201138465 六、 [0001] [0002] [0003] [0004] [0005] [0006] 發明說明： 【發明所屬之技術領域】 & 本發明涉及"·種攝像裝置及利用其m㈣定物體的 方法。 【先前技彳舒】 傳統的攝像裝置需要監看人員隨時關注監控區域的畫面 ，當不明物體出現在畫面中時’發看：員再採取相應的 防衛措施，通知監控區域的保安人貝則主處理。但是’ 如果監看人員忽略了晝面中的異常物體，將給安全監控 工作帶來隱患。而且，傳統的攝像裝置無法對㈣到的 特定物體進行動態追蹤。 【發明内容】 赛於以上内容，有必要提供-種攝像裝置及利用該攝像 裝置動態追蹤特定物體的方法，0根據預先設置的移 動路徑表，對偵測到的特定物體進行動態追縱’並自動 調整攝像裴置的偵測靈敏度。 -種攝像I置，用於對特定物禮捕動態追踉’β亥攝像 裝置透過網路與監控區域巾其他攝#·裝置建立點#點連 接，該攝像裝置包括： 圖像獲取單元，用於獲取監控區域的圖像； 物體偵測單元，用於利用物體偵測技術，從獲取的圖片 中識別特定物體； 動態追蹤單元，用於當物體偵測單元從獲取的圖片中識 別出特定物體後，從該攝像裝置的儲存裝置中讀取預先 099112877 表單編號Α0101 第4頁/共27頁 0992022800-0 [0007] 201138465 設置的移動路徑表，查詢該移動路徑表，獲取該監控區 域中下一個能識別到該特定物體的攝像裝置；及 [0008] 所述動態追蹤單元，還用於調整下一個攝像裝置的偵測 靈敏度。 [0009] —種利用攝像裝置動態追蹤特定物體的方法，該攝像裝 置透過網路與監控區域中其他攝像裝置建立點對點連接 ，該方法包括如下步驟： [0010] 透過該攝像裝置獲取監控區域的圖像； [0011] 利用物體偵測技術，從獲取的圖片中識別特定物體； [0012] 當從獲取的圖片中識別出特定物體後，從該攝像裝置的 儲存裝置中讀取預先設置的移動路徑表，查詢該移動路 徑表，獲取該監控區域中下一個能識別到該特定物體的 攝像裝置；及 [0013] 調整下一個攝像裝置的偵測靈敏度。 [0014] 相較於習知技術，所述的攝像裝置及利用其動態追蹤特 定物體的方法，利用預先設置的移動路徑表，對偵測到 的特定物體進行動態追蹤，並自動調整攝像裝置的偵測 靈敏度，提高了安全監控工作的效率和精確度。 【實施方式】 [0015] 參閱圖1所示，係本發明監控系統較佳實施方式的網路架 構圖。在本實施方式中，該監控系統12包括多台攝像裝 置A-K (即A-E、F-K)、網路（Network) 14和用戶端 電腦16。其中，該多台攝像裝置位於一個監控區域内， 099112877 表單編號A0101 第5頁/共27頁 0992022800-0 201138465 透過網路14以點對點方式建立連接，以對該監控區域進 行電子監控。用戶端電腦16可以透過網路14登入其中一 台攝像裝置（如攝像裝置A)的設定頁面，進行各項資料 的設置。當各項資料都在攝像裝置A中設置完畢後，攝像 裝置A再以點對點方式將設置的各項資料傳輸至其他攝像 裝置，實現所有攝像裝置的資料同步。網路14可以是企 業内部網路（Intranet )，也可以是網際網路（Internet) 或其他類型的通訊網路。 [0016] 參閱圖2所示，係本發明攝像裝置較佳實施方式的結構方 框圖。在本實施方式中，該攝像裝置（如攝像裝置A)包 括資料設置單元121、圖像獲取單元122、物體偵測單元 123、動態追蹤單元124、儲存裝置125和處理器126。所 述攝像裝置包括，但不限於，可由軟體或硬體電路方式 驅動的巡轉臺攝影機、快速球攝影機（SpeedDome)和 可平移（Pan)、傾斜（Tilt)、縮放（Zoom)的PTZ ( Pan/Tilt/Zoom)攝影機等。所述處理器126用於控制 上述單元121至124的執行，完成對特定物體的動態追蹤 〇 [0017] 其中，所述資料設置單元121用於設置移動路徑表及監控 參數。具體步驟如下所述。 [0018] ( 1 )用戶端電腦16透過網路14登入該監控區域的一台攝 像裝置（如攝像裝置A )的設定頁面。 [0019] ( 2 )透過該設定頁面上傳該監控區域的電子地圖，並將 該電子地圖儲存在儲存裝置125中。參閱圖3所示，是該 099112877 表單編號A0101 第6頁/共27頁 0992022800-0 201138465 [0020] [0021] Ο Ο [0022] 監控區域的電子地圖18的示意圖’其中，區塊丨至區塊* 為不可通行區域，庫房41為監控物件。 (3) 在該電子地圖18上標示所有攝像裝置的安裝位置與 IP位址，並將每個攝像裝置的安裝位置與Ιρ位址儲存在 儲存裝置125中。繼續參閱圖3所示，電子地圖18上標示 的攝像裝置有：A、B、C、D、E、F、G、H、I、J、k ( 即圖1中的多台攝像裝置）^ (4) 根據電子地圖18上標示的每個攝像袭置的位置關係 ，設置移動路徑表，並將該移動路徑表儲存在儲存裝置 125中。在本實施方式中，所述移動路徑表包括兩列，其 中一列記錄電子地圖18上的所有攝像裝置，另外一列記 錄與每個攝像裝置相鄰的攝像裝置。參閱圖4所示，是根 據圖3中的電子地圖18設置的移動路徑表。第一列（Cam_ era ID)記錄有電子地圖18上的所有攝像裝置：a、B、 C、D、E、F、G、Η、I、J、K ;第二列（Next Camera ID)記錄有與每個攝像裝置相鄰崎辨舉果置。舉例而言 ，與攝像裝置Α相鄰的攝像裝置為辦像裝置Β和D，與攝像 裝置B相鄰的攝像裝置為攝像裝置A、F和C。 (5) 在該登入的攝像裝置（即攝像裝置A)的設定頁面 中設置監控參數，並將該監控參數儲存在該攝像裝置A的 儲存裝置125中。在本實施方式中，所述監控參數包括： 物體偵測技術的類型、資料同步間隔時間和管理人員的 資料等。所述物體偵測技術的類塑包括，但不限於：人 臉偵測技術、車牌偵測技術和行人偵測技術等°所述管 理人員的資料包括：管理人員的電話號碼和郵件地址等 099112877 表單編號A0101 第7頁/共27頁 0992022800-0 201138465 [0023] [0024] [0025] [0026] (6)根據電子地圖18上其他攝像裝置的Ip位址該登入 的攝像裝置（即攝像裝置A)透過網路14以點對點方式， 將上述設置的資料傳輸給該監控區域的其他攝像裝置， 完成資料的同步。上述設置的資料包括：電子地圖、每 個攝像裝置的安裝位置與1P位址、移動路#表和監控參 數等。 另外，每隔設定的資料同步間隔時間，攝像裳置A透過綱 路丨4進行一次資料同步的操作，如果發現電子地圖18上 存在無法進行資料同步的攝像裝專，則可透過短信或郵 件的方式告知管理人員。進^步地，用戶寸以根據需要 衣该#動路徑表中新增一台攝像Μ置、刪除或修改已有 的攝像裝置，同時修改各攝像裝置的相鄰關係，然後將 修改後的移動路徑表傳送給電子地圖18上的其他攝像骏 壤，β實現資料同步。 " % 所述圚像獲取單元122用於獲取監控區域的圖像。在本實 施方式中，所述圖像獲取單元122為電荷耦合裝置（ Charge Coupled Device， CCD) 〇 所述物體偵測單元123用於利用物體偵測技術，從獲取的 圖片中識別特定物體2〇。在本實施方式中，所述特定物 體20包括人臉、車牌和行人等。舉例而言，如果$署 夏^的 物體偵測技術的類型為人臉偵測技術，則所述物體侦測 單元123可以利用膚色理論（如YCbCr色彩空間模型）從 獲取的圖片中識別臉部，也可以利用臉部特徵樣本比對 099112877 表單編號A0101 第8頁/共27頁 0992022800-0 201138465 技術（Face Template Matching)或其他面部識別技 術從獲取的圖片中識別臉部。 [0027] 所述動態追蹤單元124用於當物體偵測單元123從獲取的 圖片中識別出特定物體20後，從儲存裝置125中讀取預先 設置的移動路徑表，查詢該移動路徑表，獲取下一個可 能識別到該特定物體20的攝像裝置。在本實施方式中， Ο 下一個可能識別到該特定物體20的攝像裝置為：與當前 攝像裝置相鄰的所有攝像裝置中的任意一個攝像裝置。 參閱圖5所示，假設特定物體20的位置在攝像裝置A處， 查詢圖4中的移動路徑表可知，與攝像裝置A相鄰的攝像 裝置為攝像裝置B和D，所以，下一値可能識別到該特定 物體20的攝像裝置為B或])。 [0028] 〇 所述動態追蹤單元124還用於調整下一個攝像裝置的偵測 靈敏度。在本實施方式中，所述偵測靈敏度是指：識別 出的特定物體20的面積最小值，所述調整下一個攝像裝 置的偵測靈敏度是指：將識別出的特定物體20的面積最 小值調小。例如，假設原先的彳貞/則靈敏度設定為5 X 5 :如 果識別出的特定物體20面積小於5x5，則物體偵測單元 123忽略該特定物體20，判定沒有識別出特定物體20 ;如 果識別出的特定物體20面積大於或等於5x5，則物體偵測 單元123不再忽略該特定物體20，判定識別出特定物體20 。假設調整後的偵測靈敏度設定為2x2 :如果識別出的特 定物體20小於2x2，則物體偵測單元123忽略該特定物體 20，判定沒有識別出特定物體20 ;如果識別出的特定物 體20面積大於或等於2x2，則物體偵測單元123不再忽略 099112877 表單編號A0101 第9頁/共27頁 0992022800-0 201138465 該特定物體20，判定識別出特定物體20。 [0029] 在其他實施方式中，所述動態追蹤單元124還用於記錄下 特定物體20的移動路徑，並將該特定物體20的移動路徑 儲存在所述儲存裝置125。進一步地，所述動態追蹤單元 124還可以發送短信或電子郵件至預先設置的管理人員， 讓其進行處理。更進一步地，所述動態追蹤單元124還可 以調整下一個攝像裝置的鏡頭方向，使其對準該識別出 的特定物體20。 [0030] 參閱圖6所示，假設特定物體20的位置從攝像裝置A處移 動至攝像裝置B處，查詢圖4中的移動路徑表可知，與攝 像裝置B相鄰的攝像裝置為攝像裝置A、F和C，所以，下 一個可能識別到該特定物體20的攝像裝置為A、F或C。動 態追蹤單元124將調整攝像裝置為A、F和C的偵測靈敏度 。在圖5至圖7中，實線箭頭代表特定物體20的移動路徑 ，虛線箭頭代表特定物體20下一次可能移動的路徑。 [0031] 參閱圖7所示，假設特定物體20的位置繼續從攝像裝置B 處移動至攝像裝置F處，查詢圖4中的移動路徑表可知， 與攝像裝置F相鄰的攝像裝置為攝像裝置B、G和Η，所以 ，下一個可能識別到該特定物體20的攝像裝置為Β、G或Η 。動態追蹤單元124將調整攝像裝置為Β、G和Η的偵測靈 敏度。由於攝像裝置Η處於監控物件庫房41附近，因此， 動態追蹤單元124還可以發出警報，以提醒管理人員注意 〇 [0032] 參閱圖8所示，係本發明利用攝像裝置對特定物體20的動 099112877 表單編號Α0101 第10頁/共27頁 0992022800-0 201138465 [0033] [0034] [0035] [0036] ❹ [0037] Ο [0038] [0039] 態追蹤方法的較佳實施方式的流程圖。 步驟S1，資料設置單元設置移動路徑表及監控參數。具 體過程參見圖9的描述。 步驟S2，圖像獲取單元122獲取監控區域的圖像。 步驟S3，物體偵測單元123利用物體偵測技術，從獲取的 圖片中識別特定物體20。在本實施方式申，所述特定物 體20包括人臉、車牌和行人等。 步驟S4，物體偵測單元123判斷是否從獲取的圖片中識別 出特定物體20。如果從獲取的圖片中識別出特定物體2〇 ，執行步驟S5 ;如果沒有從獲取的圖片中識別出特定物 體20，返回步驟S2。 步驟S5，動態追蹤單元124從儲存裝置125中讀取預先設 置的移動路徑表，查詢該移動路徑表，獲取下一個可能 識別到該特定物體20的攝枣舞置匕在本實施方式中，下 一個可能識別到該特定物體2〇的攝像裝置為：與當前攝 像裝置相鄰的所有攝像裝置中的任意一個攝像I置。 步驟S6，動態追蹤單元124調整下一個攝像裝置的偵測靈 敏度。 在其他實施方式中，所述動態追蹤單元124還用於記錄特 定物體20的移動路徑，並將該特定物體20的移動路徑儲 存在儲存裝置125中。進一步地，所述動態追蹤單元124 還可以發送短信或電子郵件至預先設置的管理人員，讓 其進行處理。更進一步地，所述動態追縱單元124還可以 099112877 表單編號Α0101 第11頁/共27頁 0992022800-0 201138465 調整下一個攝像裝置的鏡頭方向，使其對準該識別出的 特定物體2〇。 [0040] [0041] [0042] [0043] [0044] [0045] 099112877 參閱圖9所示’是圖8中步驟SI的具體流程圖。 乂驟31〇 ’用戶端電腦16透過網路Η登入該監控區域的— 台攝像裝置（如攝像裝置A)的設定頁面。 步驟sii，透過該設定頁面上傳該監控區域的電子地圖18 並將遠電子地圖18儲存在攝像裝置A的儲存裝置125中 步驟Sl2，在該電子地圖18上標示所有攝像裝置的安裝位 置與卟位址，並將每個攝像襄置的安裝隹置與ιρ位址儲 存在所述儲存裝置125中。 步驟S13，根據電子地圖18上標示的每個攝像裝置的位置 關係’設置移動路徑表’並將該移動路徑表儲存在健存 裝置124。在本實施方式中，所述移動路徑表包括兩列 ，其令—列記錄電子地圖〗δ上的所有攝像裝置另外一 列記錄與每個攝像裝置相鄰翁健置（參閱圖4所示） 同時’用戶可以根據需要在該移動路徑表中新增一& 攝像裝置、刪除或修改已有的攝像裳置，同時“： 像裝置的相鄰關係，^後將修改後的移動路徑 電子地圖18上的其他攝像裝置，以實現資料同步。、° 芡騍，隹該登入的攝像裝 百而由-罟贮私么 娜1豕衮置A)的設定 頁面中认置皿控參數’並將該監控參數儲存在儲存裝置 !25中。在本實施方式中，所述監控參數包括：物體 技術的類型、資料同步間隔時間和管理 … 表單編號Α0ΗΗ 第12頁/共27頁 貞科寺。 0992022800-0 201138465 所述物體偵測技術的類型包括，但不限於：人臉偵測技 術、車牌彳貞測技術和行人彳貞測技術等。所述管理人員的 資料包括：管理人員的電話號碼和郵件地址等。 [0046] ❹ [0047] ❹ [0048] [0049] [0050] [0051] [0052] 099112877 步驟S15，根據電子地圖18上其他攝像裝置的IP位址，該 登入的攝像裝置透過網路14以點對點方式，將上述設置 的資料傳輸給該監控區域的其他攝像裝置，完成資料的 同步。上述設置的資料包括：電子地圖、每個攝像裝置 的安裝位置與IP位址、移動路徑表和監控參數等。另外 ，每隔設定的資料同步間隔時間，攝像裝置A透過網路14 進行一次資料同步的操作，如果發現電子地圖18上存在 無法進行資料同步的攝像裝置，則可透過短信或郵件的 方式告知管理人員。 最後應說明的是，以上實施方式僅用以說明本發明的技 術方案而非限制，儘管參照較佳實施方式對本發明進行 了詳細說明，本領域的普通技術人員應當理解，可以對 本發明的技術方案進行修改或等同替換，而不脫離本發 明技術方案的精神和範圍。 【圖式簡單說明】 圖1係本發明監控系統較佳實施方式的網路架構圖。 圖2係本發明攝像裝置較佳實施方式的結構方框圖。 圖3係某監控區域的電子地圖示意圖。 圖4係根據圖3中的電子地圖設置的移動路徑表示意圖。 圖5至圖7係根據圖4中的移動路徑表進行動態追蹤的示意 圖。 表單編號A0101 第13頁/共27頁 0992022800-0 201138465 [0053] 圖8係本發明利用攝像裝置動態追蹤特定物體的方法的較 佳實施方式的流程圖。 [0054] 圖9係圖8中步驟S1的具體流程圖。 【主要元件符號說明】 [0055] 監控系統：12 [0056] 攝像裝置：A-K [0057] 網路：14 [0058] 用戶端電腦：1 6 [0059] 電子地圖：18 [0060] 庫房：41 [0061] 特定物體：20 [0062] 資料設置單元： 121 [0063] 圖像獲取單元： 122 [0064] 物體偵測單元： 123 [0065] 動態追蹤單元： 124 [0066] 儲存裝置：125 [0067] 處理器：126 099112877 表單編號A0101 第14頁/共27頁 0992022800-0[Technical Field] The present invention relates to an image pickup apparatus and a method of using the same (m) . [Previous technology] The traditional camera device requires the monitor to pay attention to the screen of the monitoring area at any time. When an unknown object appears in the screen, the sender will take corresponding defense measures to notify the security guard in the monitoring area. deal with. However, if the monitor ignores the abnormal objects in the face, it will bring hidden dangers to the safety monitoring work. Moreover, the conventional camera device cannot dynamically track the specific object to (4). SUMMARY OF THE INVENTION In the above content, it is necessary to provide a camera device and a method for dynamically tracking a specific object by using the camera device, and 0 dynamically crawling the detected specific object according to a preset movement path table. Automatically adjusts the detection sensitivity of the camera. - A kind of camera I set, used for the special object ritual dynamic tracking 'β Hai camera device through the network and the monitoring area towel other camera # · device establishment point # point connection, the camera device includes: image acquisition unit, with Obtaining an image of a monitoring area; an object detecting unit for identifying a specific object from the acquired image by using an object detecting technology; and a dynamic tracking unit for identifying the specific object from the acquired image by the object detecting unit After that, read from the storage device of the camera device, the previous 099112877 form number Α 0101 page 4 / total 27 page 0992022800-0 [0007] 201138465 set the movement path table, query the movement path table, obtain the next one in the monitoring area An imaging device capable of recognizing the specific object; and [0008] the dynamic tracking unit is further configured to adjust a detection sensitivity of the next imaging device. [0009] A method for dynamically tracking a specific object by using an imaging device, the camera device establishing a point-to-point connection with other camera devices in the monitoring area through a network, the method comprising the following steps: [0010] acquiring a map of the monitoring area through the camera device [0011] using the object detection technology to identify a specific object from the acquired image; [0012] after identifying the specific object from the acquired image, reading a preset movement path from the storage device of the camera device a table, querying the moving path table, acquiring a next camera device in the monitoring area that can identify the specific object; and [0013] adjusting the detection sensitivity of the next camera device. [0014] Compared with the prior art, the camera device and the method for dynamically tracking a specific object use the preset movement path table to dynamically track the detected specific object and automatically adjust the camera device. Detection sensitivity improves the efficiency and accuracy of safety monitoring. [Embodiment] [0015] Referring to Figure 1, there is shown a network architecture of a preferred embodiment of the monitoring system of the present invention. In the present embodiment, the monitoring system 12 includes a plurality of imaging devices A-K (i.e., A-E, F-K), a network 14 and a client computer 16. Wherein, the plurality of camera devices are located in a monitoring area, 099112877 Form No. A0101 Page 5 of 27 0992022800-0 201138465 A connection is established in a point-to-point manner through the network 14 to electronically monitor the monitored area. The client computer 16 can log in to the setting page of one of the camera devices (such as the camera device A) via the network 14 to set various materials. After all the data are set in the camera device A, the camera device A transmits the set data to other camera devices in a point-to-point manner to realize data synchronization of all the camera devices. The network 14 can be an intranet or an internet or other type of communication network. [0016] Referring to Figure 2, there is shown a block diagram of a preferred embodiment of an image pickup apparatus of the present invention. In the present embodiment, the image pickup apparatus (such as the image pickup apparatus A) includes a material setting unit 121, an image acquisition unit 122, an object detection unit 123, a dynamic tracking unit 124, a storage device 125, and a processor 126. The camera device includes, but is not limited to, a patrol camera that can be driven by a software or a hardware circuit, a SpeedDome, and a Pan, a Tilt, and a Zoom PTZ (Pan). /Tilt/Zoom) Cameras, etc. The processor 126 is configured to control the execution of the units 121 to 124 to complete dynamic tracking of a specific object. [0017] The data setting unit 121 is configured to set a movement path table and monitoring parameters. The specific steps are as follows. [0018] (1) The client computer 16 logs into the setting page of a camera device (such as the camera device A) in the monitoring area via the network 14. [0019] (2) uploading an electronic map of the monitoring area through the setting page, and storing the electronic map in the storage device 125. Referring to FIG. 3, it is the 099112877 Form No. A0101 Page 6 / 27 pages 0992022800-0 201138465 [0020] [0021] [0022] Schematic diagram of the electronic map 18 of the monitoring area, where the block reaches the area Block * is an impassable area, and warehouse 41 is a monitoring object. (3) The mounting position and IP address of all the image pickup devices are indicated on the electronic map 18, and the mounting position of each camera device and the Ιρ address are stored in the storage device 125. Continuing to refer to FIG. 3, the imaging devices indicated on the electronic map 18 are: A, B, C, D, E, F, G, H, I, J, k (ie, multiple cameras in FIG. 1)^ (4) According to the positional relationship of each camera shot indicated on the electronic map 18, a movement path table is set, and the movement path table is stored in the storage device 125. In the present embodiment, the movement path table includes two columns, one of which records all of the image pickup devices on the electronic map 18, and the other of which records the image pickup devices adjacent to each of the image pickup devices. Referring to Fig. 4, it is a movement path table set based on the electronic map 18 in Fig. 3. The first column (Cam_era ID) records all the camera devices on the electronic map 18: a, B, C, D, E, F, G, Η, I, J, K; the second column (Next Camera ID) record There is a distinction between each camera and the camera. For example, the image pickup devices adjacent to the image pickup device 为 are the image pickup devices Β and D, and the image pickup devices adjacent to the image pickup device B are the image pickup devices A, F, and C. (5) The monitoring parameter is set in the setting page of the registered imaging device (i.e., imaging device A), and the monitoring parameter is stored in the storage device 125 of the imaging device A. In this embodiment, the monitoring parameters include: the type of the object detection technology, the data synchronization interval, and the information of the manager. The types of object detection technologies include, but are not limited to, face detection technology, license plate detection technology, and pedestrian detection technology, etc. The management personnel's information includes: the telephone number and email address of the manager, etc. 099112877 Form No. A0101 Page 7 / Total 27 Page 0992022800-0 201138465 [0023] [0025] [6] According to the Ip address of other camera devices on the electronic map 18, the registered camera device (ie, the camera device) A) The above-set data is transmitted to other camera devices in the monitoring area through the network 14 in a point-to-point manner to complete data synchronization. The above settings include: electronic map, installation location and 1P address of each camera, mobile ## table and monitoring parameters. In addition, every time the data synchronization interval is set, the image capture device A performs a data synchronization operation through the route 丨4, and if it is found that there is a camera installation on the electronic map 18 that cannot synchronize data, the message can be sent through a text message or an email. Inform the manager. Step by step, the user inch adds a camera device to the #moving path table according to the need, deletes or modifies the existing camera device, and simultaneously modifies the adjacent relationship of each camera device, and then moves the modified device. The path table is transmitted to other camera mounts on the electronic map 18, and the data is synchronized by β. "% The key acquisition unit 122 is configured to acquire an image of the monitoring area. In the embodiment, the image acquisition unit 122 is a charge coupled device (CCD). The object detection unit 123 is configured to identify a specific object from the acquired image by using an object detection technology. . In the present embodiment, the specific object 20 includes a face, a license plate, a pedestrian, and the like. For example, if the type of the object detection technology of the device is a face detection technology, the object detection unit 123 can use the skin color theory (such as the YCbCr color space model) to identify the face from the acquired image. Faces can also be identified from the acquired pictures using face feature sample comparison 099112877 Form No. A0101 Page 8 / Total 27 Page 0992022800-0 201138465 Technology (Face Template Matching) or other facial recognition techniques. [0027] The dynamic tracking unit 124 is configured to: after the object detecting unit 123 identifies the specific object 20 from the acquired image, read the preset moving path table from the storage device 125, query the moving path table, and obtain The next camera device that may identify the particular object 20. In the present embodiment, the next image pickup device that may recognize the specific object 20 is any one of all the image pickup devices adjacent to the current image pickup device. Referring to FIG. 5, assuming that the position of the specific object 20 is at the image pickup apparatus A, it can be seen from the movement path table in FIG. 4 that the image pickup apparatuses adjacent to the image pickup apparatus A are the image pickup apparatuses B and D, so the next possible The imaging device that recognizes the specific object 20 is B or ]). [0028] The dynamic tracking unit 124 is further configured to adjust the detection sensitivity of the next camera. In the present embodiment, the detection sensitivity refers to the identified minimum area of the specific object 20, and the adjustment sensitivity of the next imaging device refers to the minimum area of the specific object 20 to be recognized. lower. For example, assume that the original sensitivity is set to 5 X 5 : if the identified specific object 20 area is less than 5x5, the object detecting unit 123 ignores the specific object 20, determines that the specific object 20 is not recognized; if it is recognized The specific object 20 area is greater than or equal to 5x5, and the object detecting unit 123 no longer ignores the specific object 20, and determines that the specific object 20 is recognized. It is assumed that the adjusted detection sensitivity is set to 2x2: if the identified specific object 20 is smaller than 2x2, the object detecting unit 123 ignores the specific object 20, and determines that the specific object 20 is not recognized; if the identified specific object 20 area is larger than Or equal to 2x2, the object detecting unit 123 no longer ignores 099112877 Form No. A0101 Page 9/Total 27 Page 0992022800-0 201138465 The specific object 20 determines that the specific object 20 is recognized. [0029] In other embodiments, the dynamic tracking unit 124 is further configured to record a moving path of the specific object 20 and store the moving path of the specific object 20 in the storage device 125. Further, the dynamic tracking unit 124 can also send a short message or email to a preset administrator for processing. Further, the dynamic tracking unit 124 can also adjust the lens direction of the next camera to be aligned with the identified specific object 20. [0030] Referring to FIG. 6 , it is assumed that the position of the specific object 20 is moved from the imaging device A to the imaging device B. For the query of the movement path table in FIG. 4 , the imaging device adjacent to the imaging device B is the imaging device A. , F and C, so the next camera device that may recognize the particular object 20 is A, F or C. The dynamic tracking unit 124 will adjust the detection sensitivity of the camera to A, F, and C. In FIGS. 5 to 7, the solid arrows represent the moving path of the specific object 20, and the dotted arrows represent the path that the specific object 20 may move next time. [0031] Referring to FIG. 7, it is assumed that the position of the specific object 20 continues to move from the imaging device B to the imaging device F. As can be seen from the movement path table in FIG. 4, the imaging device adjacent to the imaging device F is an imaging device. B, G, and Η, so the next camera device that may recognize the particular object 20 is Β, G, or Η. The dynamic tracking unit 124 will adjust the detection sensitivity of the camera to Β, G, and Η. Since the camera device Η is in the vicinity of the monitoring object storeroom 41, the dynamic tracking unit 124 can also issue an alarm to remind the manager of the 〇 [0032] Referring to FIG. 8, the present invention utilizes the camera device to move the specific object 20 to 099112877. Form No. 101 0101 Page 10 / Total 27 Page 0992022800-0 201138465 [0034] [0036] [0039] A flowchart of a preferred embodiment of the state tracking method. In step S1, the data setting unit sets the movement path table and the monitoring parameters. See the description of Figure 9 for a specific process. In step S2, the image acquisition unit 122 acquires an image of the monitoring area. In step S3, the object detecting unit 123 uses the object detecting technology to identify the specific object 20 from the acquired image. In the present embodiment, the specific object 20 includes a face, a license plate, a pedestrian, and the like. In step S4, the object detecting unit 123 determines whether or not the specific object 20 is recognized from the acquired picture. If the specific object 2 is recognized from the acquired picture, step S5 is performed; if the specific object 20 is not recognized from the acquired picture, the process returns to step S2. In step S5, the dynamic tracking unit 124 reads the preset movement path table from the storage device 125, queries the movement path table, and acquires the next jujube dance that may identify the specific object 20 in the present embodiment. An imaging device that may recognize the specific object 2〇 is: any one of all the imaging devices adjacent to the current imaging device. In step S6, the dynamic tracking unit 124 adjusts the detection sensitivity of the next camera. In other embodiments, the dynamic tracking unit 124 is further configured to record a moving path of the specific object 20 and store the moving path of the specific object 20 in the storage device 125. Further, the dynamic tracking unit 124 can also send a short message or email to a preset administrator for processing. Further, the dynamic tracking unit 124 may also be 099112877 Form No. 1010101 Page 11 of 27 0992022800-0 201138465 Adjust the lens direction of the next camera to be aligned with the identified specific object 2〇. [0044] [0045] [0045] 099112877 Referring to FIG. 9, ' is a specific flowchart of step S1 in FIG. 8. Step 31〇 The client computer 16 logs into the setting page of the camera device (such as camera device A) in the monitoring area through the network. Step sii, uploading the electronic map 18 of the monitoring area through the setting page and storing the remote electronic map 18 in the storage device 125 of the imaging device A in step S12, and marking the installation position and the position of all the imaging devices on the electronic map 18. And storing the installation device of each camera device and the address of the camera in the storage device 125. In step S13, the movement path table is set based on the positional relationship of each of the image pickup devices indicated on the electronic map 18, and the movement path table is stored in the storage device 124. In this embodiment, the movement path table includes two columns, which causes another column of all the image capturing devices on the column-recording electronic map δδ to be recorded adjacent to each camera device (see FIG. 4). 'Users can add a & camera device to the movement path table as needed, delete or modify the existing camera mount, and at the same time ": like the adjacent relationship of the device, ^ will be modified after the moving path electronic map 18 Other camera devices on the other to achieve data synchronization., ° 芡骒, 隹 登入 隹 登入 登入 登入 登入 登入 登入 隹 隹 认 认 认 认 认 认 认 认 认 么 么 么 么 么 么 么 么 么 认 认 认 认 认 认 认The monitoring parameters are stored in the storage device ! 25. In the present embodiment, the monitoring parameters include: type of object technology, data synchronization interval time and management... Form No. ΗΗ0ΗΗ Page 12 of 27 贞科寺. 0992022800- 0 201138465 The types of object detection technologies include, but are not limited to, face detection technology, license plate detection technology and pedestrian detection technology, etc. The management personnel's information includes: management personnel Telephone number, mail address, etc. [0046] ❹ [0048] [0050] [0052] [0052] Step S15, according to the IP address of other camera devices on the electronic map 18, the login The camera device transmits the set data to other camera devices in the monitoring area through the network 14 in a point-to-point manner to complete data synchronization. The set information includes: an electronic map, an installation location and an IP address of each camera device. And moving the path table, monitoring parameters, etc. In addition, the camera A performs a data synchronization operation through the network 14 every time the data synchronization interval is set, and if there is an image device on the electronic map 18 where data synchronization cannot be performed, The management may be informed by means of a short message or an email. The above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting, although the present invention is described in detail with reference to the preferred embodiments, The skilled person will understand that the technical solutions of the present invention may be modified or equivalently replaced without departing from the present invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a network architecture diagram of a preferred embodiment of a monitoring system of the present invention. Fig. 2 is a block diagram showing the structure of a preferred embodiment of the camera apparatus of the present invention. FIG. 4 is a schematic diagram of a movement path table set according to the electronic map in FIG. 3. FIG. 5 to FIG. 7 are schematic diagrams of dynamic tracking according to the movement path table in FIG. 4. Form No. A0101 Page 13 / Total 27 pages 0992022800-0 201138465 [0053] FIG. 8 is a flow chart of a preferred embodiment of the method for dynamically tracking a specific object using an imaging device of the present invention. 9 is a specific flowchart of step S1 in FIG. 8. [Main component symbol description] [0055] Monitoring system: 12 [0056] Camera: AK [0057] Network: 14 [0058] Client computer: 1 6 [0059] Electronic map: 18 [0060] Warehouse: 41 [ 0061] Specific object: 20 [0062] Data setting unit: 121 [0063] Image acquisition unit: 122 [0064] Object detection unit: 123 [0065] Dynamic tracking unit: 124 [0066] Storage device: 125 [0067] Processor: 126 099112877 Form No. A0101 Page 14 / Total 27 Page 0992022800-0