TWI397866B - System and method for detecting object - Google Patents
System and method for detecting object Download PDFInfo
- Publication number
- TWI397866B TWI397866B TW98113132A TW98113132A TWI397866B TW I397866 B TWI397866 B TW I397866B TW 98113132 A TW98113132 A TW 98113132A TW 98113132 A TW98113132 A TW 98113132A TW I397866 B TWI397866 B TW I397866B
- Authority
- TW
- Taiwan
- Prior art keywords
- same
- unit
- information
- monitored
- monitoring
- Prior art date
Links
Landscapes
- Image Analysis (AREA)
- Alarm Systems (AREA)
Description
本發明涉及一種物體偵測系統及方法。 The invention relates to an object detection system and method.
目前，影像監控技術可應用於各個領域中，以對複雜背景環境中之狀況進行即時監控以及後續分析，如，當商家需要統計一個時間段內進入商場之人數時，可使用影像監控技術對經過商場大門之人臉進行偵測，每偵測一人記錄一次，用戶便可根據記錄結果進行人數之統計，又，於交通監控領域，可偵測經過某一站點之車輛之特徵，以記錄一時間段內該站點之車輛流通量。然，使用傳統之偵測技術無法辨識於該時間段內出現之相同監控物件，致使重複記錄該同一監控物件，影響了監控之結果。 At present, image monitoring technology can be applied to various fields to perform real-time monitoring and subsequent analysis on the situation in a complex background environment. For example, when a merchant needs to count the number of people entering the mall within a certain period of time, the image monitoring technology can be used to The face of the mall gate is detected. Each time a person is recorded, the user can count the number of people according to the recorded results. In the field of traffic monitoring, the characteristics of the vehicle passing through a certain site can be detected to record one. Vehicle traffic at the site during the time period. However, the use of conventional detection techniques does not identify the same monitored object that occurred during that time period, causing the same monitored object to be recorded repeatedly, affecting the results of the monitoring.
鑒於以上內容，有必要提供一種物體偵測系統及方法，可辨識於不同時間所偵測之物體是否為同一物體。 In view of the above, it is necessary to provide an object detection system and method for recognizing whether objects detected at different times are the same object.
一種物體偵測系統，包括一影像輸入單元、一特徵擷取單元及一存儲比較單元，該影像輸入單元用於連續獲取出現於一偵測區域中之複數監控物件之影像訊號，該特徵擷取單元根據該影像訊號擷取對應監控物件之特徵資訊，並根據擷取到之特徵資訊對對應之監控物件賦予一編號傳送至該存儲比較單元，其中，特徵資訊相同之監控物件被賦予相同之編號，該存儲比較單元用於存儲該等監控物件之編號，且該存儲比較單元對相同之編號僅 存儲一次。 An object detecting system includes an image input unit, a feature capturing unit and a storage comparing unit, wherein the image input unit is configured to continuously acquire image signals of a plurality of monitoring objects appearing in a detecting area, and the feature capturing The unit obtains the feature information of the corresponding monitoring object according to the image signal, and assigns a number to the corresponding monitoring object according to the extracted feature information, and transmits the same number to the storage comparison unit, wherein the monitoring object with the same feature information is given the same number. The storage comparison unit is configured to store the numbers of the monitored objects, and the storage comparison unit has the same number only Store once.
一種物體偵測方法，包括以下步驟：一影像輸入單元連續獲取出現於一偵測區域中之複數監控物件之影像訊號；根據該影像訊號擷取對應監控物件之特徵資訊；根據擷取到之特徵資訊對每一監控物件賦予一特定之編號，且具有相同特徵資訊之監控物件被賦予相同之編號；以及比較出現於該偵測區域中之監控物件之編號是否與已存儲之編號相同，當出現於該偵測區域中之監控物件之編號與已存儲之任一編號相同時則不重複存儲該相同之編號，當出現於該偵測區域中之監控物件之編號與已存儲之所有編號均不同時，存儲該出現於該偵測區域中之監控物件之編號。 An object detecting method includes the following steps: an image input unit continuously acquires image signals of a plurality of monitoring objects appearing in a detecting area; and extracts characteristic information of the corresponding monitoring object according to the image signal; according to the captured features The information is given a specific number for each monitored object, and the monitored object having the same characteristic information is given the same number; and the number of the monitored object appearing in the detected area is the same as the stored number when appearing When the number of the monitored object in the detection area is the same as any of the stored numbers, the same number is not repeatedly stored, and the number of the monitored object appearing in the detection area is different from all the stored numbers. The number of the monitored object appearing in the detection area is stored.
該物體偵測系統及方法根據該監控物件之特徵資訊進行編號，使同一監控物件被賦予相同之編號，並透過存儲比較單元判斷其接收到之編號是否與其所存儲之編號相同，可辨識於一預設之時間段內所偵測之物體是否為同一物體。 The object detection system and method are numbered according to the characteristic information of the monitored object, so that the same monitored object is given the same number, and the stored comparison unit determines whether the received number is the same as the number stored therein, and can be identified in one Whether the detected object is the same object within the preset time period.
請參閱圖1，本發明物體偵測系統1用於偵測並記錄出現於一偵測區域之監控物件，本實施例中，以人臉作為該監控物件進行舉例說明，於其他實施例中，該監控物件亦可為其他具備特定特徵之物體，如車牌，該物體偵測系統1之較佳實施方式包括一影像輸入單元12、一特徵擷取單元14及一存儲比較單元16。 Referring to FIG. 1 , the object detection system 1 of the present invention is used for detecting and recording a monitoring object that appears in a detection area. In this embodiment, a human face is used as the monitoring object for example. In other embodiments, The object of the object detection system 1 includes an image input unit 12, a feature extraction unit 14, and a storage comparison unit 16.
該影像輸入單元12用於連續獲取出現於該偵測區域中之 人臉之影像訊號，該影像輸入單元12包括一攝像頭、一圖像採集卡或其他可獲取物體影像之裝置。該特徵擷取單元14用於接收該人臉之影像訊號，並根據接收到之人臉之影像訊號擷取人臉之特徵資訊，以及根據擷取到之特徵資訊對每一人臉賦予一編號，本實施方式中，該特徵資訊包括但不限於該人臉之形狀、膚色以及具體之五官特徵等資訊。該存儲比較單元16用於於一預設時間段內接收並存儲該人臉之編號，該存儲比較單元16每接收到一個人臉之編號，都將該接收到之編號與其所存儲之其他編號進行比較，由於同一人臉具有相同之特徵資訊，故，同一人臉亦被賦予相同之編號，當比較出存在與該接收到之編號重複之編號時，該存儲比較單元16不會重複存儲該接收到之編號，當該存儲比較單元16比較出其接收到之編號與其所存儲之其他編號均不相同時，該存儲比較單元16存儲該接收到之編號。 The image input unit 12 is configured to continuously acquire the image appearing in the detection area. The image signal of the face, the image input unit 12 includes a camera, an image capture card or other device for acquiring an image of the object. The feature capturing unit 14 is configured to receive the image signal of the face, and extract feature information of the face according to the image signal of the received face, and assign a number to each face according to the captured feature information. In this embodiment, the feature information includes, but is not limited to, information such as the shape of the face, the skin color, and the specific facial features. The storage comparison unit 16 is configured to receive and store the number of the face within a preset time period, and the storage comparison unit 16 performs the number of the face and the other number stored by the storage comparison unit 16 for each number. In comparison, since the same face has the same feature information, the same face is also given the same number. When comparing the number with the received number, the storage comparison unit 16 does not repeatedly store the reception. To the number, when the storage comparison unit 16 compares that the number it receives is different from the other number stored therein, the storage comparison unit 16 stores the received number.
請參閱圖2，該存儲比較單元16所存儲之編號可傳送至一處理系統2中之一處理單元22，以便用戶透過該處理單元22對該存儲比較單元16所存儲之編號進行後續分析處理，如統計編號之數量，本實施例中，可對該存儲比較單元16存儲人臉編號之時間進行設置，如，可根據實際需要以及該存儲比較單元16存儲量之大小，將該存儲比較單元16存儲每一人臉編號之時間設為10分鐘或10小時，故，若一人臉於設定之時間段內重複出現於該物體偵測系統1之偵側區域內，該物體偵測系統1均能夠辨識出該相同之人臉，從而避免了重複紀錄。 Referring to FIG. 2, the number stored by the storage comparison unit 16 can be transmitted to a processing unit 22 in a processing system 2, so that the user can perform subsequent analysis processing on the number stored by the storage comparison unit 16 through the processing unit 22. For example, in the embodiment, the time when the storage comparison unit 16 stores the face number may be set. For example, the storage comparison unit 16 may be configured according to actual needs and the storage amount of the storage comparison unit 16. The time for storing each face number is set to 10 minutes or 10 hours. Therefore, if a face repeatedly appears in the detection side area of the object detection system 1 during the set time period, the object detection system 1 can recognize The same face is taken out, thus avoiding duplicate records.
該處理系統2還可包括一座標存儲單元24、一目標追蹤單元26及一顯示單元28，該特徵擷取單元14每賦予人臉一編號，都將此編號及與此編號對應之人臉出現於該偵測區域內之時間及位置資訊傳送至該處理系統2，本實施例中，該人臉之位置資訊係以座標之表示形式傳送至該座標存儲單元24，當需追蹤人臉之移動軌跡時，該目標追蹤單元26從該座標存儲單元24取出具有相同編號之人臉之座標資訊，並分別將其以時間順序進行排序後顯示於該顯示單元28上，以便用戶對每一人臉之移動軌跡進行觀測，用戶亦可根據需要鎖定一定數量之監控目標進行持續追蹤。 The processing system 2 can further include a standard storage unit 24, a target tracking unit 26, and a display unit 28. The feature extraction unit 14 displays the number and the face corresponding to the number for each face number. The time and position information in the detection area is transmitted to the processing system 2. In this embodiment, the location information of the face is transmitted to the coordinate storage unit 24 in the form of coordinates, and the movement of the face needs to be tracked. At the time of the trajectory, the target tracking unit 26 retrieves the coordinate information of the face having the same number from the coordinate storage unit 24, and sorts them in chronological order and displays them on the display unit 28 so that the user can face each face. The trajectory is observed, and the user can also lock a certain number of monitoring targets as needed to continuously track.
請參閱圖3，本發明物體偵測方法之較佳實施方式應用於圖1及圖2中之物體偵測系統1，該物體偵測方法包括以下步驟： Referring to FIG. 3, a preferred embodiment of the object detecting method of the present invention is applied to the object detecting system 1 of FIG. 1 and FIG. 2, and the object detecting method includes the following steps:
步驟S1：該影像輸入單元12於一偵測區域內連續獲取出現於該區域中之監控物件之影像訊號，並將獲取之影像訊號傳送至該特徵擷取單元14，本實施例中，該監控物件為人臉。 Step S1: The image input unit 12 continuously acquires the image signal of the monitored object appearing in the area in a detection area, and transmits the acquired image signal to the feature capturing unit 14. In this embodiment, the monitoring The object is a human face.
步驟S2：該特徵擷取單元14根據接收到之影像訊號計算該監控物件之特徵資訊，該特徵資訊包括但不限於該監控物件之形狀、紋理、顏色等，本實施例中，該特徵資訊可為該人臉之形狀、膚色以及具體之五官特徵等資訊。 Step S2: The feature capturing unit 14 calculates the feature information of the monitored object according to the received image signal, and the feature information includes, but is not limited to, the shape, texture, color, and the like of the monitored object. In this embodiment, the feature information may be Information such as the shape of the face, the skin color, and the specific features of the facial features.
步驟S3：該特徵擷取單元14根據擷取到之特徵資訊對每 一監控物件賦予一特定之編號，並將每一監控物件之編號傳送至該存儲比較單元16。 Step S3: the feature extraction unit 14 compares each of the acquired feature information A monitoring object is assigned a specific number and the number of each monitored object is transmitted to the storage comparing unit 16.
步驟S4：該存儲比較單元16存儲該監控物件之編號，並比較其接收到之編號是否與其所存儲之編號相同。 Step S4: The storage comparison unit 16 stores the number of the monitored object and compares whether the received number is the same as the number stored therein.
步驟S5：當接收到之編號與該存儲比較單元16所存儲之任一編號相同時，該存儲比較單元16忽略該接收到之編號，不對其進行存儲操作。 Step S5: When the received number is the same as any of the numbers stored by the storage comparison unit 16, the storage comparison unit 16 ignores the received number and does not perform a storage operation thereon.
步驟S6：當接收到之編號與該存儲比較單元16所存儲之編號均不相同時，該存儲比較單元16存儲該接收到之編號。 Step S6: When the received number is different from the number stored by the storage comparing unit 16, the storage comparing unit 16 stores the received number.
步驟S7：該特徵擷取單元14所輸出之編號以及該存儲比較單元16所存儲之編號均傳送至該處理系統2進行後續處理。 Step S7: The number output by the feature extraction unit 14 and the number stored by the storage comparison unit 16 are transmitted to the processing system 2 for subsequent processing.
於步驟S7中，該處理系統2透過該處理單元22統計該存儲比較單元16所存儲之編號之數量。於步驟S2中，該特徵擷取單元14每賦予人臉一編號，均將此編號及與此編號對應之人臉出現於該偵測區域內之位置資訊以座標之表示形式傳送至該座標存儲單元24，該目標追蹤單元26將每一編號對應之座標以時間順序進行排序後顯示於該顯示單元28上。 In step S7, the processing system 2 counts the number of the numbers stored by the storage comparison unit 16 through the processing unit 22. In step S2, the feature capturing unit 14 assigns a number to the face, and transmits the number and the position information of the face corresponding to the number in the detection area to the coordinate storage in the form of coordinates. The unit 24, the target tracking unit 26 displays the coordinates corresponding to each number in chronological order and displays them on the display unit 28.
綜上所述，本發明符合發明專利要件，爰依法提出專利申請。惟，以上所述者僅為本發明之較佳實施方式，舉凡熟悉本案技藝之人士，於爰依本發明精神所作之等效修飾或變化，皆應涵蓋於以下之申請專利範圍內。 In summary, the present invention complies with the requirements of the invention patent and submits a patent application according to law. The above description is only the preferred embodiment of the present invention, and equivalent modifications or variations made by those skilled in the art will be covered by the following claims.
1‧‧‧物體偵測系統 1‧‧‧ Object Detection System
12‧‧‧影像輸入單元 12‧‧‧Image input unit
14‧‧‧特徵擷取單元 14‧‧‧Character extraction unit
16‧‧‧存儲比較單元 16‧‧‧Storage comparison unit
2‧‧‧處理系統 2‧‧‧Processing system
22‧‧‧處理單元 22‧‧‧Processing unit
24‧‧‧座標存儲單元 24‧‧‧Coordinate storage unit
26‧‧‧目標追蹤單元 26‧‧‧Target Tracking Unit
28‧‧‧顯示單元 28‧‧‧Display unit
圖1係本發明物體偵測系統較佳實施方式之模塊圖。 1 is a block diagram of a preferred embodiment of the object detection system of the present invention.
圖2係圖1中物體偵測系統連接於一處理系統之模塊圖。 2 is a block diagram of the object detection system of FIG. 1 connected to a processing system.
圖3係本發明物體偵測方法較佳實施方式之流程圖。 3 is a flow chart of a preferred embodiment of the object detecting method of the present invention.
Claims (7)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW98113132A TWI397866B (en) | 2009-04-21 | 2009-04-21 | System and method for detecting object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW98113132A TWI397866B (en) | 2009-04-21 | 2009-04-21 | System and method for detecting object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201039275A TW201039275A (en) | 2010-11-01 |
| TWI397866B true TWI397866B (en) | 2013-06-01 |
Family
ID=44995417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW98113132A TWI397866B (en) | 2009-04-21 | 2009-04-21 | System and method for detecting object |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI397866B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI611712B (en) * | 2017-01-04 | 2018-01-11 | 財團法人工業技術研究院 | Object tracking system and method therewith |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005234947A (en) * | 2004-02-20 | 2005-09-02 | Omron Corp | Number-of-person counter |
| TW200835328A (en) * | 2006-09-19 | 2008-08-16 | Sony Corp | Recording device and method, program, and reproducing device and method |
| TW200905575A (en) * | 2007-04-24 | 2009-02-01 | Objectvideo Inc | Method for finding paths in video |
-
2009
- 2009-04-21 TW TW98113132A patent/TWI397866B/en not_active IP Right Cessation
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005234947A (en) * | 2004-02-20 | 2005-09-02 | Omron Corp | Number-of-person counter |
| TW200835328A (en) * | 2006-09-19 | 2008-08-16 | Sony Corp | Recording device and method, program, and reproducing device and method |
| TW200905575A (en) * | 2007-04-24 | 2009-02-01 | Objectvideo Inc | Method for finding paths in video |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI611712B (en) * | 2017-01-04 | 2018-01-11 | 財團法人工業技術研究院 | Object tracking system and method therewith |
| US10252701B2 (en) | 2017-01-04 | 2019-04-09 | Industrial Technology Research Institute | Object tracking system and method therewith |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201039275A (en) | 2010-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240221417A1 (en) | Methods and apparatus to monitor environments | |
| US10490002B2 (en) | Suspicious person report system and suspicious person report method | |
| US8254633B1 (en) | Method and system for finding correspondence between face camera views and behavior camera views | |
| US11048942B2 (en) | Method and apparatus for detecting a garbage dumping action in real time on video surveillance system | |
| TWI466065B (en) | Detection device and detection method | |
| GB2585800A (en) | People counting and tracking systems and methods | |
| CN105631515B (en) | People flow counting system | |
| CN110648352B (en) | Abnormal event detection method and device and electronic equipment | |
| CN102306304B (en) | Face occluder identification method and device | |
| CN111325089A (en) | Method and apparatus for tracking object | |
| WO2019080669A1 (en) | Method for person re-identification in enclosed place, system, and terminal device | |
| JPWO2018198373A1 (en) | Video surveillance system | |
| CN111275030A (en) | Straight running detection and timing system and method based on face and human body recognition | |
| CN112232186B (en) | Epidemic prevention monitoring method and system | |
| JP6516702B2 (en) | People count system, number count method, and view method of number count result | |
| JP2001008197A (en) | Device for analyzing human action and recording medium for recording human action analytic program | |
| TWI397866B (en) | System and method for detecting object | |
| US10783365B2 (en) | Image processing device and image processing system | |
| CN110728249B (en) | Cross-camera recognition method, device and system for target pedestrian | |
| CN112906439A (en) | Passenger flow analysis method and system based on target tracking and behavior detection | |
| Hadi et al. | Fusion of thermal and depth images for occlusion handling for human detection from mobile robot | |
| WO2020114102A1 (en) | Video tracking method and system, and storage medium | |
| JP2020095651A (en) | Productivity evaluation system, productivity evaluation device, productivity evaluation method, and program | |
| KR101355206B1 (en) | A count system of coming and going using image analysis and method thereof | |
| CN111898471A (en) | Pedestrian tracking method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |