go to step S79; S79: store the successfully compared image in the comparison success folder, and send a prompt signal. [0035] Please refer to FIG. 8 , which is a third flowchart of the second embodiment of the automatic image comparison system of the present invention. The flow chart of the manual static image recognition comparison is as follows: S81: the system provides the user to read one of the static images of one of the folders; S82: the user opens the comparison database form number A0101. Page / Total 22 pages of M410287 alignment image; S83: Begin manual comparison. [0036] Please refer to FIG. 9, which is a schematic diagram of a third embodiment of the automatic image comparison system of the present invention. This diagram will further clarify the purpose of the automatic image comparison system of this creation. The automatic image comparison system 300 can be placed beside the road to perform license plate recognition for vehicles traveling between. When the license plate is illuminated by the camera, the automatic image comparison system 300 performs image processing on the captured image, and then performs database identification comparison. In this way, it is very useful for recording vehicles that violate traffic rules or for searching for vehicles. Of course, this creation not only recognizes numbers, but also recognizes images, texts, etc., and only uses digital identification as a description, but it is not limited to this. -. People's [0037] In summary, the automatic image comparison system of the present invention can automatically and quickly identify the comparison motion images, and classify and store the processed images in the process of identification and comparison. To make the user very convenient in the operation. [0038] The foregoing is illustrative only and not limiting. Any equivalent modifications or variations that are made without departing from the spirit and scope of this creation should be included in the scope of the appended patent application. BRIEF DESCRIPTION OF THE DRAWINGS [0039] Fig. 1 is a block diagram of a first embodiment of the automatic image comparison system of the present invention; Fig. 2 is an automatic motion image recognition ratio of the automatic image comparison system of the present invention The first schematic diagram of the mode; the third diagram is the automatic dynamic image recognition form number A0101 of the automatic image comparison system of the creation, the second schematic diagram of the M410287 recognition comparison mode; the fourth figure is The schematic diagram of the semi-automatic static image recognition comparison mode of the automatic image comparison system of the present invention; FIG. 5 is a schematic diagram of the manual static image recognition comparison mode of the automatic image comparison system of the creation; FIG. The first flow chart of the second embodiment of the automatic image comparison system; the seventh figure is the second flow chart of the second embodiment of the automatic image comparison system of the present invention; The third flowchart of the second embodiment of the automatic image comparison system; and the ninth diagram is a schematic diagram of the third embodiment of the automatic image comparison system of the present invention. ^ [Main component symbol description] [0040] 100, 300: automatic image comparison system 101: comparison database 102: receiving module 103: image processing module 104: identification module 105: comparison module 106: prompt Module * 107 : Area Qualification Module 108 : Selection Module 10 9 : Storage Module 110 : Transmission Module Form No. A0101 Page 13 / Total 22 pages M410287 200 : Remote Operation Terminal S6 S69, S7 Bu S79, S8 Bu S83: Step Form No. A0101 Page 14 of 22