TWI314419B - Surveillance system having tracking functions and control method thereof - Google Patents

Description

1314419 IX. Description of the invention: [Technical field to which the invention belongs] Not poor is a method for controlling the blood muscles of the θ ... called the blood muscles, and there is a monitoring system with tracking function and its control method. Technology] ..., the heart and the society how to change 'security issues' - straight time:: avoid. Therefore, monitoring systems are important. The conventional monitoring system includes a mobile camera and a computer. The mobile camera includes a - and a moving mechanism, wherein the camera is used to shoot - shadow #, this film: if - analog or digital film. The moving mechanism is, for example, a rotary table mechanism for carrying a camera, and the rotary table mechanism allows the camera to move horizontally or vertically to expand the shooting range of the camera. The brain has a mobile camera control interface that controls the motion of the mobile camera and processes the movies captured by the mobile camera.

Although the conventional monitoring system has the function of automatically moving back and forth, it is easy to avoid the monitoring because the action is monotonous and fixed. Therefore, the additional tracking function for the monitoring system, especially the function of automatically tracking the human face, will effectively enhance the role of the monitoring system, making the monitoring system less vulnerable to dodging, and the face of the shooting will contain more photographers. Faces are noodles, making the results more valuable. However, the tracking function currently applied to the monitoring system mostly stops automatically detecting the moving object, that is, after the moving object enters the monitoring screen, the warning message is immediately sent, and the recording function is turned on at the same time, but the automatic tracking target is still missing. Features. [Summary Content] 5 1314419 The purpose of the month is to provide a monitoring system with a tracking function, and a good knowledge monitoring system is vulnerable to dodge. The other purpose of the full moon is to provide a face with a tracking face function; the screen system makes the shooting result of the monitoring system contain more of the ray of the photographer, and the above purpose is proposed. Surveillance of the tracking function - mobile camera, a computer and a tracking software, in which the camera is used to shoot a movie 'computer with a mobile camera::=: control: mobile camera, tracking software = The control module is configured to generate at least one of the digital images captured by the mobile camera. The module is used to determine whether each component of the digital image is a component. The image element's mechanism control module 3=type camera tracks the target object, wherein the correction signal is generated by the image-like prime number of the object in the second area and the second area. The monitoring system of the preferred embodiment wherein the target is 匕3 - an image of a human face. The monitoring system of the preferred embodiment, wherein the mobile phone package 3-rotating table mechanism is used to change the shooting of the mobile camera, such as the monitoring system of the preferred embodiment, wherein the digital camera captures A partial area of the shirt. A monitoring system according to a preferred embodiment of the present invention, wherein the image pixel color range of the target 1314419 includes a face skin color range in an RGB color coordinate area. According to a preferred embodiment of the present invention, the correction signal includes a horizontal direction correction signal and a vertical direction correction signal. The invention further provides a tracking software for monitoring system, which is executed on a computer. The computer has a control interface of a mobile camera for controlling the mobile camera and processing the film captured by the mobile camera. The tracking software includes an image capture system. The image capture module is configured to generate at least one digital image. The digital image is captured from a mobile camera, and the image processing module is used for the image capture module. Determining whether each constituent element of the digital image belongs to an image element of a target object, and the mechanism control module is configured to generate a correction signal for the mobile camera to track the target object, wherein the correction signal is by comparing one of the digital images The area and a second area are generated by the number of image pixels belonging to the target. The invention further provides a control method for a monitoring system, which is implemented on a computer having a control interface of a mobile camera for controlling a mobile camera and processing a movie taken by the mobile camera. The control method is firstly from the mobile type. The camera captures at least one digital image, and then determines whether each constituent element of the digital image belongs to an image pixel of a target, and finally generates a correction signal for the mobile camera to track the target, wherein the correction The signal is generated by comparing the number of image pixels belonging to the target in the first region and the second region of the digital image. The monitoring system of the present invention includes a tracking software and thus has the function of tracking an object, thereby making it difficult for the target to escape the surveillance system's monitoring range. The monitoring system of the present invention 'the target of monitoring can be set as the face of the human' and thus has the function of tracking the human face, so that the face of the shooting contains more facial images of the photographer, so that the shooting result is more Have value. [Embodiment] Referring to Figure 1, a block diagram of a monitoring system having a tracking function according to a preferred embodiment of the present invention is shown. The monitoring system i includes a computer U and a mobile camera 12. The mobile camera 12 is used to capture a video, and the computer 11 has a control interface 13 of the mobile camera 12 for controlling the mobile camera 12 and processing the images captured by the mobile camera 12. It should be noted that the monitoring system 1 includes a tracking software 14, which is executed on the computer η. The tracking software 14 includes an image capturing module 141, an image processing module 142, and a mechanism control module (4). . The image capturing module 141 is configured to generate at least one digital image. The digital image is captured by the mobile camera 12 and the image processing module 142 is used to determine whether each constituent element of the digital image belongs to The image of a target is pixel. The mechanism control module 143 is configured to generate a correction signal to cause the mobile camera 12 to track the target. ^ Referring to Figure 2, there is shown an image capture ", L-process diagram in accordance with a preferred embodiment of the present invention. First, set a range of captures in a monitor surface. Thereafter, the image of the captured range is extracted into a digital image S22. Referring to Fig. 3, the outline of the range is not set in accordance with a preferred embodiment of the present invention. The user can set a capture range of 32 in the monitor screen 31. For example, the rectangular monitoring of the surface of the palladium 3 1 from the palladium.庋軚 is (〇, 〇), the end point coordinates are (χ, γ), 8 1314419 The user, for example, sets the rectangle to take the starting coordinate of _32 $ (training, or 'user can also enter the starting coordinates (X The length and width of the training and the extracted images are i X2-X1 |, | Y2-Y!, so that the setting of the capturing range 32 can be completed. Referring to Figure 4, it is shown in accordance with the present invention. Image processing flow of a preferred embodiment. The target object tracked by the monitoring system of the present invention is, for example, a human face. The image color range of the target includes a face color range in an rgb color coordinate area. Referring to the first and fourth figures, first, the image processing module 142 If the human image captures the digital image S41' captured by the modulo 16141 and the digital image is like the color of each pixel and the color of the face. The conditional expression is compared with S42, and finally the pixel points that match the face skin condition condition are retained, and the pixel points that do not conform to the face skin condition condition are removed to separate the face color and the non-face color skin area S43. In detail, the face color condition is mainly based on digital image Face skin color and non-human skin color are statistically established in the RGB color coordinates, and the face color condition is, for example,: FUx (0·2<s<°3) & (°*3 <;r<α5)&(G>B)&(R>g+30)

Hx,y) [〇otherwise ....(1) where: r =R/(R+G+B) m ::;3; In the face color conditional expression (1), Fl(x,y) represents the composition Digital image 9 1314419 Any pixel in the image, and R, G, and B are the color values of the red, green, and blue colors, respectively. When Fi(x,y) = 〇, it represents the pixel point. The color does not fall within the skin color range of the RGB color coordinate area, that is, the color of the pixel point is not the skin color of the face', so we remove the pixel point; when FlCx'y)^! That is, the color of the pixel point falls within the range of the skin color of the face in the RGB color coordinate area, and the pixel point is reserved. In this way, the pixel points conforming to the face skin condition condition (1) can be retained, and the pixel points that do not conform to the face skin condition condition (1) can be removed to separate the face color and the non-fit face in the digital image. The area of skin color. Referring to Figure 5, there is shown a flow chart of the mechanism control in accordance with a preferred embodiment of the present invention. First, the left and right deviations of the area where the skin color of the face is located are deviated from the information S51, and then the upper and lower deviation information S52' is calculated for the area where the skin color of the face is located. Finally, the left and right deviation information and the upper and lower deviation information are used to generate a correction signal S53. Referring to FIG. 1 and FIG. 5 simultaneously, in the step S51 of calculating the left and right deviation information of the area where the skin color of the face is located, the image processed by the image processing module 142 is first divided into the left half and the right half. Then, calculate the total number of face color points in the left half and right half, and finally compare the total number of face color points in the left half and right half, if the left half has more face color The number of pixels, that is, the area where the skin color of the face is deviated from the left side of the coordinates of the center point of the digital image; if the right half of the area has more face color, the area where the skin color of the face is deviated from the digital image The right side of the center point coordinates. For example, when the image processing module 142 processes a digital image of a size (KXL), the mechanism control module (4) will complete the 1314419 separation of the face color and the area of the non-human skin color. The image is divided equally into the left half and the right half. Then, for example, the following two equations are used to determine the person.

Is the area of the face skin color deviated from the left or right of the center point of the digital image: ^-1 2 L

K L Σ Σ F1(x,y) > Σ Σ f

κ L

Hx. y) ε •(4) κ 1 (χ,y) > Σ Σ F “x,y) + a 〇γ sb 〇.(5) where ε is an error tolerance 'to prevent being a face The area where the skin color is located has only slight changes, and the overall deviation of the center point coordinates of the digital image may cause the mobile camera 12 to move back and forth. The error tolerance can be determined by the user's choice when the user selects a larger one. The error tolerance 'when the area of the face skin color deviates from the center point coordinate of the digital image, the information of the face color area deviation will be obtained. The principle of equation (4) and equation (5) is when the face color is When the area is deviated to the left of the center point of the digital image, the left half of the digital image will naturally have more face color points, that is, when the equation (4) is established, it represents the face of the left half. The skin color point is much more than the right half area, so you can get the information that the face color area deviates to the left. Similarly, when equation (5) is established, it represents the face color ratio of the right half area. The left half is coming much more, so It can get the information that the face color area is deflected to the right. The same principle as above is used to calculate the face color area and the upper side is 11 1314419

In the step of separating information, the image processed by the digital image processing module 142 is first divided into upper half and lower half regions, and then the total number of face color pixels in the upper half and the lower half is calculated. Finally, Compare the total number of face color points in the upper half and the lower half. If the upper half has more face color points, it means that the area where the skin color is located deviates from the center point coordinates of the digital image. Above; if the lower half has more face color points, that is, the area where the skin color of the face is deviated from the center point coordinates of the digital image. For example, the following two equations are used to determine whether the area of the human face color is above or below the center point of the digital image:

2 ^ KLK ....(6) ....(7) ^〇SFl(X,y) > Σ ZFi(x;y)+^ L Κ γ'1 κ L Fl(x> y) > Σ Σ F1(X , y) + ε ^Γ+1 ^〇Λ = ° When equation (6) is established, it means that the face of the upper half has a much more radiance than the lower half, so you can get the face color area. Information that deviates upwards; as well, when equation (7) is established, it means that the face color of the face in the lower half is much larger than that of the upper half, so that the information of the face color area deviating downward can be obtained. Finally, the mechanism control module 143 will use the left and right deviation information and the upper and lower deviation signal generation-correction signals to be transmitted to the control plane 71 of the mobile camera U to make the face color area and return to the digital image. The middle point and the point further cause the mobile camera 12 to track the face of the subject. 12 1314419 It is worth mentioning that, in the monitoring system 10 of the present invention, the tracking software i4 can be independently developed into an application software, and is attached to a conventional digital monitoring system by means of an external connection, so that the conventional digital monitoring system has a tracking one. The function of the target. By using the tracking software 14 to set the range of the image to be taken and the color range of the target to be tracked, the user can achieve the purpose of tracking the target. It will be apparent from the above-described preferred embodiments of the present invention that the application of the present invention has the following advantages. ~ - The monitoring system of the present invention has a tracking software and thus has the function of tracking-targeting, thereby making it difficult for the target to escape the monitoring range of the monitoring system. 2. The monitoring system of the present invention has a tracking software attached to the conventional digital monitoring system in an external manner, so that the digital monitoring system has the function of tracking the target. - The monitoring system of the present invention has the function of tracking a human face, and the captured face contains more faces of the subject's face, making the result more valuable. Although the present invention has been disclosed in a preferred embodiment as described above, it is not intended to limit the invention to any skilled person in the art, and does not depart from the scope of the invention. Retouching, therefore, the protection of the present invention is subject to the scope defined by the patent scope. BRIEF DESCRIPTION OF THE DRAWINGS In order to make the above and other objects of the present invention 'features, advantages and embodiments b obvious, the detailed description of the drawings is as follows: 13 1314419 Figure i shows a preferred embodiment in accordance with the present invention. Example of a monitoring system block diagram. Figure 2 is a flow chart showing the image browning in accordance with the preferred embodiment of the present invention. Figure 3 is a diagram showing the setting of the operating range in accordance with the preferred embodiment of the present invention. Figure 4 depicts an image processing flow not in accordance with a preferred embodiment of the present invention.

Figure 5 is a flow chart showing the mechanism control in accordance with a preferred embodiment of the present invention.

[Main component symbol description]: Monitoring system U: Computer 12·Mobile camera 13: Control interface 14: Tracking software 141: Image capturing module 142: Image processing module 143: Mechanism control module S21. Step S22 of setting a range of capturing in the face: Step of capturing the image of the captured range into a digital image 3 1 : Monitoring face 32: capturing range 541 · Reading into the image capturing module Step 542 of the digital image: Step S43 of comparing each pixel of the digital image with a face color condition: Step S51 of separating the face color and the face of the non-human skin color. S10. Step of deviating from the right information S52. Step S53 of deciding the information above and below the area where the skin color of the human face is located: Step of generating a correction signal ΙΊ·

Claims (1)

9 Towering Correction Replacement Page 1314419 Circumference: X. Patent Application 1. A monitoring system with tracking function, comprising: a mobile camera for shooting a movie; a computer having a control interface for the mobile camera, To process the video and control the mobile camera; a tracking software, executed on the computer, the tracking software includes: an image capturing module for generating at least one digital image, the digital image is captured from the video; An image processing module for determining whether each constituent element of the digital image belongs to an image pixel of a target object, and a mechanism control module for dividing the digital image into a first region and a second image a region, and by comparing the number of image pixels of the first region and the second region of the digital image that belong to the target, generating a correction signal* to cause the mobile camera to track the target. 2. The surveillance system of claim 1, wherein the image of the target comprises an image of a human face. 3. The surveillance system of claim 1, wherein the mobile camera includes a rotary table mechanism for changing a shooting position of the mobile camera. 4. The surveillance system of claim 1, wherein the digital image is a partial region taken from the film. 15 1314419 j^VW replacement page β 5. The monitoring system described in the scope of the patent application, wherein the image color range of the target contains - the face color range in the RGB color coordinate area. 6. If you apply for a patent range! The monitoring system of the present invention, wherein the mechanism control module generates a horizontal direction correction signal by comparing the number of image pixels belonging to the § target in the left half and the right half of the digital image. 7. The monitoring system of claim 1, wherein the mechanism control module generates a vertical direction correction signal by comparing the number of image pixels belonging to the object in the upper half and the lower half of the digital image. The tracking software of the monitoring system is executed on a computer having a control interface of the mobile camera for controlling the mobile camera and processing the film-image manipulation module captured by the mobile camera. The image processing module is configured to determine whether each constituent element of the digital image belongs to an image element of the target object; and - the mechanism is configured to generate an at least one digit: the image is obtained from the image; The control module is configured to divide the digital image into a first region and a second region, and generate a number of image pixels belonging to the target in the second region by comparing the (four)-regional fish of the digital image. A correction signal causes the mobile camera to track the target. 9. A tracking soft body of a surveillance system as claimed in claim 8 wherein the image of the object comprises an image of a human face. Ι〇· As claimed in the eighth paragraph of the patent application, the digital image is taken from the tracking system and captured from the local area of the film. For example, the 8th RGB color coordinate area of the patent application scope One of the face color range bodies, wherein the target (4) system of the traces of the image of the elementary color range contains one, and 2 of the patent scope of the monitoring (10) of the tracking and the second area Ryusaki compares the number (4) like the left half and the right 4 positive signals. The number of images of the standard object produces - the horizontal direction #体,二^11=_the monitoring system of the eighth item is more than the second mode _ New Zealand (4) like the upper half, the lower half is positive 3 b The number of image pixels produced by the target - the vertical direction of the brain control system with the right view system, executed in the computer, the camera and the mobile camera control The interface is used to control the mobile type: U: the film taken by the mobile camera, the method includes the image; the δ shift (four) takes the job to take it - the image takes at least one number - each component of the broken money image Whether the vegetarian element belongs to a target 17 1314419 Zhu Yue 1 Correcting a replacement page image pixel; dividing the digital image into a first region and a second region; and comparing the number of image pixels in the first region and the second region that belong to the target by comparing the digital image 'generating a correction signal to cause the mobile camera to track the target. 15. The control method of the monitoring system as described in the patent application section further includes a step of setting the range of the film capture. The control method of the monitoring system according to claim 14, wherein the determining step is to determine whether each component pixel color of the digital image is located in a face color range of an RGB color coordinate area: 17· The method for controlling a monitoring system according to claim 14, wherein the correction signal is generated by comparing the number of image pixels belonging to the target in the left half and the right half of the digital image. The method for controlling a monitoring system according to claim 14, wherein the correction signal is obtained by comparing the upper half and the lower half of the digital image to the target The number of images of the standard is produced by the number of elements. 18