六、 [0001] [0002] [0003] [0004] 097108064 發明說明： 【發南所屬之技術領域】 一 、 本發明於―種影像麵裝置，㈣係種快迷 7影像之影像存取裝置。本發明還提供—種使用該影 像存取裝置之影像存儲及讀取方法。 【先前技術】 影像資料作為一種可真實、直觀記錄所需場景之媒介， 2廣泛剌於各個領域’“，雜諸讀時往往都 己將其5類’即便記得分類亦會因需要手動輸入分類 =稱之Μ過程而將其忽略’從而導致當需要從大量存 :於存儲設備中之影像資料中尋找出自己需要之影像資 Γ得十分之繁瑣，通常需要使詩时動之方式於^ 2之影像資料中尋找，如此將需要花費相當長之時間 —找，查询過程耗> 時而枯燥。 【發明内容】 有蓉於此’有必要提供—_像存取裝置Μ 該影像存取裝置之影像存儲及讀取方法。 使用 Γ種影像存取裝置包括—存料元…-單元以及一 :音存取單元1述細W存儲影像資料 與所述影像資料組對應之語音特徵請。所述奸及 :以!示存儲於所述存儲單元中之影像資料。所述語： 存取單7L包括-麥克風' _影 9 取模组包括所述影像存 包括s。曰獲取子模組、查找子模組、判斷子模组 以及影像輸出子模組； 模組 所述語音診顿_簡由料麥^獲 表單編號觸1 第3頁/共17頁 J ° θ 1003294859-0 [0005] 1353591 100年08月15日’核正替換頁 資料； ί- I t* - ί · * I ιί；·> 1*1 - < • 、 [0006] 所述查找子模組用以由#儲單元中尋找是否存在與由所 述語音獲取子模組所獲取之當前語音資料相似之語音特 徵資料； [0007] 所述判斷子模組用以判斷是否找到與當前語音資料相似 之語音特徵資料； [0008] 影像輸出子模組用以當找到與當前語音資料相似之語音 特徵資料時，將與所述語音特徵資料對應之影像資料組 | 輸出至顯示單元。 [0009] 一種影像存儲方法其包括以下步驟： [0010] 獲取一待存影像資料； f [0011] 藉由麥克風獲取一與待存影像資料之内容相對應之當前 語音資料； [0012] 根據該當前語音資料從一存儲單元中尋找與當前語音資 料最相近之語音特徵資料； · [0013] 判斷是否找到與當前語音資料相似之語音特徵資料； [0014] 若找到與當前語音資料相似之語音特徵資料，則將所述 待存影像資料存儲於與該語音特徵資料相對應之影像資 料組中； [0015] 若未找到與當前語音資料相似之語音特徵資料，則確認 是否需要建立新的語音特徵資料，若係則建立新的語音 特徵資料及與該語音特徵資料對應之影像資料組，並將 097108064 表單編號A0101 第4頁/共17頁 1003294859-0 1353591 1 « 100年08月15日慘正替换頁 待存影像資料存儲於該語音特徵資料所對應之影像資料 組中。 、 · [0016] 一種影像讀取方法，其包括以下步驟： [0017] 藉由麥克風獲取一當前語音資料； [0018] 根據當前語音資料從一存儲單元中尋找與當前語青資料 最相近之語音特徵資料； [0019] 當找到與當前語音資料最相近之語音特徵資料時，將存 • 儲於所述存儲單元中與該語音特徵資料對應之影像資料 傳輸至顯示單元顯示。 [0020] 相較先前技術，本發明中採用語音存儲方式對影像資料 進行分類存儲及讀取，替代了手動存儲及讀取之繁瑣過 程，極大之方便了使用者對大量之影像資料進行存儲及 讀取之工作。 [0021] • 【實施方式】 請參閱圖1，本發明影像存取裝置100包括一存儲單元110 、一顯示單元120、一輸入單元130以及一語音存取單元 140 ° ' [0022] 所述存儲單元110用以存儲影像資料112以及語音特徵列 表114。所述影像資料112根據其所記載之内容之不同類 型分為不同之影像資料組112a，且每一影像資料組112a 均有自己特定之特徵碼如a、b、c等。所述語音特徵列表 114中包括多個與所述影像資料組112a中之特徵碼a、b 、（:等對應之語音特徵資料114a如1、2、3等。 097108064 表單編號A0101 第5頁/共17頁 1003294859-0 丄二丄[0001] [0002] [0003] [0003] 097108064 Description of the invention: [Technical field of the South] 1. The present invention is a kind of image surface device, and (4) an image access device for 7 images. The present invention also provides an image storage and reading method using the image access device. [Prior Art] As a medium for real and intuitive recording of required scenes, 2 imagery is widely used in various fields. “When reading, it is often used to classify it.” Even if you remember to classify it, you will need to manually enter the classification. = Ignore the process and ignore it', which leads to the cumbersome task of finding the image that you need from a large amount of storage: the image data in the storage device. It is usually necessary to make the poetry movement method ^ 2 Looking for the image data, it will take a long time - find, query process consumption > sometimes boring. [Summary of the content] There is a need to provide -_ image access device Μ the image access device Image storage and reading method. The image access device includes: a storage unit...-unit and a: audio access unit 1 to describe the voice characteristics of the image data and the image data group. The image data stored in the storage unit is displayed in the storage unit. The language: the access ticket 7L includes a microphone _ shadow 9 取 module includes the image storage includes s. 曰 acquisition sub-module, check Sub-module, judging sub-module and image output sub-module; module said voice diagnosis_simple material me ^ get form number touch 1 page 3 / total 17 pages J ° θ 1003294859-0 [0005] 1353591 On August 15th, 100th, 'the nuclear replacement page information; ί- I t* - ί · * I ιί;·> 1*1 - < • , [0006] The search sub-module is used to store by # The unit is configured to find whether there is a voice feature data similar to the current voice data acquired by the voice acquisition sub-module; [0007] the determining sub-module is configured to determine whether to find a voice feature data similar to the current voice data; [0008] The image output sub-module is configured to output the image data group corresponding to the voice feature data to the display unit when the voice feature data similar to the current voice material is found. [0009] An image storage method includes The following steps are performed: [0010] acquiring a current voice data corresponding to the content of the image data to be stored by using a microphone; [0012] searching for a current memory data according to the current voice data Most with current voice material Similar voice feature data; [0013] determining whether to find a voice feature data similar to the current voice material; [0014] if the voice feature data similar to the current voice profile is found, storing the to-be-stored image data in the In the image data group corresponding to the voice feature data; [0015] if no voice feature data similar to the current voice data is found, it is confirmed whether new voice feature data needs to be established, and if the system establishes new voice feature data and The image data set corresponding to the voice feature data, and 097108064 Form No. A0101 Page 4 of 17 1003294859-0 1353591 1 « 100 years of August 15th, the replacement page to be stored image data is stored in the voice feature data corresponding to In the image data group. [0016] An image reading method, comprising the following steps: [0017] acquiring a current voice data by using a microphone; [0018] searching for a voice closest to the current language data from a storage unit according to the current voice data Feature data; [0019] When the voice feature data closest to the current voice material is found, the image data stored in the storage unit corresponding to the voice feature data is transmitted to the display unit for display. [0020] Compared with the prior art, the present invention uses a voice storage method to classify, store and read image data, instead of the cumbersome process of manual storage and reading, which greatly facilitates the user to store a large amount of image data and Read the work. [0021] The first embodiment of the present invention includes a storage unit 110, a display unit 120, an input unit 130, and a voice access unit 140[ '. The unit 110 is configured to store the image data 112 and the voice feature list 114. The image data 112 is divided into different image data sets 112a according to different types of contents, and each image data set 112a has its own specific feature code such as a, b, c, and the like. The voice feature list 114 includes a plurality of voice feature materials 114a corresponding to the feature codes a, b, (:, etc. in the image data group 112a, such as 1, 2, 3, etc. 097108064 Form No. A0101 Page 5 / Total 17 pages 1003294859-0 丄二丄
[0025] Γ4Γ存取單元14G包括—麥克風141、—影像存取模 [0026]所述麥克風l4i用 以將使用者之語音信號轉化為語音 資料 所述影像存取模組142用以藉由語音 料或者讀取存儲於所述存儲單元U0中之影像資料H 像存取模_包__+_422\== 組1424 查找子模組 、 儲子模电判斷子棋组1428、影像存 ^子模⑴43G、影像輪出子模組咖以及 建立子模組1434。 〜像資科組 =子模組1422可用以直接 獲取之待存=裝置 Η22亦可以從其他之存 ％像㈣子模組 取所述財影像資料。'J， «、⑽卡中獲 _===，所述麥克-_當 集子模組！422所獲取二2 =用者根據所述影像採 相應特徵之語音信號。存㈣f料之内容而發出具有 [0030] 所述查找子模組1426用 097108064 [0028] 从於存儲單元110中尋找是否存在 第6頁/共17 表單鴆號aoi〇1 頁 1003294859-0 [0031] [0032] [0033] [0034] [0035] [0036] 097108064 1100 年 08 月 15Q 與由所述語㈣取子馳1424所魏之當祕音資料相 似之語音特徵資料114a。 所述判斷子模組1428用㈣斷是否_與《語音資料 相似之語音特徵資料： 所述影像存館子模組⑽用以#_子模組“Μ判斷找 當前語音資料相似之語音特徵資料ma時，將所 像資料存儲於與所述語音特徵fml4a對應 之衫像資料組112a中； 到了:Ϊί出子模組1432用以當判斷子模組1428判斷找 二之語音特徵資料一，將存 資料組心輸出至顯=5徵料仙對應之影像 料組建立子模組1434用於當未找到與當前語音資 钭相似之語音特徵資料⑴ 資 〜與該新的語音特徵資料=的語音特徵資料 ll2a。 貫抖ll4a對應之影像資科組 垓影像存取模組！ 42還可 於找到所查詢之影像資料％^模式選擇子模組咖用以 於所述顯示單元120令，/後選擇以何種方式顯示 流覽兩種方式。 、、不方式包括自動流覽和手動 發月令採用語音存儲方式 讀取，替代了手動_切^像㈣特分類存館及 了使用者對大量之影像資之繁_程，極大之方便 n連行存儲及讀取之工作。 表單鵠號 1003294859-0 1353591 100年08月15日τ核正替換頁 [0037] 請參閱圖2，一種影像存儲方法，其包括以下步驟： [0038] S210,建立語音特徵資料及與該語音特徵資料對應之影像 資料組，該步驟具體包括以下步驟： [0039] 藉由所述麥克風141獲取語音； [0040] 從所述語音中萃取出當前語音特徵資料114a ; [0041] 查找在存儲單元110中是否已存在與所萃取之當前語音特 徵資料114a相似之語音特徵資料； [0042] 若未找到與所萃取之當前語音特徵資料114 a相似之語音 特徵資料，則建立與所述當前語音特徵資料114a對應之 影像資料組112a ; [0043] 將所述當前語音特徵資料114a以及與其對應之影像資料 組112a存儲於存儲單元110中。 [0044] S220,獲取一待存影像資料； [0045] S230,藉由麥克風獲取一與待存影像資料之内容相對應之 當前語音資料； [0046] S240,根據該當前語音資料從語音特徵列表114中尋找與 當前語音資料最相近之語音特徵資料114a; [0047] S250,判斷是否找到與當前語音資料相似之語音特徵資料 114a; [0048] S260,若找到與當前語音資料相似之語音特徵資料，則將 所述待存影像資料存儲於與該語音特徵資料114a相對應 之影像資料組112a中； 097108064 表單编號 A0101 第 8 頁/共 17 頁 1003294859-0 [0049] )91 奶〇，若未制與當前語音⑽相似喊 114a，則確認是否需要建立新的語音特徵資料ul; [0050] S280,若是’則建立新的語音特徵資料114a絲該w 特徵資1關^w_U2a，並料存影^資 料存儲於⑼音特徵資料U4a所對應之影像資料组112a 中。若否則結束該影像存儲。 [0051] [0052] [0053] 請參閱圖3 ’―種影像讀取方法，其包如下步驟： S310,藉由麥克風獲取—當前語音資料； S320根據該當前語音⑽於—存儲單元⑴ 前語音資料最相近之語音特徵資料U4a; 、田 [0054] 是邱到與當料音㈣最㈣之語音特徵 料114a; 資 [0055] [0056] s::〇’當找到與當前語音資料最相近之語音特徵資料 1 a > 自動流覽模式或者 與該^音特徵資料仙所對應之影像資料組覽 ;方未找料料音f_目近之語音特 重複步驟_至$34〇。 韻4a時則 m存铸於所述存儲單元110中與該語音特徵資料 义影像資料傳輪至顯示單元顯示120。 發明之範圍並不以上=明之較佳實施方式，本 藝之人埃依林1 式為限’軌絲本案技 本發月之精神所作之等效修飾或變化，皆 097108064 表單編號A0101 笫9頁/共17頁 1003294859-0 [0057] 1353591 ___:_— 100年08月15日•修正替換頁 應涵蓋於以下申請專利範圍内。 ^ 【圖式簡單說明】 [0058] 圖1係本發明影像存取裝置框架圖； [0059] 圖2係本發明影像存儲方法流程圖； [0060] 圖3係本發明影像讀取方法流程圖。 【主要元件符號說明】 [0061] 影像存取裝置：100 [0062] 存儲單元：110 φ [0063] 影像資料：112 [0064] 影像資料組：112a [0065] 語音特徵列表：114 [0066] 語音特徵資料：114a [0067] 顯示單元：120 [0068] 輸入單元：130 φ [0069] 語音存取單元：140 [0070] 麥克風：141 [0071] 影像存取模組：142 [0072] 影像採集子模組：1422 [0073] 語音獲取子模組：1424 [0074] 查找子模組：1426 097108064 表單編號A0101 第10頁/共17頁 1003294859-0 100年08月15日修正替換頁 1353591 [0075] 判斷子模組：1428 [0076] 影像存儲子模'組：1 4 3 0 [0077] 影像輸出子模組：1432 [0078] 模式選擇子模組：1436 [0079] 影像資料組建立子模組：1434 參[0025] The access unit 14G includes a microphone 141, an image access module [0026], the microphone 14i is used to convert a user's voice signal into voice data, and the image access module 142 is used for voice Read or read the image data H stored in the storage unit U0. Image access module_package__+_422\== group 1424 search sub-module, memory model judging board 1428, image storage sub-module (1) 43G The image wheel is sub-module and the sub-module 1434 is created. ~ Like the group of subordinates = sub-module 1422 can be used to directly obtain the pending = device Η 22 can also take the financial image data from other stored (like) sub-modules. 'J, «, (10) get _=== in the card, the Mike-_ when the set sub-module! 422 acquires 2 2 = the user uses the voice signal of the corresponding feature according to the image. The content of the (4) f material is issued with [0030] The search sub-module 1426 is used with 097108064 [0028] From the storage unit 110 to find out whether there is a sixth page / a total of 17 forms nickname aoi 〇 1 page 1003294859-0 [0031 [0036] [0036] [0036] 097108064 1100 August 15Q and the speech feature data 114a similar to the secret music data of the 1424 Weizhi. The determining sub-module 1428 uses (four) to break whether or not the voice feature data similar to the voice data: the image storage sub-module (10) is used for the #_sub-module to determine the voice feature data similar to the current voice data. In the case of ma, the image data is stored in the shirt image data set 112a corresponding to the voice feature fml4a; and the mouse module 1432 is used to determine that the child module 1428 determines the voice feature data of the second voice. The image data set output sub-module 1434 corresponding to the data set core output to the display group 5 is used for not finding the voice feature data similar to the current voice resource (1) - the voice with the new voice feature data = The feature data ll2a. The video image group corresponding to the image 资 ll4a 42 存取 a 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 42 ll ll ll Choose which way to display the two ways of browsing.,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, Image _程, great convenience n continuous storage and reading work. Form nickname 1003294859-0 1353591 100 years of August 15th τ nuclear replacement page [0037] Please refer to Figure 2, an image storage method, including the following steps [0038] S210: Establish a voice feature data and a video data group corresponding to the voice feature data, where the step specifically includes the following steps: [0039] acquiring a voice by using the microphone 141; [0040] extracting from the voice Presenting the current voice feature data 114a; [0041] finding whether the voice feature data similar to the extracted current voice feature data 114a already exists in the storage unit 110; [0042] if the extracted current voice feature data 114a is not found The similar voice feature data is used to create a video data set 112a corresponding to the current voice feature data 114a. [0043] The current voice feature data 114a and the corresponding image data set 112a are stored in the storage unit 110. 0044] S220, acquiring a to-be-stored image data; [0045] S230, acquiring, by the microphone, a current voice data corresponding to the content of the image data to be stored; [0 046] S240, searching for the voice feature data 114a closest to the current voice data from the voice feature list 114 according to the current voice data; [0047] S250, determining whether to find the voice feature data 114a similar to the current voice material; [0048] S260. If the voice feature data similar to the current voice data is found, the image data to be stored is stored in the image data group 112a corresponding to the voice feature data 114a; 097108064 Form No. A0101 Page 8 of 17 Page 1003294859-0 [0049] 91 milk, if not made to call 114a similar to the current voice (10), then confirm whether it is necessary to establish a new voice feature data ul; [0050] S280, if yes, then create a new voice feature data 114a The silk feature 1 is closed to ^w_U2a, and the stored image is stored in the image data set 112a corresponding to the (9) sound feature data U4a. If not, the image storage is ended. [0053] Please refer to FIG. 3, a method for reading an image, which includes the following steps: S310: acquiring a current voice data by a microphone; S320 according to the current voice (10) to a voice before the storage unit (1) The most similar phonetic feature data U4a; , Tian [0054] is the voice feature material 114a of Qiuzhi and Dangyin (4) most (four); [0055] [0056] s::〇' when found the closest to the current phonetic data The voice feature data 1 a > automatic browsing mode or the image data group corresponding to the sound feature data fairy; the sound is not found in the material f_ near the voice repeat steps _ to $ 34 〇. When the rhyme 4a is stored in the storage unit 110, the audio feature data is transmitted to the display unit display 120. The scope of the invention is not the above-mentioned preferred embodiment of the invention. The Ei-Lin 1 formula of the art is equivalent to the equivalent modification or change of the spirit of the moon in the style of the wire, and all of them are 097108064 Form No. A0101 笫 9 pages / Total 17 pages 1003294859-0 [0057] 1353591 ___: _ - August 15, 100 • The revised replacement page should be covered by the following patent application. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an image access device of the present invention; FIG. 2 is a flow chart of a method for image storage according to the present invention; [0060] FIG. . [Main component symbol description] [0061] Image access device: 100 [0062] Storage unit: 110 φ [0063] Image data: 112 [0064] Image data set: 112a [0065] Voice feature list: 114 [0066] Voice Characteristic data: 114a [0067] Display unit: 120 [0068] Input unit: 130 φ [0069] Voice access unit: 140 [0070] Microphone: 141 [0071] Image access module: 142 [0072] Image acquisition sub Module: 1422 [0073] Voice Acquisition Sub-module: 1424 [0074] Search Sub-Module: 1426 097108064 Form Number A0101 Page 10/Total 17 Page 1003294859-0 100 August 15 Revision Replacement Page 1353591 [0075] Judgment sub-module: 1428 [0076] Image storage sub-mode 'group: 1 4 3 0 [0077] Image output sub-module: 1432 [0078] Mode selection sub-module: 1436 [0079] Image data group creation sub-module :1434 ginseng
097108064 表單編號A0101 第11頁/共17頁 1003294859-0097108064 Form number A0101 Page 11 of 17 1003294859-0