201003578 九、發明說明： 【發明所屬之技術領域】 本發明涉及-種監控系統及其監控方法。 【先前技術】 :統之監視系統模式’是由多台監 2幕及錄放影機所組成。該監㈣_式由 ^ 監::，所拍攝之視訊藉由視訊傳: 丄點所設置之監看螢幕顯示視訊， 相由錄放純將其監看榮幕之視訊儲存於錄影帶中。如 此於使用上，可即時監蒼芬 低犯罪之效果。像’而達到維護治安降 ::而’為了找出犯罪嫌疑人之做案證據，必須將 ▼找出，韻影帶記憶之影像巾逐步查看該犯罪嫌疑人是 否出現及是否做過案’從而可找出證據。如此將耗費大量 之查找時間。同時人工查看，如果查看之時間過長，由於 疲勞也合易出現差錯’將犯罪嫌疑人從監控錄影中遺漏。 【發明内容】 ^有鑒於此，有必要提供一種可從錄取之影像中快速查 詢目標人物之影像之監控系統及其監控方法。 — 一種監控系統，其包括： 一攝像單元； 一人臉識別單元，用於該攝像單元所擷取之影像中識 別所有人之人臉，並對應各人臉分別生成一人臉輪廓信息，· 一记憶單兀，用於記憶該攝像單元擷取之影像及由人 6 201003578 臉識別單元生成之人臉輪廓信息； 一搜尋單元，用於針對目標人物於記憶單元内搜尋是 否有與先前所記憶之人臉輪廓信息相同之人臉； 一顯不單兀，用於顯示攝像單元擷取之影像及搜尋結 果。 一種監控方法，其包括以下步驟： (1)擷取影像； (2 )從擷取之影像中識別人臉，對應不同人臉分別生 成一人臉輪廟信息； (3) 記憶擷取之影像及人臉識別單元生成人臉輪廓信 息； (4) 針對一目標人物查詢是否與先前所記憶之人臉輪 廓信息相同； (5 )顯示查尋結果。 相較於先前技術，該人臉識別單元從該擷取之影像中 識別所有人之人臉，並對應各人臉分別生成一人臉輪廓信 息，该§己憶單元§己憶人臉識別單元識別之人臉輪廓信息， 搜錄單元可於對目標人物攝像，人臉識別單元生成之該人 之人臉輪廓信息後，於記憶單元内搜尋是否有與先前所記 憶之人臉輪廓信息相同之人臉，如此於所記憶之影像中可 快速查找具有相同之人臉信息之人臉。 【實施方式】 下面將結合附圖，對本發明作進一步之詳細說明。 請一併參閱圖1與圖2 ’為本發明提供之一種監控系統 7 201003578 100,其包括一攝像單元10、一轉送單元20、一顯示單元 30、一人臉識別單元40、一記憶單元50及一搜尋單元60。 該攝像單元10用於擷取被攝物之影像。該轉送單元20用 於將攝像單元10擷取之影像轉送至該人臉識別單元40及 顯示單元30。該人臉識別單元40從該擷取之影像中識別所 有人之人臉，並對應各人臉分別生成一人臉輪廓信息，該 記憶單元50記憶該攝像單元10擷取之影像及人臉識別單 元40生成之人臉輪廓信息。該搜尋單元60可針對目標人 物，於記憶單元50内搜尋是否有與先前所記憶之人臉輪廓 信息相同之人臉。該顯示單元30可顯示搜尋單元搜尋之結 果及顯示擷取之影像。 該一攝像單元10分別可包括複數攝像裝置19分別設 置於不同之監控區域。每個攝像裝置19包括一鏡頭單元11 及一感測單元12。該鏡頭單元11為一鏡頭模組，其包括至 少一光學鏡片。該鏡頭單元11用於彙聚光訊號，並將該光 訊號傳輸至感測單元12。該感測單元12為CCD ( Charge Coupled Device，電荷耦合組件感測器）或 CMOS (Complementary Metal Oxide Semiconductor,互補性金屬 氧化物感測器）。該感測單元12用於將鏡頭單元11彙聚之 光訊號轉換為影像訊號。 本實施方式中，該傳送單元20為一光纖，其用於將攝 像單元10擷取之影像訊號分別傳送至人臉識別單元40及 顯示單元30。 該顯示單元30可包括多台顯示器，於使用該攝像單元 8 201003578 10拍攝圖像時即時觀察藉由該轉送單元20傳送來之影像 擷取單元10所擷取之被攝物之影像。 實際應用中，當攝像單元10只有一攝像裝置19,而顯 示單元30只有一顯示器時，該攝像單元10擷取之影像訊 號也可直接傳輸至顯示單元30而不用藉由傳送單元2〇之 轉輸。 該人臉識別單元40用在於該攝像單元1〇所擷取之影 像中識別所有人之人臉，並對應各人臉分別生成一人臉輪 廓信息。 該記憶單元50用於記憶該至少一攝像單元擷取之影像 及人臉識別單元40生成之人臉輪廓信息。 該搜尋單元60其可針對目標人物於記憶單元4〇内搜 尋是否與先前所記憶之人臉輪廓信息相同。具體使用時， 如果需對一目標人物進行查詢，則攝像單元10對該目標人 物進行攝像，此時人臉識別單元40同時生成該人之人臉輪 廓信息’搜尋單元60把該目標人物之人臉輪廓信息與該記 隐單元50存儲之所有人臉輪廓信息比較。由於不同之人具 有不同之人臉輪廓信息’若搜尋單元60從記憶單元5〇搜 尋與此人相同之人臉輪廓信息，則將該目標人物之監視錄 相藉由顯示單元30顯示出來。以達到從記憶單元5〇存儲 之影像中快速查詢目標人物是否出現過之目的。 §亥監控系統100還進一步包括一壓縮控制單元及一 壓縮單元80。該壓縮控制單元70用於控制壓縮單元80是 否對攝像單元10擷取之影像進行壓縮處理，以便節約記憶 9 201003578 體空間量。使用時’當人臉識別罝—/ , _ , 臉％別早7L 4〇沒有識別到人臉時 (也即沒有人出現於監控區域内. L·· X N Bf )，該壓縮控制單元7〇 啟動該壓縮單元8G對影像進行I縮處理；反之，當人臉識 =單元40識別到人臉時，該壓縮控制單元％關閉該壓縮 單元80工作。如此，可诘，丨、兮格触 t J成j。己隐體，進一步達到快速查找 之目的。 請參閱圖3,為本發明之監控方法，其包括以下步驟： S102 :擷取影像； 該攝像單元之鏡頭單元u彙聚被攝物之光訊號，並將 該光訊號傳輸至感光單元12，該感光單元12將鏡頭單元 11彙聚之光訊號轉換為影像訊號。 S104 :從擷取之影像中識別人臉，對應不同人臉分別 生成一人臉輪廓信息； 該人臉識別單元40從該攝像單元10擷取之影像中識 別所有人之人臉’同時分別對應各人臉分別生成一人臉輪 廓信息。 S106 :記憶擷取之影像及人臉識別單元生成之人臉輪 廓信息； 該記憶單元50存儲該至少一攝像單元1〇擷取之影像 及人臉識別單元40生成之各人臉輪廓信息。 S108 :針對一目標人物查詢是否與先前所記憶之人臉 輪靡信息相同； 如果需對一目標人物進行查詢，則攝像單元10對該目 標人物進行攝像，此時人臉識別單元40同時生成該人之人 201003578 臉輪廊信息’搜尋單元6〇把該目標人物之人臉輪廓信息與 該記憶單元50存儲之所有人臉輪廓信息比較。由於不同之 人具有不同之人臉輪廓信息，若搜尋單元6〇從記憶單元5〇 搜尋與此人相同之人臉輪廓信息，則將該目標人物之監視 錄相藉由顯示單元30顯示出來。以達到從記憶單元5〇存 儲之影像中快速查詢目標人物是否出現過之目的。 S110 :顯示查尋結果。 優選地，於步驟S1〇4與步驟S1〇6之間進一步包括一 步驟：當沒有識別到人臉時，對所擷取之影像進行壓縮 理。 ' 當人臉識別單元40沒有識別到人臉時（也即沒有人出 現於監控區域内時）’該壓縮控制單元70啟動該壓縮單元 8〇對影像進行壓縮處理。反之，當人臉識別單元4〇識別到 人臉時，該壓縮控制單元70關閉該壓縮單元8〇工作。如 此，可減小記憶體，進一步達到快速查找之目的。 該人臉識別單元從該擷取之影像中識別所有人之人 臉，並對應各人臉分別生成一人臉輪廓信息，該記憶單元 記憶人臉識別單元識別之人臉輪廓信息，搜錄單元可於對 目標人物攝像，人臉朗單元生叙該人之人臉作自 於記憶單元内搜尋是否有與切所記憶之人臉輪^ 心相同之人臉，如此於所記憶之影像中可快速查找具 同之人臉信息之人臉。 综上所述，本發明確已符合發明專利之要件， 提出專利申請。惟，以上所述者料本發明之較佳實: 11 201003578 式，自不能以此限制本案之申請專利範固 技敲之人4" Μ 举1凡熟悉本案 L之人外依本發明之精神所作之等效修飾或變化 應涵蓋於以下申請專利範圍内。 【圖式簡單說明】 圖1係本發明提供之監控系統之功能模組示意圖； 圖2係圖1中監控系統之攝像單元之攝像裝置之功能 模組示意圖； 圖 圖3係本發明提供之監控方法流程 【主要元件符號說明】 監控系統 100 攝像單元 10 轉送單元 20 顯示單元 30 人臉識別單元 40 記憶單元 50 搜尋單元 60 攝像裝置 19 鏡頭單元 11 感光單元 12 壓縮控制單元 70 壓縮單元 80 12201003578 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a monitoring system and a monitoring method thereof. [Prior Art]: The unified monitoring system mode is composed of multiple monitors and video recorders. The Supervisor (4) _-type by ^ Supervisor::, the video was taken by video transmission: The surveillance screen set by the video display video, and the video of the surveillance screen was stored in the videotape. In this way, you can immediately monitor the effects of low crime. Like 'to achieve the maintenance of law and order:: and 'in order to find out the evidence of the suspect's case, you must find out, the image of the image with the memory of the shadow gradually check to see if the suspect appears and has done the case' Find evidence. This will take a lot of time to find. At the same time, manual inspection, if the viewing time is too long, due to fatigue, it is easy to make mistakes. The suspects are omitted from the surveillance video. SUMMARY OF THE INVENTION In view of the above, it is necessary to provide a monitoring system and a monitoring method thereof for quickly inquiring an image of a target person from a recorded image. A monitoring system, comprising: a camera unit; a face recognition unit, configured to identify a face of all people in the image captured by the camera unit, and generate a face contour information corresponding to each face, Recalling a single image for memorizing the image captured by the camera unit and the face contour information generated by the face recognition unit of the person 2010 201003578; a search unit for searching for the target person in the memory unit for the previous memory The face with the same face contour information; one is not only used to display the image captured by the camera unit and the search result. A monitoring method includes the following steps: (1) capturing an image; (2) identifying a face from the captured image, and generating a face face temple information corresponding to different faces; (3) capturing the captured image and The face recognition unit generates face contour information; (4) whether a target person query is the same as the previously recorded face contour information; (5) displaying the search result. Compared with the prior art, the face recognition unit recognizes the faces of all the people from the captured image, and generates a face contour information corresponding to each face, and the § recall unit § recalls the face recognition unit identification The face contour information, the search unit can search for the target person, the face recognition information generated by the face recognition unit, and search the memory unit for the same person as the previously recorded face contour information. Face, so you can quickly find faces with the same face information in the images you remember. [Embodiment] Hereinafter, the present invention will be further described in detail with reference to the accompanying drawings. Referring to FIG. 1 and FIG. 2 together, a monitoring system 7 201003578 100 according to the present invention includes a camera unit 10, a transfer unit 20, a display unit 30, a face recognition unit 40, a memory unit 50, and A search unit 60. The camera unit 10 is used to capture an image of a subject. The transfer unit 20 is configured to transfer the image captured by the image pickup unit 10 to the face recognition unit 40 and the display unit 30. The face recognition unit 40 identifies a face of the person from the captured image, and generates a face contour information corresponding to each face, and the memory unit 50 memorizes the image captured by the camera unit 10 and the face recognition unit. 40 generated face contour information. The search unit 60 can search the memory unit 50 for the target person to have the same face as the previously recorded face contour information. The display unit 30 can display the result of the search unit search and display the captured image. The camera units 10 can each include a plurality of cameras 19 disposed in different monitoring areas. Each camera device 19 includes a lens unit 11 and a sensing unit 12. The lens unit 11 is a lens module that includes at least one optical lens. The lens unit 11 is configured to collect optical signals and transmit the optical signals to the sensing unit 12. The sensing unit 12 is a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). The sensing unit 12 is configured to convert the optical signal concentrated by the lens unit 11 into an image signal. In this embodiment, the transmitting unit 20 is an optical fiber for transmitting the image signals captured by the camera unit 10 to the face recognition unit 40 and the display unit 30, respectively. The display unit 30 can include a plurality of displays, and instantly capture an image of the captured object captured by the image capturing unit 10 transmitted by the transfer unit 20 when the image is captured using the image capturing unit 8 201003578 10 . In an actual application, when the camera unit 10 has only one camera device 19 and the display unit 30 has only one display, the image signal captured by the camera unit 10 can also be directly transmitted to the display unit 30 without being transferred by the transfer unit 2 lose. The face recognition unit 40 recognizes the faces of all persons in the image captured by the image pickup unit 1 and generates a face contour information corresponding to each face. The memory unit 50 is configured to memorize the image captured by the at least one camera unit and the face contour information generated by the face recognition unit 40. The search unit 60 can search for the target person in the memory unit 4 to see if it is the same as the face profile information previously memorized. In a specific use, if a target person needs to be queried, the camera unit 10 images the target person, and at this time, the face recognition unit 40 simultaneously generates the face profile information of the person, the search unit 60, the person of the target person. The face contour information is compared with all face contour information stored by the token unit 50. Since different people have different face contour information', if the search unit 60 searches for the same face contour information from the memory unit 5, the monitoring record of the target person is displayed by the display unit 30. In order to quickly query whether the target person has appeared from the image stored in the memory unit 5〇. The hi-monitoring system 100 further includes a compression control unit and a compression unit 80. The compression control unit 70 is configured to control whether the compression unit 80 compresses the image captured by the camera unit 10 to save memory 9 201003578. When using 'When face recognition 罝—/ , _ , face % 7L 4〇 does not recognize the face (that is, no one appears in the monitoring area. L·· XN Bf ), the compression control unit 7〇 The compression unit 8G is activated to perform the image reduction processing; otherwise, when the face recognition unit 40 recognizes the human face, the compression control unit % closes the compression unit 80 to operate. In this way, you can, 丨, 兮 触 touch t J into j. It has been hidden to further achieve the purpose of quick search. Referring to FIG. 3, the monitoring method of the present invention includes the following steps: S102: capturing an image; the lens unit u of the camera unit converges the optical signal of the object, and transmits the optical signal to the photosensitive unit 12, The light sensing unit 12 converts the optical signal concentrated by the lens unit 11 into an image signal. S104: Recognizing a face from the captured image, and generating a face contour information corresponding to different faces; the face recognition unit 40 identifies a face of the person from the image captured by the camera unit 10 The face generates a face contour information. S106: the captured image and the facial contour information generated by the face recognition unit; the memory unit 50 stores the image captured by the at least one imaging unit 1 and each facial contour information generated by the face recognition unit 40. S108: Querying whether a target person is the same as the previously recorded face rim information; if a target person needs to be queried, the camera unit 10 images the target person, and the face recognition unit 40 simultaneously generates the Person's Person 201003578 The face gallery information 'search unit 6' compares the face contour information of the target person with all face profile information stored by the memory unit 50. Since different people have different face contour information, if the search unit 6 searches for the same face contour information from the memory unit 5〇, the monitoring video of the target person is displayed by the display unit 30. In order to quickly find out whether the target person has appeared from the image stored in the memory unit 5〇. S110: Display the search result. Preferably, a step is further included between the step S1〇4 and the step S1〇6: when the human face is not recognized, the captured image is compressed. When the face recognition unit 40 does not recognize the face (i.e., when no one appears in the monitoring area), the compression control unit 70 activates the compression unit 8 to perform compression processing on the image. On the other hand, when the face recognition unit 4 recognizes the face, the compression control unit 70 closes the compression unit 8 to operate. In this way, the memory can be reduced to further achieve the purpose of quick search. The face recognition unit identifies a face of the person from the captured image, and generates a face contour information corresponding to each face, the memory unit memorizing the face contour information recognized by the face recognition unit, and the search unit can In the case of the target person, the human face is said to be the face of the person from the memory unit to search for the same face as the face of the person who remembers the face, so that it can be quickly seen in the recorded image. Find faces with the same face information. In summary, the present invention has indeed met the requirements of the invention patent and filed a patent application. However, the above description is based on the preferred embodiment of the present invention: 11 201003578, which cannot limit the patent application of the patent in this case. 4" 举 1 1 Whoever is familiar with the case L is in accordance with the spirit of the present invention Equivalent modifications or variations are intended to be included within the scope of the following claims. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of functional modules of a monitoring system provided by the present invention; FIG. 2 is a schematic diagram of functional modules of an imaging device of an imaging unit of the monitoring system of FIG. 1; FIG. 3 is a monitoring provided by the present invention Method Flow [Main Component Symbol Description] Monitoring System 100 Camera Unit 10 Transfer Unit 20 Display Unit 30 Face Recognition Unit 40 Memory Unit 50 Search Unit 60 Camera Unit 19 Lens Unit 11 Photosensitive Unit 12 Compression Control Unit 70 Compression Unit 80 12