201220215 六、發明說明： 【發明所屬之技術領域】 [0001]本發明涉及一種監控系統及方法，尤其是關於一種可疑 目標識別及追蹤系統及方法。 【先前技術】 [〇〇〇2]隨著電腦系統的應用、普及和網路技術的迅速發展’透 過電子數碼產品如攝像頭及電腦網路實現遠端監控已被 越來越多的行業所採用。其中，尤其是帶自動追蹤功能 的攝像頭倍受歡迎。目前，帶自動追蹤功能的攝像頭一 般是根據可疑目標的形體特徵對可疑目標進行識別、追 蹤。然而，若可疑目標混入擁擠的人群，由於其他人物 的干擾，攝像頭不易根據形體特徵對可疑目標進行識別 、追蹤。 【發明内容】 [0003] 鑒於以上内容，有必要提供一種可疑目標識別及追縱系 統及方法，可以排除監控區域内其他人物的干擾，對可 疑目標進行識別及追蹤。 [0004]201220215 VI. Description of the Invention: [Technical Field of the Invention] [0001] The present invention relates to a monitoring system and method, and more particularly to a suspicious object recognition and tracking system and method. [Prior Art] [〇〇〇2] With the rapid development of the application, popularity and network technology of computer systems, remote monitoring through electronic digital products such as cameras and computer networks has been adopted by more and more industries. . Among them, especially cameras with automatic tracking are popular. Currently, cameras with automatic tracking are generally able to identify and track suspicious targets based on the physical characteristics of suspicious targets. However, if suspicious targets are mixed into a crowd, the camera is not easy to identify and track suspicious targets based on physical characteristics due to interference from other characters. SUMMARY OF THE INVENTION [0003] In view of the above, it is necessary to provide a suspicious target recognition and tracking system and method, which can eliminate interference of other characters in the monitoring area, and identify and track suspicious objects. [0004]
—種可疑目標識別及追蹤系統，應用於電腦《該電腦透 過信號線與雲台控制器相連接，該雲台控制器透過信號 線分別與紅外發光二極體、攝像頭及紅外光電管相連接 。該系統包括接收模組、選擇模組、目標追縱模級、命 令生成模組及判斷模組。接收模組，用於接收攝像頭， 採集的監控區域不同時刻的視頻圖像。選擇模組，用於 在—張視頻圖像中選擇一個運動的人物作為追縱目f 目標追蹤模組，用於根據可疑目標追蹤演算法確定H 099138366 表單蝙號A0101 第4頁/共20頁 201220215 Ο 蹤目標在不同時刻視頻圖像中的位置，根據該追蹤目標 在不同時刻視頻圖像令的位置計算該追蹤目標在監控區 域的移動方向及移動距離。命令生成模組，用於根據上 述計算結果產生控制命令，控制雲台控制器驅動攝像頭 向相應方向移動相應距離，使得該追踉目標始終處於攝 像頭的視角範圍内，並控制雲台控制器驅動紅外發光二 極體產生紅外線照射該追蹤目標。所述紅外光電管感測 所述紅外線穿透追蹤目標後的光強度變化，根據光強度 變化產生脈衝電信號，並透過信號線將該脈衝電信號傳 送給雲台控制器’雲台控制器將該滅衝電信號轉換為數 位信號後回傳給所述電腦。所述判斷模组，用於判斷該 數位彳§號的頻率疋否超過人的正常心跳頻率範圍，若超 過人的正常心跳頻率範圍，則認定該追蹤目標為可疑目 標，利用所述目標追蹤模組及命令生成模叙的功能繼續 追蹤該追蹤目標。 [0005] 一種可疑目標識別及追蹤芳法，應用於電腦。該電腦透 〇 過k號線與雲台控制器相連接，該雲台控制器透過信號 線分別與紅外發光二極體、攝像頭及紅外光電管相連接 。該方法包括：（a)接收攝像頭採集的監控區域的一張 視頻圖像，在該視頻圖像中選擇一個運動的人物作為追 蹤目標·，（b)根據可疑目標追蹤演算法確定該追蹤目ζ 在該視頻圖像中的位置；⑹根據可疑目標追蹤演算法 確疋该追蹤目標在下一張視頻圖像中的位置·，（d)根據 該追蹤目標在該兩張視頻圖像中的位置計算該追蹤目標 在監控區域的移動方向及移動距離；（e)根據上述計^ 099138366 表單編號A0101 第5頁/共20頁 201220215 結果產生控制命令，根據該控制命令控制雲台控制器驅 動攝像頭向相應方向移動相應距離，使得該追蹤目標始 終處於攝像頭的視角範圍内；（f)控制雲台控制器驅動 紅外發光二極體產生紅外線照射該追蹤目標，所述紅外 光電管感測所述紅外線穿透追蹤目標後的光強度變化， 根據光強度變化產生脈衝電信號，並透過信號線將該脈 衝電信號傳送給雲台控制器，雲台控制器將該脈衝電信 號轉換為數位信號後回傳給電腦；及（g)判斷該數位信 號的頻率是否超過人的正常心跳頻率範圍，若未超過人 的正常心跳頻率範圍，則停止追蹤該追蹤目標，流程結 束，若超過人的正常心跳頻率範圍，則認定該追蹤目標 為可疑目標，流程返回步驟（C )，繼續追蹤該追蹤目標。 [0006] 相較於習知技術，本發明所提供的可疑目標識別及追蹤 系統及方法，可以排除監控區域内其他人物的干擾，對 可疑目標進行識別及追蹤。 【實施方式】 [0007] 參閱圖1所示，係本發明可疑目標識別及追蹤系統100 ( 以下簡稱“系統100”）較佳實施例之應用環境圖。該系 統10 0應用於電腦10，該電腦10透過信號線與雲台控制器 20相連接。該雲台控制器20透過信號線分別與紅外發光 二極體30、攝像頭40及紅外光電管50相連接。需要指出 的是，圖1僅表示上述各電子裝置之間的電連接關係，並 不代表各電子裝置的空間位置關係。在本實施例中，紅 外發光二極體30可以安裝於攝像頭40正上方並與攝像頭 40綁定，雲台控制器20驅動攝像頭40運動時，紅外發光 099138366 表單編號A0101 第6頁/共20頁 0992066853-0 201220215 —極體30作與攝像頭40相同的運動。 [0008] 攝像頭40採集監控區域的視頻圖像，並透過雲台控制器 將採集得到的視頻圖像傳送至電腦i卜系統⑽分析所 述視頻圖像，根據一個追縱目標（如圖】中人物7〇)在不 5、見頻圖像中的位置，計异該追縱目標在監控區域的位 移貝訊，並根據該位移資訊發送控制命令至雲台控制器 雲台控制器20根據控制命令驅動攝像頭4〇運動對該 Ο [0009] 追蹤目標進行追蹤，使得該追縱目標始終處於攝像頭40 的視角範圍之内。 在驅動攝像頭40運動的同時，雲台控制器2〇驅動紅外發 光一極體30產生紅外線（如圖1中虛線所示）照射該追蹤 目標。紅外光電管5 〇感測強紅外線穿透追縱目標後的光 強度變化’根據光強度變化產生脈衝電信號，並透過信 號線將該脈衝電信號傳送給雲台控制器2〇。在本實施例 中’紅外光電管50安裝在監控區域内的地板6〇下方。在 ;： .:丨M;'.. , i '；：： ：；：; .：；, i： Ο [0010] 其他實施例中，紅外光電管5〇也可以夫裝在監控區域内 的侧面牆壁内。 隨著人體心臟的波動，人體組織的半透明度會發生改變 ’進而會引起穿過人體某一部位（例如耳垂）的紅外線 的光強度發生變化。人體組織的半透明度變化頻率與心 跳頻率—致，而穿過人體的紅外線的光強度變化頻率又 與人體組織的半透明度變化頻率一致。故透過穿過人體 的紅外線的光強度變化頻率可以得知運動人物的心跳頻 率。 099138366 表單編號A0101 第7頁/共20頁 0992066853-0 201220215 [0011] 紅外光電管50感測穿過Λ, %人體的紅外線的光強變化，並根 據光強變化產生脈衝電信號1台控髮_㈣㈣ 信號轉換為數位信號後傳送給電腦10。系統100判斷該數 位信號的頻率是否超過人的正常心跳頻率範圍，若超過 人的正常心跳頻率乾圍，則趣定該追蹤目標為可疑目標 ，繼續追蹤該追蹤目操。 $ [0012] 參閲圖2所示’係系統⑽較隹實施例之功能模組圖。亨 系統_包括接收模組⑴、選擇模議、目標造縱模 組130、命令生成模組Ι4ΰ、_模議及報警模/ 160 闕接收模組11〇用於接收攝像頭4〇採集的監控區域不同時刻 的視頻圖像。 [0014] [0015] 選擇模組1则於在—張視頻_巾_- 作為追蹤目標。例如，選擇 頻圖像中選擇-«動的錢作輕m 目標追縱独丨_練據顿目標料㈣法確定节追 縱目標在不同時刻視頻圖像中的位^n(xn，yn，Zd 本實施例中，該可疑目標追_算法為連續自適應平均 值遷移演算法（Continuously Apative Mean_Shift ’CamShift)。該演算法根據追蹤目標的顏色特徵在視 頻圖像中搜索追蹤目標’確定追縱目襟在視頻圖像中的 位置和大小。例如’在本實施例中，假設追蹤目標著裝 的主要顏色為深紅色’則目標追蹤楔奴13〇以包含所述深 紅色範圍的最小矩形為追蹤目標在刻視頻圖像中 099138366 表單編號A0101 第8頁/共20頁 0992066853-0 201220215 的大小’以該最小矩形的中心點的位置為追縱目標在t = 1 時刻視頰圖像中的位置pl(xi，yi，zi)。 [0016] 目標追縱模組1 30還用於根據該追蹤目標在不同時刻視頻 圖像中的位置計算該追蹤目標在監控區域的移動方向及 移動距離。例如，假設該追縱目標在t = 2s時刻視頻圖像 中的位置為p2(x2，y2，z2),則目標追蹤模組130計算 P2(x2，y2，z2)與Pi(xl，yl，zl)的差值，得到該追蹤目 標在監控區域的移動方向及移動距離。 0 [0017] 〇 [0018] [0019] 命令生成模組14〇用於根據上述計算結果產生控制命令’ 並發送該控制命令至雲台控制器20，雲台控制器20根據 該控制命令控制攝像頭40向相應方向移動相應距離，使 得該追蹤目標始終處於攝像頭40的視角範圍内。例如， 雲台控制器20根據該控制命令控制攝像頭40向X軸方向移 動（χ2-χ1)的距離，γ軸方向移動（y2-yi)的距離，z轴 方向移動（z2-zl)的距離。同時，雲台控制器20驅動紅 外發光二極體3 〇產生紅外線照射該追縱目標。紅外光電 管50感測紅外線穿透追鞭目標後的光強度變化，根據光 強度變化產生脈衝電信號，並透過信號線將該脈衝電信 號傳送給雲台控制器20。雲台控制器2〇將該脈衝電信號 轉換為數位信號。 接收模組110還用於接收雲台控制器20回傳的數位信號。 判斷模組15 0 #情雜純號㈣率是否超過人的正常心 跳頻率範圍，例如6G~1GG次/每分鐘。若該數位信號的頻 率超過人的正常心跳頻率範圍，則認定該追蹤目標為可 099138366 表單編號A0101 第9頁/共20頁 0992066853-0 201220215 疑目標，繼續追蹤該追蹤目標。 [0020] [0021] [0022] [0023] 報警模組1 6 G用於當該數位信號的頻率超過人的正常心跳 頻率範圍’開啟電腦的報警裝置，例如LED警示燈以通 知監控人員監控區域有可疑目標出現。 參閱圖3所示，係本發明可疑目標識別及追蹤方法較佳實 施例之流程圖。 步驟S101，接收模組110接收攝像頭4〇採集的監控區域 的一張視頻圖像（例如t = ls時刻的一張視頻圖像），選 擇模組120在該視頻圖像中選擇一俩運動的人物作為追蹤 目標。 步驟S103，目標追蹤模組13〇根據可疑目標追蹤演算法確 定該追蹤目標在該視頻圖像中的位置卩丨^】^，“）。在 本實施例中，該可疑目標追蹤演算法為連續自適應平均 值遷移演异法（Continuously Apative Mean-Shift ’ CamShift)。該演算法根據追蹤目標的顏色特徵在視 頻圖像中搜索追蹤目標，確定追縱目標在視頻圖像中的 位置和大小。例如，在本實施例中，假設追蹤目標著裝 的主要顏色為深紅色’則目標追蹤模組丨3〇以包含所述深 紅色範圍的最小矩形為追蹤目標在t = s〗時刻視頻圖像中 的大小，以該最小矩形的中心點的位置為追蹤目標在 t = sl時刻視頻圖像中的位。 步驟S105’目標追蹤模組130根據可疑目標追蹤演算法確 定該追蹤目標在下一張視頻圖像（例如t = 2s時刻的一張 視頻圖像）中的位置1)2(\2，卩2,22)。具體方法與步驟 099138366 表單編號A0101 第10頁/共20頁 0992066853-0 [0024] 201220215 [0025] [0026] ❹ [0027]A suspicious target recognition and tracking system for computers. The computer is connected to the pan/tilt controller via a signal line. The pan/tilt controller is connected to the infrared light-emitting diode, the camera and the infrared photocell through signal lines. The system includes a receiving module, a selection module, a target tracking module level, a command generation module, and a determining module. The receiving module is configured to receive the camera and collect video images of the monitoring area at different times. The selection module is used to select a moving character in the video image as the target tracking module, and is used to determine the H099138366 form bat number A0101 according to the suspicious target tracking algorithm. Page 4 of 20 201220215 The position of the target in the video image at different moments is tracked, and the moving direction and the moving distance of the tracking target in the monitoring area are calculated according to the position of the video image at different times. The command generation module is configured to generate a control command according to the calculation result, and control the PTZ controller to drive the camera to move the corresponding distance to the corresponding direction, so that the tracking target is always within the viewing angle range of the camera, and control the PTZ controller to drive the infrared The light-emitting diode generates infrared rays to illuminate the tracking target. The infrared photocell senses a change in light intensity after the infrared ray penetrates the tracking target, generates a pulse electrical signal according to the change of the light intensity, and transmits the pulse electrical signal to the pan-tilt controller via the signal line. The extinguishing electrical signal is converted to a digital signal and transmitted back to the computer. The determining module is configured to determine whether the frequency of the digital number exceeds a normal heartbeat frequency range of the person, and if the normal heartbeat frequency range of the person is exceeded, determining that the tracking target is a suspicious target, and using the target tracking mode The group and command generation features continue to track the tracking target. [0005] A suspicious target recognition and tracking method is applied to a computer. The computer is connected to the PTZ controller via the k-line, and the PTZ controller is connected to the infrared LED, the camera and the infrared photocell through the signal line. The method comprises: (a) receiving a video image of a surveillance area acquired by a camera, selecting a moving character as a tracking target in the video image, and (b) determining the tracking target according to a suspicious target tracking algorithm. Position in the video image; (6) confirming the position of the tracking target in the next video image according to the suspicious target tracking algorithm, (d) calculating the position in the two video images according to the tracking target The tracking direction is in the direction of movement and the distance of movement of the monitoring area; (e) according to the above calculation, 099138366 Form No. A0101, Page 5 / Total 20 pages 201220215, a control command is generated, and the PTZ controller is driven to drive the camera according to the control command. The direction moves the corresponding distance so that the tracking target is always within the viewing angle range of the camera; (f) controlling the pan-tilt controller to drive the infrared light-emitting diode to generate infrared rays to illuminate the tracking target, and the infrared photoelectric tube senses the infrared penetration tracking The change of light intensity after the target, generating a pulse electrical signal according to the change of the light intensity, and transmitting the pulse through the signal line The signal is transmitted to the PTZ controller, and the PTZ controller converts the pulse electrical signal into a digital signal and transmits it back to the computer; and (g) determines whether the frequency of the digital signal exceeds a normal heartbeat frequency range of the person, if not exceeded If the normal heartbeat frequency range stops tracking the tracking target, the process ends. If the normal heartbeat frequency range of the person is exceeded, the tracking target is determined to be a suspicious target, and the process returns to step (C) to continue tracking the tracking target. Compared with the prior art, the suspicious target recognition and tracking system and method provided by the present invention can eliminate the interference of other characters in the monitoring area, and identify and track the suspicious target. [Embodiment] [0007] Referring to Fig. 1, an application environment diagram of a preferred embodiment of the suspicious object recognition and tracking system 100 (hereinafter referred to as "system 100") of the present invention is shown. The system 100 is applied to a computer 10 which is connected to a pan/tilt controller 20 via a signal line. The pan/tilt controller 20 is connected to the infrared illuminating diode 30, the camera 40, and the infrared photo transistor 50 via signal lines. It is to be noted that Fig. 1 only shows the electrical connection relationship between the above electronic devices, and does not represent the spatial positional relationship of each electronic device. In this embodiment, the infrared light emitting diode 30 can be mounted directly above the camera 40 and bound to the camera 40. When the pan/tilt controller 20 drives the camera 40 to move, the infrared light is 099138366. Form number A0101 Page 6 of 20 0992066853-0 201220215 - The polar body 30 performs the same motion as the camera 40. [0008] The camera 40 collects the video image of the monitoring area, and transmits the captured video image to the computer through the PTZ controller (10) to analyze the video image according to a tracking target (as shown in the figure). Character 7〇) In the position of the video image, the position of the tracking target in the monitoring area is calculated, and the control command is sent according to the displacement information to the pan/tilt controller PTZ controller 20 according to the control. The command-driven camera 4 〇 motion tracks the tracking target [0009] so that the tracking target is always within the viewing angle of the camera 40. While driving the camera 40 to move, the pan/tilt controller 2 drives the infrared illuminating body 10 to generate infrared rays (shown by broken lines in Fig. 1) to illuminate the tracking target. The infrared photocell 5 detects the intensity of the light after the strong infrared rays penetrate the target. The pulse electrical signal is generated according to the change of the light intensity, and the pulse electrical signal is transmitted to the pan-tilt controller 2 through the signal line. In the present embodiment, the infrared photocell 50 is mounted under the floor 6〇 in the monitoring area. In:: .: 丨M; '.. , i ';:: :;:; .:;, i: Ο [0010] In other embodiments, the infrared photocell 5 〇 can also be mounted on the side of the monitoring area Inside the wall. As the human heart fluctuates, the translucency of the human tissue changes, which in turn causes changes in the intensity of the infrared light that passes through a part of the body, such as the earlobe. The translucency of the human tissue changes in frequency and the frequency of the heartbeat, and the frequency of change in the intensity of the light passing through the human body is consistent with the frequency of change in the translucency of the human tissue. Therefore, the heartbeat frequency of the moving person can be known by the frequency of the change in the intensity of the light passing through the infrared light of the human body. 099138366 Form No. A0101 Page 7 / Total 20 Page 0992066853-0 201220215 [0011] Infrared photocell 50 senses the change of the intensity of infrared rays passing through the Λ, % human body, and generates a pulsed electrical signal according to the change of light intensity. (4) (4) The signal is converted to a digital signal and transmitted to the computer 10. The system 100 determines whether the frequency of the digital signal exceeds a normal heartbeat frequency range of the person. If the normal heartbeat frequency of the person is exceeded, the tracking target is determined to be a suspicious target, and the tracking of the tracking operation is continued. $ [0012] Referring to FIG. 2, the functional system diagram of the system (10) is compared to the embodiment. The heng system _ includes a receiving module (1), a selection module, a target aging module 130, a command generation module Ι4ΰ, a _ mode, and an alarm module/160 阙 receiving module 11 接收 for receiving the monitoring area of the camera 4〇 Video images at different times. [0015] [0015] The selection module 1 is then used as a tracking target in the video. For example, in the selection frequency image, select - «moving money for light m target tracking 縱 丨 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Zd In this embodiment, the suspicious target chase algorithm is a continuously adaptive average migration algorithm (Continuously Apative Mean_Shift 'CamShift). The algorithm searches for a tracking target in the video image according to the color feature of the tracking target. The position and size of the video in the video image. For example, 'in this embodiment, it is assumed that the main color of the tracking target dress is dark red', then the target tracking wedge is 13 to include the smallest rectangle of the deep red range as the tracking target. In the engraved video image 099138366 Form No. A0101 Page 8 / Total 20 Page 0992066853-0 The size of 201220215 'The position of the center point of the smallest rectangle is the position of the target in the image of the cheek image at t = 1 [0016] The target tracking module 1 30 is further configured to calculate a moving direction and a moving distance of the tracking target in the monitoring area according to the position of the tracking target in the video image at different times. For example, if the location of the tracking target is p2 (x2, y2, z2) in the video image at t = 2s, the target tracking module 130 calculates P2 (x2, y2, z2) and Pi (xl, The difference between yl, zl) is obtained in the moving direction and moving distance of the tracking target in the monitoring area. [0017] [0018] [0019] The command generating module 14 is configured to generate a control command based on the above calculation result The control command is sent to the pan-tilt controller 20, and the pan-tilt controller 20 controls the camera 40 to move the corresponding distance in the corresponding direction according to the control command, so that the tracking target is always within the viewing angle range of the camera 40. For example, the pan-tilt controller 20 According to the control command, the distance in which the camera 40 moves in the X-axis direction (χ2-χ1), the distance in the γ-axis direction (y2-yi), and the distance in the z-axis direction (z2-zl) are controlled. Meanwhile, the pan/tilt controller 20 drives the infrared light-emitting diode 3 to generate infrared rays to illuminate the target. The infrared phototube 50 senses the change of the light intensity after the infrared light penetrates the target, generates a pulse electric signal according to the change of the light intensity, and transmits the pulse through the signal line. Electric signal transmission The pan-tilt controller 20 converts the pulse electrical signal into a digital signal. The receiving module 110 is further configured to receive the digital signal returned by the pan-tilt controller 20. The determining module 15 0 Whether the pure (four) rate exceeds the normal heartbeat frequency range of the person, for example, 6G~1GG times/minute. If the frequency of the digital signal exceeds the normal heartbeat frequency range of the person, the tracking target is determined to be 099138366. Form No. A0101 Page 9 / Total 20 pages 0992066853-0 201220215 Doubtful target, continue to track the tracking target. [0022] [0023] [0023] The alarm module 1 6 G is used when the frequency of the digital signal exceeds the normal heartbeat frequency range of the person 'turn on the computer's alarm device, such as an LED warning light to notify the monitoring personnel to monitor the area. There are suspicious targets. Referring to Figure 3, there is shown a flow chart of a preferred embodiment of the suspicious object recognition and tracking method of the present invention. Step S101, the receiving module 110 receives a video image of the monitoring area collected by the camera 4 (for example, a video image at time t = ls), and the selection module 120 selects a motion of the two in the video image. The character is the target of tracking. Step S103, the target tracking module 13 determines the position of the tracking target in the video image according to the suspicious target tracking algorithm. In this embodiment, the suspicious target tracking algorithm is continuous. Continuously Apative Mean-Shift 'CamShift. This algorithm searches the video image for the tracking target according to the color feature of the tracking target, and determines the position and size of the tracking target in the video image. For example, in the present embodiment, it is assumed that the main color of the tracking target dress is dark red', and the target tracking module 丨3〇 is the minimum rectangle containing the deep red range as the tracking target in the video image at time t=s The size of the center point of the minimum rectangle is the bit in the video image of the tracking target at t = sl. Step S105' The target tracking module 130 determines the tracking target in the next video image according to the suspicious target tracking algorithm. Position 1) 2 (\2, 卩 2, 22) in the image (for example, a video image at time t = 2s). Specific method and step 099138366 Form number A0101 Page 10 / Total 20 pages 0992066853-0 [0024] 201220215 [0025] [0026] [0027]
[0028] [0029] S103類似，在此不再贅述。 步驟S1 0 7 ^目標追縱模組13 0根據該追縱目標在該兩張視 頻圖像中的位置計算該追蹤目標在監控區域的移動方向 及移動距離。例如，目標追蹤模組130計算 p2(x2，y2，z2)與pl(xl，yl，zl)的差值，得到該追蹤目 標在t = Is至t = 2s時間内在監控區域的移動方向及移動距 離。 步驟S109，命令生成模組140根據上述計算結果產生控制 命令，並發送該控制命令至雲台控制器20，雲台控制器 20根據該控制命令控制攝像頭40向相應方向移動相應距 離，使得該追蹤目標始終處於攝像頭40的視角範圍内。 例如，雲台控制器20根據該控制命令控制攝像頭40向X軸 方向移動（x2-xl)的距離，Y軸方向移動（y2-yl)的距離 ，Z軸方向移動（z2-zl)的距離。 步驟S111，雲台控制器20驅動紅外發光二極體30產生紅 外線照射該追蹤目標。 步驟S113，紅外光電管50感測紅外線穿透追蹤目標後的 光強度變化，根據光強度變化產生脈衝電信號，並透過 信號線將該脈衝電信號傳送給雲台控制器20。 步驟S115，雲台控制器20將該脈衝電信號轉換為數位信 號，並透過信號線將該數位信號發送給電腦10。 步驟S117，判斷模組150判斷該數位信號的頻率是否超過 人的正常心跳頻率範圍，例如60〜100次/每分鐘。若該數 位信號的頻率超過人的正常心跳頻率範圍，則流程進入 099138366 表單編號A0101 第11頁/共20頁 0992066853-0 [0030] 201220215 步驟SI 19。若該數位信號的頻率未超過人的正常 率範圍’則流程進入步驟S1 21。 [0031] 步驟S119，判斷模組150認定該追蹤目標為可_目枳 警模組160開啟電腦的報警裝置，例如LED警示垵，^報 / W通 知監控人員監控區域有可疑目楳出現。之後，後程自歩 驟s 1 0 5開始重複，以繼續追跟該追縱目‘。 [0032] 步驟S121 ,目標追蹤模組停止追蹤該追蹤目榡。> ^ <後 ，流程結束。 【圖式簡單說明】 [0033] 圖1係本發明可疑目標識別及追蹤系統較佳實施例之應用 環境圖。 [0034] 圖2係本發明可疑目標識別及追蹤系統較佳實施例之功能 模組圖。 [0035] 圖3係本發明可疑目標識別及追蹤方法較佳實施例之流程 圖。 【主要元件符號說明】 [0036] 電腦：10 [0037] 可疑目標識別及追蹤系統：1〇〇 [0038] 接收模組：110 [0039] 選擇模組：120 [0040] 目標追蹤模組：130 [0041] 命令生成模組：140 099138366 表單編號A0101 第12頁/共20頁 0992066853-0 201220215 [0042] 判斷模組：150 [0043] 報警模組：160 [0044] 雲台控制器：2 0 [0045] 紅外發光二極體：30 [0046] 攝像頭：40 [0047] 紅外光電管：50 [0048] 地板：60 [0049] 人物：70 Ο 099138366 表單編號Α0101 第13頁/共20頁 0992066853-0[0029] S103 is similar, and details are not described herein again. Step S1 0 7 ^ The target tracking module 130 calculates the moving direction and the moving distance of the tracking target in the monitoring area according to the position of the tracking target in the two video images. For example, the target tracking module 130 calculates the difference between p2(x2, y2, z2) and pl(xl, yl, zl), and obtains the moving direction and movement of the tracking target in the monitoring area from t = Is to t = 2s. distance. In step S109, the command generation module 140 generates a control command according to the calculation result, and sends the control command to the pan-tilt controller 20, and the pan-tilt controller 20 controls the camera 40 to move the corresponding distance to the corresponding direction according to the control command, so that the tracking is performed. The target is always within the viewing angle of the camera 40. For example, the pan/tilt controller 20 controls the distance that the camera 40 moves (x2-xl) in the X-axis direction, the distance in the Y-axis direction (y2-yl), and the distance in the Z-axis direction (z2-zl) according to the control command. . In step S111, the pan/tilt controller 20 drives the infrared illuminating diode 30 to generate an infrared ray to illuminate the tracking target. In step S113, the infrared photocell 50 senses the change of the light intensity after the infrared rays penetrate the tracking target, generates a pulse electrical signal according to the change of the light intensity, and transmits the pulse electrical signal to the pan-tilt controller 20 through the signal line. In step S115, the pan/tilt controller 20 converts the pulse electrical signal into a digital signal and transmits the digital signal to the computer 10 through the signal line. In step S117, the determining module 150 determines whether the frequency of the digital signal exceeds a normal heartbeat frequency range of the person, for example, 60 to 100 times/minute. If the frequency of the digital signal exceeds the normal heartbeat frequency range of the person, the flow proceeds to 099138366 Form No. A0101 Page 11 of 20 0992066853-0 [0030] 201220215 Step SI 19. If the frequency of the digital signal does not exceed the normal rate range of the person', the flow advances to step S21. [0031] Step S119, the determining module 150 determines that the tracking target is an alarm device that can be turned on by the police module 160, such as an LED warning, and the notification/w notification that the monitoring personnel has a suspicious target in the monitoring area. After that, the back-end is repeated from step s 1 0 5 to continue to follow the tracking ‘. [0032] Step S121, the target tracking module stops tracking the tracking target. > ^ < After the process ends. BRIEF DESCRIPTION OF THE DRAWINGS [0033] FIG. 1 is an application environment diagram of a preferred embodiment of the suspicious object recognition and tracking system of the present invention. 2 is a functional block diagram of a preferred embodiment of the suspicious object recognition and tracking system of the present invention. 3 is a flow chart of a preferred embodiment of the suspicious object identification and tracking method of the present invention. [Main component symbol description] [0036] Computer: 10 [0037] Suspicious target recognition and tracking system: 1〇〇[0038] Receiver module: 110 [0039] Selection module: 120 [0040] Target tracking module: 130 [0041] Command generation module: 140 099138366 Form number A0101 Page 12 / Total 20 pages 0992066853-0 201220215 [0042] Judgment module: 150 [0043] Alarm module: 160 [0044] PTZ controller: 2 0 [0045] Infrared LED: 30 [0046] Camera: 40 [0047] Infrared Phototube: 50 [0048] Floor: 60 [0049] People: 70 Ο 099138366 Form No. 101 0101 Page 13 / Total 20 Page 0992066853-0